System

Abstract

A system is provided.SOLUTION: A system including means for recording voice data, means for transmitting the recorded voice data to a server, means for converting the voice data into text in the server, means for analyzing the text data and extracting a keyword indicating loneliness or stress, means for generating feedback based on an analysis result and transmitting the feedback to a medical expert or a caregiver, means for displaying or notifying the elderly person of a message for prompting regular interaction, means for proposing a social activity suitable for the elderly person, and means for evaluating a state of the elderly person after the proposal and improving accuracy of the system.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In an aging society, many elderly people feel loneliness and social isolation, which can have a negative impact on their physical and mental health. If this situation is left unchecked, the sense of isolation will deepen, increasing the risk of later-stage depression and dementia. Therefore, there is a need to identify situations in which elderly people feel lonely early and take appropriate measures, but currently there is a lack of efficient and effective means to do so. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides the following means.

[0006] means for recording audio data;

[0007] means for transmitting the recorded voice data to a server;

[0008] means for converting voice data into text at the server;

[0009] A means of analyzing text data and extracting keywords that indicate loneliness and stress;

[0010] a means for generating and transmitting feedback based on the analysis results to healthcare professionals and caregivers;

[0011] A means for displaying or notifying a message to encourage regular dialogue to the elderly person;

[0012] A means of suggesting suitable social activities for older people;

[0013] A means to evaluate the condition of the elderly person after the proposal and improve the accuracy of the system;

[0014] It is a system including:

[0015] This system can detect feelings of loneliness early through everyday conversations among elderly people and provide appropriate measures and advice, thereby reducing feelings of loneliness among the elderly and promoting their physical and mental health.

[0016] "Voice data" refers to electrical signals or files that digitally record the voice of the user (elderly person).

[0017] "Recording means" refers to a device or function that records the elderly person's everyday conversations and saves the audio data.

[0018] The "server" is a central control system that receives voice data sent from the terminals, analyzes it, and generates feedback.

[0019] A "text conversion means" is a process or device that uses speech recognition technology to convert voice data into written information.

[0020] The "keyword extraction means" is a function or algorithm that identifies and extracts specific words or phrases that indicate loneliness or stress from text data.

[0021] A "feedback means" is a process or device that generates and transmits appropriate information or advice to medical professionals or caregivers based on the results of the analysis.

[0022] "Dialogue promotion means" refers to a method or device for displaying or notifying elderly people of messages that encourage them to have regular conversations.

[0023] "Social activity suggestion means" refers to a function or system that informs elderly people of social activities and events suitable for them based on the analysis results.

[0024] The "condition evaluation means" refers to a process or device for re-analyzing the condition of the elderly person after the proposal and evaluating the effectiveness and accuracy of the system. [Brief explanation of the drawings]

[0025] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0026] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0027] First, the terms used in the following description will be explained.

[0028] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0029] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0030] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0031] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0032] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0033] [First embodiment]

[0034] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0035] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0036] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0037] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0038] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0039] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0040] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0041] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0042] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0043] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0044] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0045] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0046] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0047] The main components of this system are: "means for recording voice data," "means for sending voice data to a server," "means for converting voice data into text," "means for analyzing text data and extracting keywords," "means for generating feedback and sending it to medical professionals and caregivers," "means for encouraging regular dialogue with the elderly," "means for suggesting social activities," and "means for assessing the condition of the elderly and improving the accuracy of the system."

[0048] An embodiment of this system is described in detail below.

[0049] Data collection

[0050] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[0051] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[0052] Conversation Analysis

[0053] The server processes the received voice data and first converts the voice into text using voice recognition technology.

[0054] The server uses a generative AI to extract keywords that indicate loneliness or stress from the text data. For example, if keywords such as "lonely" or "solitude" appear frequently, it is determined that the user is feeling lonely.

[0055] The server also performs sentiment analysis, assessing the overall emotional tone (negative, positive, neutral) of the text data.

[0056] Feedback and Advice

[0057] The server generates a feedback report based on the analysis, including keyword frequency and emotional tone, which is sent to medical professionals and caregivers.

[0058] For example, specific advice is provided such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[0059] Promoting dialogue

[0060] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0061] Social activity proposals

[0062] The server will suggest suitable social activities for seniors, such as sending notifications recommending them to join hobby clubs or online meetups.

[0063] Specifically, you might receive a notification such as, "There's a painting class at your local community center this Saturday. Would you like to join?"

[0064] Evaluation and Improvement

[0065] The server then reanalyzes the elderly person's conversation data after the suggestions are made and evaluates whether their sense of loneliness has been alleviated, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[0066] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system.

[0067] Specific examples

[0068] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server, converted into text, and analyzed. The result is that the keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in dialogue by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and their quality of life improves.

[0069] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[0073] Step 2:

[0074] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[0075] Step 3:

[0076] The server decrypts the received encrypted voice data and converts the voice into text information. It uses voice recognition technology to carry out the process of converting the voice data into text data.

[0077] Step 4:

[0078] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0079] Step 5:

[0080] The server evaluates the emotional tone of the entire text data, analyzing it as negative, positive, or neutral to understand the user's emotional state.

[0081] Step 6:

[0082] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords and an evaluation of emotional tone, and is sent to medical professionals and caregivers.

[0083] Step 7:

[0084] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[0085] Step 8:

[0086] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0087] Step 9:

[0088] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[0089] Step 10:

[0090] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0091] Step 11:

[0092] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0093] Step 12:

[0094] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate voice recognition technology.

[0095] Example 1

[0096] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0097] Elderly people are prone to feelings of loneliness and stress, which often have a negative impact on their health. Conventional measures make it difficult to detect early signs of loneliness and stress and take appropriate measures. In addition, feedback and responses are often inappropriate, leading to the problem of not being able to adequately support the physical and mental health of elderly people.

[0098] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0099] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data using a generative AI model and extracting keywords indicative of loneliness or stress, a means for generating a feedback report based on the extracted keywords and emotional tone and sending it to a medical professional or caregiver, a means for displaying or notifying a message encouraging the elderly to have regular conversations, a means for suggesting social activities suitable for the elderly, and a means for reanalyzing the elderly's conversation data after the suggestions and evaluating it to improve the accuracy of the system. This makes it possible to detect signs of loneliness or stress early through the elderly's everyday conversations and take appropriate countermeasures.

[0100] "Voice data" is data in digital form obtained from a user's voice.

[0101] A "terminal" is an electronic device for recording and transmitting voice data, including smart speakers and tablets.

[0102] A "server" is a computer system for receiving and processing audio data.

[0103] The "means for converting to text" is a function for converting voice data into text information, and voice recognition technology is used for this.

[0104] A "generative AI model" is an artificial intelligence model that uses machine learning and natural language processing techniques to analyze data and extract specific information.

[0105] "Keywords" are specific words or phrases that indicate loneliness or stress.

[0106] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[0107] "Messages encouraging regular dialogue" are notifications and messages that are displayed to encourage everyday dialogue with elderly people.

[0108] "Means for suggesting social activities" is a suggestion function for providing elderly people with opportunities for community activities and interaction.

[0109] The "means of evaluation" is an evaluation function for reanalyzing the condition of the elderly person and improving the accuracy of the system.

[0110] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0111] Data collection

[0112] 1. The device records voice data. For example, an elderly person (user) uses a smart speaker or tablet to record their everyday conversations. The recorded voice data is periodically encrypted by the device and sent to a server in a privacy-protected manner. AES (Advanced Encryption Standard) technology is used for encryption, and the data is sent using the HTTPS protocol.

[0113] Conversation Analysis

[0114] 2. The server processes the received voice data and first converts it into text using the Google Cloud Speech-to-Text API. For example, the resulting text is "I was gardening today."

[0115] 3. The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data. Specifically, it uses a machine learning algorithm to extract keywords such as "lonely," "solitude," and "alone." An example of a prompt is, "Please extract keywords indicating loneliness from this text data."

[0116] 4. The server then performs further sentiment analysis to assess the overall emotional tone (negative, positive, neutral) of the text data using the Sentiment Analysis API.

[0117] Feedback and Advice

[0118] 5. The server generates a feedback report based on the analysis results. The report includes the frequency of keywords and emotional tone, and is sent to medical professionals and caregivers. For example, a report might say, "User A has been using the word lonely a lot recently. The emotional tone is negative."

[0119] Promoting dialogue

[0120] 6. The device will display or notify the elderly with messages encouraging regular conversations. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" will be sent.

[0121] Social activity proposals

[0122] 7. The server will suggest suitable social activities for the elderly. For example, a notification will be sent to the elderly saying, "There will be an art class at the local community center this Saturday. Would you like to join?"

[0123] Evaluation and Improvement

[0124] 8. The server reanalyzes the elderly person's conversation data after the suggestion and evaluates whether the sense of loneliness has been reduced, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[0125] 9. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system, which involves retraining the machine learning algorithm and introducing new datasets.

[0126] Specific examples

[0127] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server and converted into text using the Google Cloud Speech-to-Text API. A generative AI model then analyzes the text data, finding that the keyword "lonely" appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A with a message asking, "Have you been able to talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases and his quality of life improves.

[0128] In this way, the present invention is a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] Step 1:

[0131] The device records voice data. As input, it acquires the elderly person's (user's) everyday conversation. As operation, when the user speaks to the smart speaker or tablet, the device records the voice in digital format. As output, it generates the recorded voice data.

[0132] Step 2:

[0133] The terminal encrypts the recorded voice data and sends it to the server. As input, there is the recorded voice data. As operation, the terminal encrypts the voice data using AES technology to ensure the security of the data. Then, it sends the encrypted data to the server via HTTPS protocol. As output, the encrypted voice data is sent to the server.

[0134] Step 3:

[0135] The server converts the received voice data into text. As input, it receives encrypted voice data. In operation, the server uses speech recognition technology to convert the data into text format. Specifically, it uses the Google Cloud Speech-to-Text API. As output, it generates text data.

[0136] Step 4:

[0137] The server uses a generative AI model to extract keywords indicating loneliness and stress from text data. The input is text data. The operation is to apply a machine learning algorithm to detect specific keywords. An example of a prompt is "Please extract keywords indicating loneliness from this text data." The output is a list of extracted keywords.

[0138] Step 5:

[0139] The server performs sentiment analysis and evaluates the emotional tone of the entire text data. The input is text data. The operation is to evaluate the emotional tone of the data using the Sentiment Analysis API. The output is an evaluation result of the emotional tone (negative, positive, neutral).

[0140] Step 6:

[0141] The server generates a feedback report based on the analysis results. The inputs are a list of keywords and the emotional tone evaluation results. In operation, the server integrates these data to create a feedback report. As an output, a feedback report is generated. This report includes the frequency of keywords and the emotional tone.

[0142] Step 7:

[0143] The server sends the generated feedback report to the healthcare professional or caregiver. As input, there is a feedback report. As action, the server sends the report using secure email or a dedicated dashboard. As output, the feedback report has been sent.

[0144] Step 8:

[0145] The device displays or notifies the elderly with messages encouraging regular conversation. The input is an instruction from the system. The action is to display a notification on the screen or play a message by voice. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" is sent. The output is to encourage the elderly to have a conversation.

[0146] Step 9:

[0147] The server suggests suitable social activities for the elderly. As input, it has the feedback report and the profile data of the elderly. As operation, the system selects suitable activities for the elderly and generates a notification. For example, a notification may be generated such as "There is a painting class at your local community center this Saturday. Would you like to join?" As output, the suggested social activities are sent to the elderly.

[0148] Step 10:

[0149] The server re-analyzes the elderly person's conversation data after the suggestions and evaluates whether their feelings of loneliness have been reduced. The input is the new speech data after the suggestions. The operation is to re-apply the method from steps 3 to 5 described above to obtain new analysis results. The output is an evaluation result of whether the signs of loneliness have been reduced.

[0150] Step 11:

[0151] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy. The input is the reanalysis results. The operation involves retraining the machine learning algorithm and introducing a new dataset. The output is a new system with improved algorithm accuracy.

[0152] (Application example 1)

[0153] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0154] Situations in which elderly people are prone to feeling loneliness and stress in their daily lives can have a negative impact on their physical and mental health. Furthermore, if efforts to reduce loneliness are insufficient, there is a higher risk that an emergency at home will go unnoticed. The present invention aims to collect and analyze the everyday conversations of elderly people, detect signs of loneliness early, provide appropriate countermeasures, and reduce the risks associated with loneliness.

[0155] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0156] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and extracting keywords indicating loneliness or stress using a generative AI model, and a means for promptly notifying family members or caregivers if an abnormality is detected. This makes it possible to detect loneliness or stress in the elderly person's everyday conversations early on and take necessary measures promptly.

[0157] "Voice data" refers to data in which voice is electronically recorded, and includes everyday conversations between elderly people.

[0158] "Recording means" refers to devices or technologies for storing audio data, including, for example, smartphones, tablets, and smart speakers.

[0159] A "server" is a computer system that receives, processes, and analyzes voice data.

[0160] "Means for converting to text" refers to technology for converting voice data into text data in sentence format, and includes voice recognition software and APIs.

[0161] "Keyword extraction means" refers to technology for identifying and extracting specific words and phrases related to loneliness and stress from text data.

[0162] A "generative AI model" is an artificial intelligence algorithm that analyzes the meaning and sentiment of text data based on large datasets.

[0163] The "means for generating feedback" is a technology that generates information to suggest measures to medical professionals and caregivers based on the results of text analysis.

[0164] "Means of notification" refers to technologies for quickly transmitting analysis results and feedback information to family members and caregivers, and includes email and messaging services.

[0165] "Means for encouraging regular dialogue" refers to technology that displays or notifies elderly people of messages encouraging dialogue on a regular basis.

[0166] "Means for suggesting social activities" refers to techniques for suggesting events and club activities suitable for the elderly.

[0167] "Means for improving the accuracy of the system" refers to methods or techniques for updating the system's algorithms or databases based on the analysis results, thereby improving the accuracy and efficiency of the analysis.

[0168] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0169] Data collection

[0170] The device records audio data. For example, an elderly user might use a smartphone to record everyday conversations. The device periodically encrypts the recorded audio data and transmits it to a server in a privacy-protected manner. The hardware used is a standard Android or iOS device, and encryption is performed using the Python "pycryptodome" library.

[0171] Conversation Analysis

[0172] The server processes the received voice data and first converts the speech to text. This conversion is performed using Google's Cloud Speech-to-Text API. Next, the server uses a generative AI model (e.g., OpenAI GPT-4) to extract keywords indicating loneliness or stress from the text data. For example, if keywords such as "lonely" and "solitude" appear frequently, it is determined that the user is feeling lonely. The server then performs sentiment analysis to evaluate the emotional tone (negative, positive, or neutral) of the entire text data.

[0173] Feedback and Advice

[0174] The server generates a feedback report based on the analysis results. This report includes keyword frequency, emotional tone, and the results of emotion analysis using a generative AI model, and is sent to family members or caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently. You should talk to him more often" is provided. If an abnormality is detected, the family or caregiver is promptly notified.

[0175] Promoting dialogue

[0176] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0177] Social activity proposals

[0178] The server suggests social activities suitable for seniors. For example, notifications are sent to seniors recommending them to join hobby clubs or online social gatherings. Specifically, a notification might say, "There's a painting class at the local community center this Saturday. Would you like to join?"

[0179] Evaluation and Improvement

[0180] The server reanalyzes the elderly person's conversation data after the proposal and evaluates whether their feelings of loneliness have been alleviated. For example, it checks whether the frequency of keywords indicating loneliness has decreased. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy.

[0181] Specific examples

[0182] For example, user A might say on his smartphone, "I've been alone a lot lately and I feel lonely." This voice data is sent to the server, converted into text, and analyzed. The keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to the family, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in conversation by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and his quality of life improves.

[0183] Prompt Sentence Examples

[0184] Audio data: "I've been spending a lot of time alone lately and I feel lonely."

[0185] Text data: "I've been spending a lot of time alone lately and I feel lonely."

[0186] Keywords: "Spending a lot of time alone", "lonely"

[0187] Feedback: "User B feels lonely. Please consider visiting or contacting them regularly."

[0188] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0189] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0190] Step 1:

[0191] A user uses a smartphone to have everyday conversations.

[0192] (Input) User's everyday conversational voice.

[0193] (Output) Audio data recorded on a smartphone.

[0194] (Specific operation) When a user says "I'm lonely" to their smartphone, the voice is recorded by the smartphone's built-in microphone.

[0195] Step 2:

[0196] The device periodically encrypts the recorded audio data and sends it to the server.

[0197] (Input) Recorded audio data.

[0198] (Output) Encrypted audio data.

[0199] (Specific operation) The smartphone encrypts the voice data using AES (Advanced Encryption Standard) and sends it to the server via HTTPS.

[0200] Step 3:

[0201] The server converts the received voice data into text.

[0202] (Input) Encrypted audio data.

[0203] (Output) Text data.

[0204] (Specific operation) The server decrypts the encrypted voice data and converts the voice data into text data using Google's Cloud Speech-to-Text API.

[0205] Step 4:

[0206] The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data.

[0207] (Input) Text data.

[0208] (Output) Extracted keywords.

[0209] (Specific operation) The server inputs the text data into a natural language processing algorithm (e.g., OpenAI GPT-4) and extracts keywords such as "lonely" and "solitude."

[0210] Step 5:

[0211] The server also evaluates the emotional tone of the entire text data.

[0212] (Input) Text data and extracted keywords.

[0213] (Output) Sentiment analysis result (negative, positive, neutral).

[0214] (Specific operation) Using a generative AI model, the emotional tone of the entire text data is analyzed to determine emotional categories such as "negative" or "positive."

[0215] Step 6:

[0216] The server generates a feedback report based on the analysis results.

[0217] (Input) Extracted keywords and sentiment analysis results.

[0218] (Output) Feedback report.

[0219] (Specific operation) Based on the extracted keywords and the results of sentiment analysis, the server generates a feedback report such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[0220] Step 7:

[0221] If an abnormality is detected, the server will promptly notify family members or caregivers.

[0222] (Input) Feedback report.

[0223] (Output) A warning notification.

[0224] (Specific operation) The server uses the contact information of family members or caregivers to send messages via email or SMS such as "User A is feeling lonely, so we recommend regular phone consultations."

[0225] Step 8:

[0226] The device displays or notifies the elderly person of messages encouraging regular dialogue.

[0227] (Input) Instructions from the server.

[0228] (Output) Messages that are displayed on the terminal.

[0229] (Specific action) The smartphone displays a message such as, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0230] Step 9:

[0231] The server suggests social activities suitable for seniors.

[0232] (Input) Analysis results and social activity information in the database.

[0233] (Output) The proposal message.

[0234] (Specific operation) The server generates a message such as "There's a painting class at the local community center this Saturday. Would you like to join?" and sends it to the terminal.

[0235] Step 10:

[0236] The server reanalyzes the elderly's conversation data after the proposal to improve the accuracy of the system.

[0237] (Input) Re-collected audio data.

[0238] (Output) Updated algorithm to improve system accuracy.

[0239] (Specific operation) The server reanalyzes the elderly's conversation data to see if the frequency of keywords indicating loneliness has decreased, and uses the results to update the algorithm.

[0240] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0241] This invention relates to a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes their emotions. The system aims to reduce the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[0242] The main components of this system are: a means for recording voice data, a means for sending voice data to a server, a means for converting voice data into text, a means for analyzing text data and extracting keywords, an emotion engine, a means for generating feedback and sending it to medical professionals and caregivers, a means for encouraging regular dialogue with the elderly, a means for suggesting social activities, and a means for assessing the condition of the elderly and improving the accuracy of the system.

[0243] An embodiment of this system is described in detail below.

[0244] Data collection

[0245] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[0246] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[0247] Conversation Analysis

[0248] The server processes the received voice data and first converts the voice data into text, then uses voice recognition technology to convert the voice data into text.

[0249] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0250] emotion recognition

[0251] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[0252] Feedback and Advice

[0253] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[0254] For example, specific advice may be provided such as, "User A has been using the word lonely a lot recently, and the emotion engine's evaluation has confirmed that he has strong negative emotions. We recommend that you increase the frequency of conversations with him and suggest some hobby activities."

[0255] Promoting dialogue

[0256] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0257] Social activity proposals

[0258] The server suggests social activities suitable for seniors, using generative AI to select activities based on the user's interests and sending that information to the device.

[0259] Specifically, the device will send a notification to the elderly such as, "There will be a painting class at the local community center this Saturday. Would you like to join?"

[0260] Evaluation and Improvement

[0261] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been reduced. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0262] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[0263] Specific examples

[0264] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text. The server performs analysis and detects negative emotions through the keyword "lonely" and evaluation by the emotion engine. The server then generates a feedback report and notifies medical professionals that "User B is likely feeling very lonely." The device then sends User B messages such as "Have you talked to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[0265] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[0266] The processing flow will be explained below.

[0267] Step 1:

[0268] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[0269] Step 2:

[0270] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[0271] Step 3:

[0272] The server decrypts the received encrypted voice data and converts the voice data into text data. It uses voice recognition technology to carry out the process of converting the voice data into text information.

[0273] Step 4:

[0274] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0275] Step 5:

[0276] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[0277] Step 6:

[0278] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[0279] Step 7:

[0280] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[0281] Step 8:

[0282] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0283] Step 9:

[0284] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[0285] Step 10:

[0286] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0287] Step 11:

[0288] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0289] Step 12:

[0290] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[0291] Specific examples

[0292] For example, if user C says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text data.

[0293] The server then performs an analysis and detects that the keyword "lonely" appears frequently, and the emotion engine identifies negative emotions.

[0294] The server then generates a feedback report and notifies the medical professional of the assessment result that "User C is likely to be feeling a strong sense of loneliness."

[0295] The terminal sends a message to user C prompting a conversation, such as "Have you talked to anyone today?"

[0296] Additionally, the server suggests social activities to User C, such as "Would you like to join a local online book club?"

[0297] This reduces User C's sense of loneliness and improves his quality of life.

[0298] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[0299] Example 2

[0300] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0301] There is a lack of systems that can detect elderly people's feelings of loneliness and emotional states early and take appropriate countermeasures. Conventional systems have difficulty analyzing elderly people's emotional states from their everyday conversations and providing feedback. Furthermore, there are limited means to effectively communicate this data to medical professionals and caregivers. The present invention aims to solve these problems, reduce elderly people's feelings of loneliness, and improve their quality of life.

[0302] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0303] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and voice data and evaluating the emotional state, and a means for generating a feedback report based on the analysis results and sending it to a medical professional or caregiver, thereby enabling the emotional state of an elderly person to be analyzed quickly and accurately from their everyday conversations and appropriate countermeasures to be implemented.

[0304] "Audio data" refers to digital audio files that record everyday conversations between elderly people.

[0305] A "server" is an information processing device that processes and analyzes voice data and generates feedback.

[0306] "Text data" refers to character information converted from voice data using voice recognition technology.

[0307] "Keywords" are important words in the text data that indicate feelings of loneliness or stress.

[0308] An "emotion engine" is software that analyzes voice and text data to assess emotional states.

[0309] "Analysis results" are information obtained from analyzing text data and evaluating it using an emotion engine.

[0310] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[0311] "Medical professionals" are professionals such as doctors and nurses who receive the feedback reports and use them to help care for older people.

[0312] A "caregiver" is a person who receives the feedback report and provides support for the elderly person's daily life.

[0313] "Promoting regular dialogue" means displaying or notifying elderly people of messages encouraging dialogue on a regular basis.

[0314] "Social activity suggestions" refers to providing information on social activities that encourage participation based on the interests of the elderly.

[0315] "Evaluating the state" involves reanalyzing the elderly person's conversation data after the proposal to confirm any changes in their feelings of loneliness or emotional state.

[0316] "Improving the accuracy of the system" means updating the analysis algorithm, introducing new keywords, and improving emotion recognition technology to improve the system's performance.

[0317] This invention is a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes emotions. The main purpose of this system is to alleviate the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[0318] Data collection

[0319] Users use devices such as smart speakers and tablets to record voice data. For example, a user may record their everyday conversations using a smart speaker (a common name for such devices). The device periodically encrypts the recorded voice data and transmits it to a server in a privacy-protected state.

[0320] Conversation Analysis

[0321] The server processes the received voice data and first converts it into text using speech recognition technology (e.g., a speech recognition API). The server then analyzes the text using a generative AI model (e.g., a natural language processing tool). The analysis includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0322] emotion recognition

[0323] The server's emotion engine analyzes both voice and text data to assess the elderly person's emotional state. An emotion recognition API (generic name) can be used to infer emotions from tone of voice, speaking style, and text content.

[0324] Feedback and Advice

[0325] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[0326] Promoting dialogue

[0327] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0328] Social activity proposals

[0329] The server suggests social activities suitable for seniors. Using a generative AI model, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior, such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0330] Evaluation and Improvement

[0331] The server re-analyzes the elderly person's conversation data after the proposal and evaluates whether their sense of loneliness has been reduced. It then performs another speech-to-text conversion and keyword extraction to confirm any changes in their condition. It also aims to improve the system's accuracy based on the evaluation results, by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[0332] Specific examples

[0333] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text using a speech recognition API. The server then analyzes the text using a generative AI model and detects negative emotions based on the keyword "lonely" and the evaluation of an emotion engine. The server then generates a feedback report and notifies medical professionals with specific content, such as "User B is likely feeling very lonely." The device then sends User B a message such as "Did you talk to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[0334] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[0335] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0336] Step 1: Recording audio data

[0337] The device records the user's everyday conversations. For example, when a user says "Good morning" using a smart speaker or tablet, the conversation is recorded as audio data. The input is the user's voice, and the output is digital audio data.

[0338] Step 2: Encrypt and transmit the audio data

[0339] The device encrypts the recorded voice data and sends it to the server. For example, voice data recorded at a fixed time every day is encrypted using the AES (Advanced Encryption Standard) algorithm. The encrypted data is then uploaded to the server. The input is voice data, and the output is encrypted voice data.

[0340] Step 3: Convert audio data to text

[0341] The server decrypts the received encrypted voice data and converts it into text data using a speech recognition API (for example, Google Speech-to-Text API). The input is encrypted voice data, and the output is text data. For example, voice data saying "Good morning" is converted into the string "Good morning."

[0342] Step 4: Analyzing the text data

[0343] The server analyzes the text data using a generative AI model (e.g., a natural language processing tool). The analysis involves extracting keywords that indicate loneliness or stress. The input is the text data, and the output is the analysis results. For example, keywords such as "lonely" and "alone" are extracted, and their frequency is calculated.

[0344] Step 5: Emotion Recognition

[0345] The server's emotion engine analyzes both voice and text data to evaluate the user's emotional state. It uses emotion recognition APIs (such as IBM Watson or Microsoft Azure Emotion API). The input is text and voice data, and the output is an emotional evaluation result. For example, emotions such as "sadness" or "anxiety" are detected.

[0346] Step 6: Generate a feedback report

[0347] The server generates a feedback report based on the analysis results. The report includes the frequency of extracted keywords, emotional tone, and the emotion engine's evaluation results. The input is the analysis results and the emotion evaluation results, and the output is a feedback report. For example, it may contain specific content such as, "User A has been using the word 'lonely' a lot recently, and the emotion evaluation has confirmed a negative state."

[0348] Step 7: Submit your feedback

[0349] The server generates and sends the feedback report to the healthcare professional or caregiver. For example, the report is sent via email or a web application using the notification function of the internal system. The input is the feedback report, and the output is the report being sent to the healthcare professional or caregiver.

[0350] Step 8: Promote regular dialogue

[0351] The terminal displays or notifies the user of messages that prompt regular interactions. For example, a voice or text message such as "Tell me how you're feeling today" is emitted from the terminal at a fixed time every day. The input is an instruction from the server, and the output is a message that is displayed to the user.

[0352] Step 9: Propose a social activity

[0353] The server selects social activities that match the user's interests and sends that information to the device. For example, a notification may be sent saying, "There's an online book club this Saturday. Would you like to join?" The input is the user's interest data and the analysis results of the generative AI, and the output is the suggested content.

[0354] Step 10: Assess the condition and improve the system

[0355] The server reanalyzes the user's conversation data after the proposal and evaluates whether the sense of loneliness has been reduced. Based on the evaluation results, the system's performance can be improved by updating the analysis algorithm or introducing new keywords. The input is the reanalyzed data, and the output is the system improvements. For example, "Since a reduction in the sense of loneliness was observed, we will introduce a new emotion recognition algorithm."

[0356] (Application example 2)

[0357] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0358] The challenge is to provide an effective dialogue system that can quickly detect and alleviate the feelings of loneliness and negative emotions that elderly people may experience in brick-and-mortar stores. It is also necessary to support elderly people in communicating more smoothly with store staff and other customers.

[0359] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for recording voice data, means for transmitting the recorded voice data to the server, means for converting the voice data to text in the server, means for analyzing the text data and extracting keywords indicating loneliness or stress, means for generating feedback based on the analysis results and transmitting the feedback to a medical professional or caregiver, means for displaying or notifying the elderly person of a message encouraging regular conversation, means for suggesting social activities suitable for the elderly, means for evaluating the elderly person's condition after the suggestions and improving the system's accuracy, means for suggesting conversation topics based on the analysis results, and means for providing store staff with information useful for conversations with the elderly. This makes it possible to detect elderly people's loneliness or negative emotions early on and to suggest effective conversations and social activities.

[0360] "Means for recording voice data" refers to a device or program for collecting and recording voices uttered by the elderly person in digital form.

[0361] The "means for transmitting recorded voice data to a server" refers to a device or program for securely transferring recorded voice data to a server via a network.

[0362] The "means for converting voice data into text on the server" is a program that converts voice data into text format using natural language processing technology.

[0363] The "means for analyzing text data and extracting keywords that indicate loneliness and stress" is a program that analyzes text data using machine learning models and natural language processing technology to identify keywords related to loneliness and stress.

[0364] The "means for generating feedback based on the analysis results and sending it to a medical professional or caregiver" is a program that creates a feedback report based on the analysis results and sends it to a medical professional or caregiver.

[0365] "Means for displaying or notifying messages encouraging elderly people to have regular conversations" refers to a device or program that displays or notifies elderly people in voice or text format with messages to encourage them to have regular conversations.

[0366] The "means for suggesting social activities suitable for the elderly" is a program that suggests appropriate social activities based on the interests and emotional state of each elderly person.

[0367] The "means of evaluating the elderly person's condition after the proposal and improving the accuracy of the system" is a program for reanalyzing the elderly person's reaction and condition after the proposal and improving the accuracy of the system's analysis and the effectiveness of the proposal.

[0368] The "means for proposing a topic for conversation based on the analysis result" is a program that provides an appropriate topic for conversation in accordance with the analyzed emotional state.

[0369] "Means for providing store staff with information that will be useful when interacting with the elderly" refers to a program that provides store staff with the information and suggestions they need to smoothly interact with the elderly.

[0370] The present invention relates to a system for elderly people to use in brick-and-mortar stores. This system is designed to detect loneliness and negative emotions in elderly people at an early stage and promote dialogue. Specific embodiments of the system are described below.

[0371] Data collection

[0372] The device records voice data. When an elderly person (user) speaks using smart glasses or a smartphone in a physical store, the voice data is recorded. For example, in a cafe or shopping store, the user may say, "I've been feeling lonely lately because I haven't been able to see my friends."

[0373] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner using a secure communication protocol (e.g., HTTPS).

[0374] Conversation Analysis

[0375] The server processes the received voice data and first converts it into text using software with speech recognition technology (for example, Python's speech_recognition library).

[0376] The server analyzes the text data using a generative AI model. This analysis involves extracting keywords that indicate loneliness or stress and measuring the frequency of these keywords. For example, if keywords such as "lonely" or "alone" appear frequently, it determines that particular attention is needed.

[0377] Emotion Recognition and Feedback

[0378] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, using technology that infers emotions from tone of voice, speaking style, and text content.

[0379] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[0380] Promoting dialogue and proposing social activities

[0381] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0382] The server suggests social activities suitable for seniors. Using generative AI, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0383] Dialogue suggestions

[0384] The server then suggests conversation topics based on the analysis results. This function allows users to have smoother conversations with store staff and other customers. For example, if the emotion analysis results indicate "loneliness," the server might suggest a conversation topic such as, "How about talking about the recent weather or your hobbies?"

[0385] Examples and prompts

[0386] For example, if an elderly person says, "I've been feeling lonely lately because I haven't been able to see my friends," at a brick-and-mortar cafe, this voice data is sent to the server and converted into text. The analysis results indicate the emotion of "loneliness," and the app suggests topics for conversation, such as, "How about talking about the weather recently or your hobbies?"

[0387] Specific prompt examples:

[0388] User text: "I've been missing my friends lately."

[0389] Emotion analysis result: {'emotion': 'sadness'}

[0390] Suggested topic: "Why don't we talk about the weather these days or our hobbies?"

[0391] This will help reduce the sense of loneliness among the elderly and improve their quality of life, while providing store staff with specific support to facilitate smoother conversations with the elderly.

[0392] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0393] Step 1:

[0394] The device records the voice data spoken by elderly people in a physical store. For example, if a user says, "I miss my friends these days," in a cafe, the device digitally records the voice. The input is the user's voice data, and the output is the recorded voice file.

[0395] Step 2:

[0396] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner. A secure communication protocol (HTTPS) is used for data transmission. The input is the recorded audio file, and the output is the transmission of the encrypted audio file.

[0397] Step 3:

[0398] The server converts the received voice data into text using speech recognition technology. Here, we use the Python speech_recognition library. The input is an encrypted audio file, and the output is text data.

[0399] Step 4:

[0400] The server uses a generative AI model to analyze the text data and extract keywords that indicate loneliness and stress. The input is the text data converted from speech, and the output is the extracted keywords and their frequency. For example, keywords such as "lonely" and "alone" are output.

[0401] Step 5:

[0402] The server's emotion engine further analyzes the voice and text data to assess the elderly person's emotional state. The emotion engine uses a generative AI model, whose input is the extracted keywords and their frequency, and whose output is the emotional state (e.g., "loneliness").

[0403] Step 6:

[0404] The server generates a feedback report based on the analysis results and sends it to medical professionals and caregivers. The input is the evaluation result of the emotion engine, and the output is the feedback report. The report includes the frequency of keywords and emotional tone.

[0405] Step 7:

[0406] The device displays or notifies the elderly with messages encouraging regular conversation. The input is a feedback report, and the output is a notification message (e.g., "Talking to someone every day is good for your health. Please tell us how you feel today.").

[0407] Step 8:

[0408] The server suggests social activities suitable for seniors. It uses generative AI to select activities based on the user's interests and sends that information to the device. The input is the analysis results from the generative AI model, and the output is suggested social activity information. For example, a suggestion might be output such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0409] Step 9:

[0410] The server then suggests conversation topics based on the analysis results. The input is emotional state and keywords, and the output is the conversation topic. For example, a suggestion might be, "Why don't we talk about the recent weather and your hobbies?"

[0411] Step 10:

[0412] The server provides store staff with information that is useful when interacting with elderly people. The input is the emotion analysis results and conversation topics, and the output is conversation support information. This allows staff to communicate smoothly with elderly people.

[0413] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0414] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0415] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0416] [Second embodiment]

[0417] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0418] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0419] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0420] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0421] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0422] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0423] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0424] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0425] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0426] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0427] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0428] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0429] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0430] The main components of this system are: "means for recording voice data," "means for sending voice data to a server," "means for converting voice data into text," "means for analyzing text data and extracting keywords," "means for generating feedback and sending it to medical professionals and caregivers," "means for encouraging regular dialogue with the elderly," "means for suggesting social activities," and "means for assessing the condition of the elderly and improving the accuracy of the system."

[0431] An embodiment of this system is described in detail below.

[0432] Data collection

[0433] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[0434] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[0435] Conversation Analysis

[0436] The server processes the received voice data and first converts the voice into text using voice recognition technology.

[0437] The server uses a generative AI to extract keywords that indicate loneliness or stress from the text data. For example, if keywords such as "lonely" or "solitude" appear frequently, it is determined that the user is feeling lonely.

[0438] The server also performs sentiment analysis, assessing the overall emotional tone (negative, positive, neutral) of the text data.

[0439] Feedback and Advice

[0440] The server generates a feedback report based on the analysis, including keyword frequency and emotional tone, which is sent to medical professionals and caregivers.

[0441] For example, specific advice is provided such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[0442] Promoting dialogue

[0443] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0444] Social activity proposals

[0445] The server will suggest suitable social activities for seniors, such as sending notifications recommending them to join hobby clubs or online meetups.

[0446] Specifically, you might receive a notification such as, "There's a painting class at your local community center this Saturday. Would you like to join?"

[0447] Evaluation and Improvement

[0448] The server then reanalyzes the elderly person's conversation data after the suggestions are made and evaluates whether their sense of loneliness has been alleviated, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[0449] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system.

[0450] Specific examples

[0451] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server, converted into text, and analyzed. The result is that the keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in dialogue by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and their quality of life improves.

[0452] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0453] The processing flow will be explained below.

[0454] Step 1:

[0455] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[0456] Step 2:

[0457] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[0458] Step 3:

[0459] The server decrypts the received encrypted voice data and converts the voice into text information. It uses voice recognition technology to carry out the process of converting the voice data into text data.

[0460] Step 4:

[0461] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0462] Step 5:

[0463] The server evaluates the emotional tone of the entire text data, analyzing it as negative, positive, or neutral to understand the user's emotional state.

[0464] Step 6:

[0465] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords and an evaluation of emotional tone, and is sent to medical professionals and caregivers.

[0466] Step 7:

[0467] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[0468] Step 8:

[0469] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0470] Step 9:

[0471] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[0472] Step 10:

[0473] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0474] Step 11:

[0475] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0476] Step 12:

[0477] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate voice recognition technology.

[0478] Example 1

[0479] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0480] Elderly people are prone to feelings of loneliness and stress, which often have a negative impact on their health. Conventional measures make it difficult to detect early signs of loneliness and stress and take appropriate measures. In addition, feedback and responses are often inappropriate, leading to the problem of not being able to adequately support the physical and mental health of elderly people.

[0481] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0482] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data using a generative AI model and extracting keywords indicative of loneliness or stress, a means for generating a feedback report based on the extracted keywords and emotional tone and sending it to a medical professional or caregiver, a means for displaying or notifying a message encouraging the elderly to have regular conversations, a means for suggesting social activities suitable for the elderly, and a means for reanalyzing the elderly's conversation data after the suggestions and evaluating it to improve the accuracy of the system. This makes it possible to detect signs of loneliness or stress early through the elderly's everyday conversations and take appropriate countermeasures.

[0483] "Voice data" is data in digital form obtained from a user's voice.

[0484] A "terminal" is an electronic device for recording and transmitting voice data, including smart speakers and tablets.

[0485] A "server" is a computer system for receiving and processing audio data.

[0486] The "means for converting to text" is a function for converting voice data into text information, and voice recognition technology is used for this.

[0487] A "generative AI model" is an artificial intelligence model that uses machine learning and natural language processing techniques to analyze data and extract specific information.

[0488] "Keywords" are specific words or phrases that indicate loneliness or stress.

[0489] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[0490] "Messages encouraging regular dialogue" are notifications and messages that are displayed to encourage everyday dialogue with elderly people.

[0491] "Means for suggesting social activities" is a suggestion function for providing elderly people with opportunities for community activities and interaction.

[0492] The "means of evaluation" is an evaluation function for reanalyzing the condition of the elderly person and improving the accuracy of the system.

[0493] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0494] Data collection

[0495] 1. The device records voice data. For example, an elderly person (user) uses a smart speaker or tablet to record their everyday conversations. The recorded voice data is periodically encrypted by the device and sent to a server in a privacy-protected manner. AES (Advanced Encryption Standard) technology is used for encryption, and the data is sent using the HTTPS protocol.

[0496] Conversation Analysis

[0497] 2. The server processes the received voice data and first converts it into text using the Google Cloud Speech-to-Text API. For example, the resulting text is "I was gardening today."

[0498] 3. The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data. Specifically, it uses a machine learning algorithm to extract keywords such as "lonely," "solitude," and "alone." An example of a prompt is, "Please extract keywords indicating loneliness from this text data."

[0499] 4. The server then performs further sentiment analysis to assess the overall emotional tone (negative, positive, neutral) of the text data using the Sentiment Analysis API.

[0500] Feedback and Advice

[0501] 5. The server generates a feedback report based on the analysis results. The report includes the frequency of keywords and emotional tone, and is sent to medical professionals and caregivers. For example, a report might say, "User A has been using the word lonely a lot recently. The emotional tone is negative."

[0502] Promoting dialogue

[0503] 6. The device will display or notify the elderly with messages encouraging regular conversations. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" will be sent.

[0504] Social activity proposals

[0505] 7. The server will suggest suitable social activities for the elderly. For example, a notification will be sent to the elderly saying, "There will be an art class at the local community center this Saturday. Would you like to join?"

[0506] Evaluation and Improvement

[0507] 8. The server reanalyzes the elderly person's conversation data after the suggestion and evaluates whether the sense of loneliness has been reduced, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[0508] 9. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system, which involves retraining the machine learning algorithm and introducing new datasets.

[0509] Specific examples

[0510] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server and converted into text using the Google Cloud Speech-to-Text API. A generative AI model then analyzes the text data, finding that the keyword "lonely" appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A with a message asking, "Have you been able to talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases and his quality of life improves.

[0511] In this way, the present invention is a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0512] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0513] Step 1:

[0514] The device records voice data. As input, it acquires the elderly person's (user's) everyday conversation. As operation, when the user speaks to the smart speaker or tablet, the device records the voice in digital format. As output, it generates the recorded voice data.

[0515] Step 2:

[0516] The terminal encrypts the recorded voice data and sends it to the server. As input, there is the recorded voice data. As operation, the terminal encrypts the voice data using AES technology to ensure the security of the data. Then, it sends the encrypted data to the server via HTTPS protocol. As output, the encrypted voice data is sent to the server.

[0517] Step 3:

[0518] The server converts the received voice data into text. As input, it receives encrypted voice data. In operation, the server uses speech recognition technology to convert the data into text format. Specifically, it uses the Google Cloud Speech-to-Text API. As output, it generates text data.

[0519] Step 4:

[0520] The server uses a generative AI model to extract keywords indicating loneliness and stress from text data. The input is text data. The operation is to apply a machine learning algorithm to detect specific keywords. An example of a prompt is "Please extract keywords indicating loneliness from this text data." The output is a list of extracted keywords.

[0521] Step 5:

[0522] The server performs sentiment analysis and evaluates the emotional tone of the entire text data. The input is text data. The operation is to evaluate the emotional tone of the data using the Sentiment Analysis API. The output is an evaluation result of the emotional tone (negative, positive, neutral).

[0523] Step 6:

[0524] The server generates a feedback report based on the analysis results. The inputs are a list of keywords and the emotional tone evaluation results. In operation, the server integrates these data to create a feedback report. As an output, a feedback report is generated. This report includes the frequency of keywords and the emotional tone.

[0525] Step 7:

[0526] The server sends the generated feedback report to the healthcare professional or caregiver. As input, there is a feedback report. As action, the server sends the report using secure email or a dedicated dashboard. As output, the feedback report has been sent.

[0527] Step 8:

[0528] The device displays or notifies the elderly with messages encouraging regular conversation. The input is an instruction from the system. The action is to display a notification on the screen or play a message by voice. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" is sent. The output is to encourage the elderly to have a conversation.

[0529] Step 9:

[0530] The server suggests suitable social activities for the elderly. As input, it has the feedback report and the profile data of the elderly. As operation, the system selects suitable activities for the elderly and generates a notification. For example, a notification may be generated such as "There is a painting class at your local community center this Saturday. Would you like to join?" As output, the suggested social activities are sent to the elderly.

[0531] Step 10:

[0532] The server re-analyzes the elderly person's conversation data after the suggestions and evaluates whether their feelings of loneliness have been reduced. The input is the new speech data after the suggestions. The operation is to re-apply the method from steps 3 to 5 described above to obtain new analysis results. The output is an evaluation result of whether the signs of loneliness have been reduced.

[0533] Step 11:

[0534] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy. The input is the reanalysis results. The operation involves retraining the machine learning algorithm and introducing a new dataset. The output is a new system with improved algorithm accuracy.

[0535] (Application example 1)

[0536] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0537] Situations in which elderly people are prone to feeling loneliness and stress in their daily lives can have a negative impact on their physical and mental health. Furthermore, if efforts to reduce loneliness are insufficient, there is a higher risk that an emergency at home will go unnoticed. The present invention aims to collect and analyze the everyday conversations of elderly people, detect signs of loneliness early, provide appropriate countermeasures, and reduce the risks associated with loneliness.

[0538] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0539] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and extracting keywords indicating loneliness or stress using a generative AI model, and a means for promptly notifying family members or caregivers if an abnormality is detected. This makes it possible to detect loneliness or stress in the elderly person's everyday conversations early on and take necessary measures promptly.

[0540] "Voice data" refers to data in which voice is electronically recorded, and includes everyday conversations between elderly people.

[0541] "Recording means" refers to devices or technologies for storing audio data, including, for example, smartphones, tablets, and smart speakers.

[0542] A "server" is a computer system that receives, processes, and analyzes voice data.

[0543] "Means for converting to text" refers to technology for converting voice data into text data in sentence format, and includes voice recognition software and APIs.

[0544] "Keyword extraction means" refers to technology for identifying and extracting specific words and phrases related to loneliness and stress from text data.

[0545] A "generative AI model" is an artificial intelligence algorithm that analyzes the meaning and sentiment of text data based on large datasets.

[0546] The "means for generating feedback" is a technology that generates information to suggest measures to medical professionals and caregivers based on the results of text analysis.

[0547] "Means of notification" refers to technologies for quickly transmitting analysis results and feedback information to family members and caregivers, and includes email and messaging services.

[0548] "Means for encouraging regular dialogue" refers to technology that displays or notifies elderly people of messages encouraging dialogue on a regular basis.

[0549] "Means for suggesting social activities" refers to techniques for suggesting events and club activities suitable for the elderly.

[0550] "Means for improving the accuracy of the system" refers to methods or techniques for updating the system's algorithms or databases based on the analysis results, thereby improving the accuracy and efficiency of the analysis.

[0551] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0552] Data collection

[0553] The device records audio data. For example, an elderly user might use a smartphone to record everyday conversations. The device periodically encrypts the recorded audio data and transmits it to a server in a privacy-protected manner. The hardware used is a standard Android or iOS device, and encryption is performed using the Python "pycryptodome" library.

[0554] Conversation Analysis

[0555] The server processes the received voice data and first converts the speech to text. This conversion is performed using Google's Cloud Speech-to-Text API. Next, the server uses a generative AI model (e.g., OpenAI GPT-4) to extract keywords indicating loneliness or stress from the text data. For example, if keywords such as "lonely" and "solitude" appear frequently, it is determined that the user is feeling lonely. The server then performs sentiment analysis to evaluate the emotional tone (negative, positive, or neutral) of the entire text data.

[0556] Feedback and Advice

[0557] The server generates a feedback report based on the analysis results. This report includes keyword frequency, emotional tone, and the results of emotion analysis using a generative AI model, and is sent to family members or caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently. You should talk to him more often" is provided. If an abnormality is detected, the family or caregiver is promptly notified.

[0558] Promoting dialogue

[0559] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0560] Social activity proposals

[0561] The server suggests social activities suitable for seniors. For example, notifications are sent to seniors recommending them to join hobby clubs or online social gatherings. Specifically, a notification might say, "There's a painting class at the local community center this Saturday. Would you like to join?"

[0562] Evaluation and Improvement

[0563] The server reanalyzes the elderly person's conversation data after the proposal and evaluates whether their feelings of loneliness have been alleviated. For example, it checks whether the frequency of keywords indicating loneliness has decreased. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy.

[0564] Specific examples

[0565] For example, user A might say on his smartphone, "I've been alone a lot lately and I feel lonely." This voice data is sent to the server, converted into text, and analyzed. The keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to the family, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in conversation by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and his quality of life improves.

[0566] Prompt Sentence Examples

[0567] Audio data: "I've been spending a lot of time alone lately and I feel lonely."

[0568] Text data: "I've been spending a lot of time alone lately and I feel lonely."

[0569] Keywords: "Spending a lot of time alone", "lonely"

[0570] Feedback: "User B feels lonely. Please consider visiting or contacting them regularly."

[0571] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0572] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0573] Step 1:

[0574] A user uses a smartphone to have everyday conversations.

[0575] (Input) User's everyday conversational voice.

[0576] (Output) Audio data recorded on a smartphone.

[0577] (Specific operation) When a user says "I'm lonely" to their smartphone, the voice is recorded by the smartphone's built-in microphone.

[0578] Step 2:

[0579] The device periodically encrypts the recorded audio data and sends it to the server.

[0580] (Input) Recorded audio data.

[0581] (Output) Encrypted audio data.

[0582] (Specific operation) The smartphone encrypts the voice data using AES (Advanced Encryption Standard) and sends it to the server via HTTPS.

[0583] Step 3:

[0584] The server converts the received voice data into text.

[0585] (Input) Encrypted audio data.

[0586] (Output) Text data.

[0587] (Specific operation) The server decrypts the encrypted voice data and converts the voice data into text data using Google's Cloud Speech-to-Text API.

[0588] Step 4:

[0589] The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data.

[0590] (Input) Text data.

[0591] (Output) Extracted keywords.

[0592] (Specific operation) The server inputs the text data into a natural language processing algorithm (e.g., OpenAI GPT-4) and extracts keywords such as "lonely" and "solitude."

[0593] Step 5:

[0594] The server also evaluates the emotional tone of the entire text data.

[0595] (Input) Text data and extracted keywords.

[0596] (Output) Sentiment analysis result (negative, positive, neutral).

[0597] (Specific operation) Using a generative AI model, the emotional tone of the entire text data is analyzed to determine emotional categories such as "negative" or "positive."

[0598] Step 6:

[0599] The server generates a feedback report based on the analysis results.

[0600] (Input) Extracted keywords and sentiment analysis results.

[0601] (Output) Feedback report.

[0602] (Specific operation) Based on the extracted keywords and the results of sentiment analysis, the server generates a feedback report such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[0603] Step 7:

[0604] If an abnormality is detected, the server will promptly notify family members or caregivers.

[0605] (Input) Feedback report.

[0606] (Output) A warning notification.

[0607] (Specific operation) The server uses the contact information of family members or caregivers to send messages via email or SMS such as "User A is feeling lonely, so we recommend regular phone consultations."

[0608] Step 8:

[0609] The device displays or notifies the elderly person of messages encouraging regular dialogue.

[0610] (Input) Instructions from the server.

[0611] (Output) Messages that are displayed on the terminal.

[0612] (Specific action) The smartphone displays a message such as, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0613] Step 9:

[0614] The server suggests social activities suitable for seniors.

[0615] (Input) Analysis results and social activity information in the database.

[0616] (Output) The proposal message.

[0617] (Specific operation) The server generates a message such as "There's a painting class at the local community center this Saturday. Would you like to join?" and sends it to the terminal.

[0618] Step 10:

[0619] The server reanalyzes the elderly's conversation data after the proposal to improve the accuracy of the system.

[0620] (Input) Re-collected audio data.

[0621] (Output) Updated algorithm to improve system accuracy.

[0622] (Specific operation) The server reanalyzes the elderly's conversation data to see if the frequency of keywords indicating loneliness has decreased, and uses the results to update the algorithm.

[0623] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0624] This invention relates to a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes their emotions. The system aims to reduce the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[0625] The main components of this system are: a means for recording voice data, a means for sending voice data to a server, a means for converting voice data into text, a means for analyzing text data and extracting keywords, an emotion engine, a means for generating feedback and sending it to medical professionals and caregivers, a means for encouraging regular dialogue with the elderly, a means for suggesting social activities, and a means for assessing the condition of the elderly and improving the accuracy of the system.

[0626] An embodiment of this system is described in detail below.

[0627] Data collection

[0628] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[0629] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[0630] Conversation Analysis

[0631] The server processes the received voice data and first converts the voice data into text, then uses voice recognition technology to convert the voice data into text.

[0632] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0633] emotion recognition

[0634] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[0635] Feedback and Advice

[0636] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[0637] For example, specific advice may be provided such as, "User A has been using the word lonely a lot recently, and the emotion engine's evaluation has confirmed that he has strong negative emotions. We recommend that you increase the frequency of conversations with him and suggest some hobby activities."

[0638] Promoting dialogue

[0639] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0640] Social activity proposals

[0641] The server suggests social activities suitable for seniors, using generative AI to select activities based on the user's interests and sending that information to the device.

[0642] Specifically, the device will send a notification to the elderly such as, "There will be a painting class at the local community center this Saturday. Would you like to join?"

[0643] Evaluation and Improvement

[0644] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been reduced. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0645] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[0646] Specific examples

[0647] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text. The server performs analysis and detects negative emotions through the keyword "lonely" and evaluation by the emotion engine. The server then generates a feedback report and notifies medical professionals that "User B is likely feeling very lonely." The device then sends User B messages such as "Have you talked to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[0648] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[0649] The processing flow will be explained below.

[0650] Step 1:

[0651] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[0652] Step 2:

[0653] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[0654] Step 3:

[0655] The server decrypts the received encrypted voice data and converts the voice data into text data. It uses voice recognition technology to carry out the process of converting the voice data into text information.

[0656] Step 4:

[0657] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0658] Step 5:

[0659] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[0660] Step 6:

[0661] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[0662] Step 7:

[0663] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[0664] Step 8:

[0665] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0666] Step 9:

[0667] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[0668] Step 10:

[0669] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0670] Step 11:

[0671] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0672] Step 12:

[0673] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[0674] Specific examples

[0675] For example, if user C says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text data.

[0676] The server then performs an analysis and detects that the keyword "lonely" appears frequently, and the emotion engine identifies negative emotions.

[0677] The server then generates a feedback report and notifies the medical professional of the assessment result that "User C is likely to be feeling a strong sense of loneliness."

[0678] The terminal sends a message to user C prompting a conversation, such as "Have you talked to anyone today?"

[0679] Additionally, the server suggests social activities to User C, such as "Would you like to join a local online book club?"

[0680] This reduces User C's sense of loneliness and improves his quality of life.

[0681] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[0682] Example 2

[0683] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0684] There is a lack of systems that can detect elderly people's feelings of loneliness and emotional states early and take appropriate countermeasures. Conventional systems have difficulty analyzing elderly people's emotional states from their everyday conversations and providing feedback. Furthermore, there are limited means to effectively communicate this data to medical professionals and caregivers. The present invention aims to solve these problems, reduce elderly people's feelings of loneliness, and improve their quality of life.

[0685] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0686] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and voice data and evaluating the emotional state, and a means for generating a feedback report based on the analysis results and sending it to a medical professional or caregiver, thereby enabling the emotional state of an elderly person to be analyzed quickly and accurately from their everyday conversations and appropriate countermeasures to be implemented.

[0687] "Audio data" refers to digital audio files that record everyday conversations between elderly people.

[0688] A "server" is an information processing device that processes and analyzes voice data and generates feedback.

[0689] "Text data" refers to character information converted from voice data using voice recognition technology.

[0690] "Keywords" are important words in the text data that indicate feelings of loneliness or stress.

[0691] An "emotion engine" is software that analyzes voice and text data to assess emotional states.

[0692] "Analysis results" are information obtained from analyzing text data and evaluating it using an emotion engine.

[0693] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[0694] "Medical professionals" are professionals such as doctors and nurses who receive the feedback reports and use them to help care for older people.

[0695] A "caregiver" is a person who receives the feedback report and provides support for the elderly person's daily life.

[0696] "Promoting regular dialogue" means displaying or notifying elderly people of messages encouraging dialogue on a regular basis.

[0697] "Social activity suggestions" refers to providing information on social activities that encourage participation based on the interests of the elderly.

[0698] "Evaluating the state" involves reanalyzing the elderly person's conversation data after the proposal to confirm any changes in their feelings of loneliness or emotional state.

[0699] "Improving the accuracy of the system" means updating the analysis algorithm, introducing new keywords, and improving emotion recognition technology to improve the system's performance.

[0700] This invention is a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes emotions. The main purpose of this system is to alleviate the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[0701] Data collection

[0702] Users use devices such as smart speakers and tablets to record voice data. For example, a user may record their everyday conversations using a smart speaker (a common name for such devices). The device periodically encrypts the recorded voice data and transmits it to a server in a privacy-protected state.

[0703] Conversation Analysis

[0704] The server processes the received voice data and first converts it into text using speech recognition technology (e.g., a speech recognition API). The server then analyzes the text using a generative AI model (e.g., a natural language processing tool). The analysis includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0705] emotion recognition

[0706] The server's emotion engine analyzes both voice and text data to assess the elderly person's emotional state. An emotion recognition API (generic name) can be used to infer emotions from tone of voice, speaking style, and text content.

[0707] Feedback and Advice

[0708] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[0709] Promoting dialogue

[0710] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0711] Social activity proposals

[0712] The server suggests social activities suitable for seniors. Using a generative AI model, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior, such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0713] Evaluation and Improvement

[0714] The server re-analyzes the elderly person's conversation data after the proposal and evaluates whether their sense of loneliness has been reduced. It then performs another speech-to-text conversion and keyword extraction to confirm any changes in their condition. It also aims to improve the system's accuracy based on the evaluation results, by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[0715] Specific examples

[0716] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text using a speech recognition API. The server then analyzes the text using a generative AI model and detects negative emotions based on the keyword "lonely" and the evaluation of an emotion engine. The server then generates a feedback report and notifies medical professionals with specific content, such as "User B is likely feeling very lonely." The device then sends User B a message such as "Did you talk to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[0717] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[0718] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0719] Step 1: Recording audio data

[0720] The device records the user's everyday conversations. For example, when a user says "Good morning" using a smart speaker or tablet, the conversation is recorded as audio data. The input is the user's voice, and the output is digital audio data.

[0721] Step 2: Encrypt and transmit the audio data

[0722] The device encrypts the recorded voice data and sends it to the server. For example, voice data recorded at a fixed time every day is encrypted using the AES (Advanced Encryption Standard) algorithm. The encrypted data is then uploaded to the server. The input is voice data, and the output is encrypted voice data.

[0723] Step 3: Convert audio data to text

[0724] The server decrypts the received encrypted voice data and converts it into text data using a speech recognition API (for example, Google Speech-to-Text API). The input is encrypted voice data, and the output is text data. For example, voice data saying "Good morning" is converted into the string "Good morning."

[0725] Step 4: Analyzing the text data

[0726] The server analyzes the text data using a generative AI model (e.g., a natural language processing tool). The analysis involves extracting keywords that indicate loneliness or stress. The input is the text data, and the output is the analysis results. For example, keywords such as "lonely" and "alone" are extracted, and their frequency is calculated.

[0727] Step 5: Emotion Recognition

[0728] The server's emotion engine analyzes both voice and text data to evaluate the user's emotional state. It uses emotion recognition APIs (such as IBM Watson or Microsoft Azure Emotion API). The input is text and voice data, and the output is an emotional evaluation result. For example, emotions such as "sadness" or "anxiety" are detected.

[0729] Step 6: Generate a feedback report

[0730] The server generates a feedback report based on the analysis results. The report includes the frequency of extracted keywords, emotional tone, and the emotion engine's evaluation results. The input is the analysis results and the emotion evaluation results, and the output is a feedback report. For example, it may contain specific content such as, "User A has been using the word 'lonely' a lot recently, and the emotion evaluation has confirmed a negative state."

[0731] Step 7: Submit your feedback

[0732] The server generates and sends the feedback report to the healthcare professional or caregiver. For example, the report is sent via email or a web application using the notification function of the internal system. The input is the feedback report, and the output is the report being sent to the healthcare professional or caregiver.

[0733] Step 8: Promote regular dialogue

[0734] The terminal displays or notifies the user of messages that prompt regular interactions. For example, a voice or text message such as "Tell me how you're feeling today" is emitted from the terminal at a fixed time every day. The input is an instruction from the server, and the output is a message that is displayed to the user.

[0735] Step 9: Propose a social activity

[0736] The server selects social activities that match the user's interests and sends that information to the device. For example, a notification may be sent saying, "There's an online book club this Saturday. Would you like to join?" The input is the user's interest data and the analysis results of the generative AI, and the output is the suggested content.

[0737] Step 10: Assess the condition and improve the system

[0738] The server reanalyzes the user's conversation data after the proposal and evaluates whether the sense of loneliness has been reduced. Based on the evaluation results, the system's performance can be improved by updating the analysis algorithm or introducing new keywords. The input is the reanalyzed data, and the output is the system improvements. For example, "Since a reduction in the sense of loneliness was observed, we will introduce a new emotion recognition algorithm."

[0739] (Application example 2)

[0740] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0741] The challenge is to provide an effective dialogue system that can quickly detect and alleviate the feelings of loneliness and negative emotions that elderly people may experience in brick-and-mortar stores. It is also necessary to support elderly people in communicating more smoothly with store staff and other customers.

[0742] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for recording voice data, means for transmitting the recorded voice data to the server, means for converting the voice data to text in the server, means for analyzing the text data and extracting keywords indicating loneliness or stress, means for generating feedback based on the analysis results and transmitting the feedback to a medical professional or caregiver, means for displaying or notifying the elderly person of a message encouraging regular conversation, means for suggesting social activities suitable for the elderly, means for evaluating the elderly person's condition after the suggestions and improving the system's accuracy, means for suggesting conversation topics based on the analysis results, and means for providing store staff with information useful for conversations with the elderly. This makes it possible to detect elderly people's loneliness or negative emotions early on and to suggest effective conversations and social activities.

[0743] "Means for recording voice data" refers to a device or program for collecting and recording voices uttered by the elderly person in digital form.

[0744] The "means for transmitting recorded voice data to a server" refers to a device or program for securely transferring recorded voice data to a server via a network.

[0745] The "means for converting voice data into text on the server" is a program that converts voice data into text format using natural language processing technology.

[0746] The "means for analyzing text data and extracting keywords that indicate loneliness and stress" is a program that analyzes text data using machine learning models and natural language processing technology to identify keywords related to loneliness and stress.

[0747] The "means for generating feedback based on the analysis results and sending it to a medical professional or caregiver" is a program that creates a feedback report based on the analysis results and sends it to a medical professional or caregiver.

[0748] "Means for displaying or notifying messages encouraging elderly people to have regular conversations" refers to a device or program that displays or notifies elderly people in voice or text format with messages to encourage them to have regular conversations.

[0749] The "means for suggesting social activities suitable for the elderly" is a program that suggests appropriate social activities based on the interests and emotional state of each elderly person.

[0750] The "means of evaluating the elderly person's condition after the proposal and improving the accuracy of the system" is a program for reanalyzing the elderly person's reaction and condition after the proposal and improving the accuracy of the system's analysis and the effectiveness of the proposal.

[0751] The "means for proposing a topic for conversation based on the analysis result" is a program that provides an appropriate topic for conversation in accordance with the analyzed emotional state.

[0752] "Means for providing store staff with information that will be useful when interacting with the elderly" refers to a program that provides store staff with the information and suggestions they need to smoothly interact with the elderly.

[0753] The present invention relates to a system for elderly people to use in brick-and-mortar stores. This system is designed to detect loneliness and negative emotions in elderly people at an early stage and promote dialogue. Specific embodiments of the system are described below.

[0754] Data collection

[0755] The device records voice data. When an elderly person (user) speaks using smart glasses or a smartphone in a physical store, the voice data is recorded. For example, in a cafe or shopping store, the user may say, "I've been feeling lonely lately because I haven't been able to see my friends."

[0756] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner using a secure communication protocol (e.g., HTTPS).

[0757] Conversation Analysis

[0758] The server processes the received voice data and first converts it into text using software with speech recognition technology (for example, Python's speech_recognition library).

[0759] The server analyzes the text data using a generative AI model. This analysis involves extracting keywords that indicate loneliness or stress and measuring the frequency of these keywords. For example, if keywords such as "lonely" or "alone" appear frequently, it determines that particular attention is needed.

[0760] Emotion Recognition and Feedback

[0761] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, using technology that infers emotions from tone of voice, speaking style, and text content.

[0762] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[0763] Promoting dialogue and proposing social activities

[0764] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0765] The server suggests social activities suitable for seniors. Using generative AI, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0766] Dialogue suggestions

[0767] The server then suggests conversation topics based on the analysis results. This function allows users to have smoother conversations with store staff and other customers. For example, if the emotion analysis results indicate "loneliness," the server might suggest a conversation topic such as, "How about talking about the recent weather or your hobbies?"

[0768] Examples and prompts

[0769] For example, if an elderly person says, "I've been feeling lonely lately because I haven't been able to see my friends," at a brick-and-mortar cafe, this voice data is sent to the server and converted into text. The analysis results indicate the emotion of "loneliness," and the app suggests topics for conversation, such as, "How about talking about the weather recently or your hobbies?"

[0770] Specific prompt examples:

[0771] User text: "I've been missing my friends lately."

[0772] Emotion analysis result: {'emotion': 'sadness'}

[0773] Suggested topic: "Why don't we talk about the weather these days or our hobbies?"

[0774] This will help reduce the sense of loneliness among the elderly and improve their quality of life, while providing store staff with specific support to facilitate smoother conversations with the elderly.

[0775] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0776] Step 1:

[0777] The device records the voice data spoken by elderly people in a physical store. For example, if a user says, "I miss my friends these days," in a cafe, the device digitally records the voice. The input is the user's voice data, and the output is the recorded voice file.

[0778] Step 2:

[0779] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner. A secure communication protocol (HTTPS) is used for data transmission. The input is the recorded audio file, and the output is the transmission of the encrypted audio file.

[0780] Step 3:

[0781] The server converts the received voice data into text using speech recognition technology. Here, we use the Python speech_recognition library. The input is an encrypted audio file, and the output is text data.

[0782] Step 4:

[0783] The server uses a generative AI model to analyze the text data and extract keywords that indicate loneliness and stress. The input is the text data converted from speech, and the output is the extracted keywords and their frequency. For example, keywords such as "lonely" and "alone" are output.

[0784] Step 5:

[0785] The server's emotion engine further analyzes the voice and text data to assess the elderly person's emotional state. The emotion engine uses a generative AI model, whose input is the extracted keywords and their frequency, and whose output is the emotional state (e.g., "loneliness").

[0786] Step 6:

[0787] The server generates a feedback report based on the analysis results and sends it to medical professionals and caregivers. The input is the evaluation result of the emotion engine, and the output is the feedback report. The report includes the frequency of keywords and emotional tone.

[0788] Step 7:

[0789] The device displays or notifies the elderly with messages encouraging regular conversation. The input is a feedback report, and the output is a notification message (e.g., "Talking to someone every day is good for your health. Please tell us how you feel today.").

[0790] Step 8:

[0791] The server suggests social activities suitable for seniors. It uses generative AI to select activities based on the user's interests and sends that information to the device. The input is the analysis results from the generative AI model, and the output is suggested social activity information. For example, a suggestion might be output such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0792] Step 9:

[0793] The server then suggests conversation topics based on the analysis results. The input is emotional state and keywords, and the output is the conversation topic. For example, a suggestion might be, "Why don't we talk about the recent weather and your hobbies?"

[0794] Step 10:

[0795] The server provides store staff with information that is useful when interacting with elderly people. The input is the emotion analysis results and conversation topics, and the output is conversation support information. This allows staff to communicate smoothly with elderly people.

[0796] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0797] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0798] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0799] [Third embodiment]

[0800] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0801] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0802] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0803] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0804] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0805] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0806] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0807] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0808] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0809] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0810] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0811] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0812] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0813] The main components of this system are: "means for recording voice data," "means for sending voice data to a server," "means for converting voice data into text," "means for analyzing text data and extracting keywords," "means for generating feedback and sending it to medical professionals and caregivers," "means for encouraging regular dialogue with the elderly," "means for suggesting social activities," and "means for assessing the condition of the elderly and improving the accuracy of the system."

[0814] An embodiment of this system is described in detail below.

[0815] Data collection

[0816] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[0817] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[0818] Conversation Analysis

[0819] The server processes the received voice data and first converts the voice into text using voice recognition technology.

[0820] The server uses a generative AI to extract keywords that indicate loneliness or stress from the text data. For example, if keywords such as "lonely" or "solitude" appear frequently, it is determined that the user is feeling lonely.

[0821] The server also performs sentiment analysis, assessing the overall emotional tone (negative, positive, neutral) of the text data.

[0822] Feedback and Advice

[0823] The server generates a feedback report based on the analysis, including keyword frequency and emotional tone, which is sent to medical professionals and caregivers.

[0824] For example, specific advice is provided such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[0825] Promoting dialogue

[0826] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0827] Social activity proposals

[0828] The server will suggest suitable social activities for seniors, such as sending notifications recommending them to join hobby clubs or online meetups.

[0829] Specifically, you might receive a notification such as, "There's a painting class at your local community center this Saturday. Would you like to join?"

[0830] Evaluation and Improvement

[0831] The server then reanalyzes the elderly person's conversation data after the suggestions are made and evaluates whether their sense of loneliness has been alleviated, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[0832] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system.

[0833] Specific examples

[0834] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server, converted into text, and analyzed. The result is that the keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in dialogue by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and their quality of life improves.

[0835] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0836] The processing flow will be explained below.

[0837] Step 1:

[0838] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[0839] Step 2:

[0840] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[0841] Step 3:

[0842] The server decrypts the received encrypted voice data and converts the voice into text information. It uses voice recognition technology to carry out the process of converting the voice data into text data.

[0843] Step 4:

[0844] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[0845] Step 5:

[0846] The server evaluates the emotional tone of the entire text data, analyzing it as negative, positive, or neutral to understand the user's emotional state.

[0847] Step 6:

[0848] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords and an evaluation of emotional tone, and is sent to medical professionals and caregivers.

[0849] Step 7:

[0850] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[0851] Step 8:

[0852] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0853] Step 9:

[0854] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[0855] Step 10:

[0856] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[0857] Step 11:

[0858] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[0859] Step 12:

[0860] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate voice recognition technology.

[0861] Example 1

[0862] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0863] Elderly people are prone to feelings of loneliness and stress, which often have a negative impact on their health. Conventional measures make it difficult to detect early signs of loneliness and stress and take appropriate measures. In addition, feedback and responses are often inappropriate, leading to the problem of not being able to adequately support the physical and mental health of elderly people.

[0864] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0865] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data using a generative AI model and extracting keywords indicative of loneliness or stress, a means for generating a feedback report based on the extracted keywords and emotional tone and sending it to a medical professional or caregiver, a means for displaying or notifying a message encouraging the elderly to have regular conversations, a means for suggesting social activities suitable for the elderly, and a means for reanalyzing the elderly's conversation data after the suggestions and evaluating it to improve the accuracy of the system. This makes it possible to detect signs of loneliness or stress early through the elderly's everyday conversations and take appropriate countermeasures.

[0866] "Voice data" is data in digital form obtained from a user's voice.

[0867] A "terminal" is an electronic device for recording and transmitting voice data, including smart speakers and tablets.

[0868] A "server" is a computer system for receiving and processing audio data.

[0869] The "means for converting to text" is a function for converting voice data into text information, and voice recognition technology is used for this.

[0870] A "generative AI model" is an artificial intelligence model that uses machine learning and natural language processing techniques to analyze data and extract specific information.

[0871] "Keywords" are specific words or phrases that indicate loneliness or stress.

[0872] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[0873] "Messages encouraging regular dialogue" are notifications and messages that are displayed to encourage everyday dialogue with elderly people.

[0874] "Means for suggesting social activities" is a suggestion function for providing elderly people with opportunities for community activities and interaction.

[0875] The "means of evaluation" is an evaluation function for reanalyzing the condition of the elderly person and improving the accuracy of the system.

[0876] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0877] Data collection

[0878] 1. The device records voice data. For example, an elderly person (user) uses a smart speaker or tablet to record their everyday conversations. The recorded voice data is periodically encrypted by the device and sent to a server in a privacy-protected manner. AES (Advanced Encryption Standard) technology is used for encryption, and the data is sent using the HTTPS protocol.

[0879] Conversation Analysis

[0880] 2. The server processes the received voice data and first converts it into text using the Google Cloud Speech-to-Text API. For example, the resulting text is "I was gardening today."

[0881] 3. The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data. Specifically, it uses a machine learning algorithm to extract keywords such as "lonely," "solitude," and "alone." An example of a prompt is, "Please extract keywords indicating loneliness from this text data."

[0882] 4. The server then performs further sentiment analysis to assess the overall emotional tone (negative, positive, neutral) of the text data using the Sentiment Analysis API.

[0883] Feedback and Advice

[0884] 5. The server generates a feedback report based on the analysis results. The report includes the frequency of keywords and emotional tone, and is sent to medical professionals and caregivers. For example, a report might say, "User A has been using the word lonely a lot recently. The emotional tone is negative."

[0885] Promoting dialogue

[0886] 6. The device will display or notify the elderly with messages encouraging regular conversations. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" will be sent.

[0887] Social activity proposals

[0888] 7. The server will suggest suitable social activities for the elderly. For example, a notification will be sent to the elderly saying, "There will be an art class at the local community center this Saturday. Would you like to join?"

[0889] Evaluation and Improvement

[0890] 8. The server reanalyzes the elderly person's conversation data after the suggestion and evaluates whether the sense of loneliness has been reduced, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[0891] 9. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system, which involves retraining the machine learning algorithm and introducing new datasets.

[0892] Specific examples

[0893] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server and converted into text using the Google Cloud Speech-to-Text API. A generative AI model then analyzes the text data, finding that the keyword "lonely" appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A with a message asking, "Have you been able to talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases and his quality of life improves.

[0894] In this way, the present invention is a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0895] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0896] Step 1:

[0897] The device records voice data. As input, it acquires the elderly person's (user's) everyday conversation. As operation, when the user speaks to the smart speaker or tablet, the device records the voice in digital format. As output, it generates the recorded voice data.

[0898] Step 2:

[0899] The terminal encrypts the recorded voice data and sends it to the server. As input, there is the recorded voice data. As operation, the terminal encrypts the voice data using AES technology to ensure the security of the data. Then, it sends the encrypted data to the server via HTTPS protocol. As output, the encrypted voice data is sent to the server.

[0900] Step 3:

[0901] The server converts the received voice data into text. As input, it receives encrypted voice data. In operation, the server uses speech recognition technology to convert the data into text format. Specifically, it uses the Google Cloud Speech-to-Text API. As output, it generates text data.

[0902] Step 4:

[0903] The server uses a generative AI model to extract keywords indicating loneliness and stress from text data. The input is text data. The operation is to apply a machine learning algorithm to detect specific keywords. An example of a prompt is "Please extract keywords indicating loneliness from this text data." The output is a list of extracted keywords.

[0904] Step 5:

[0905] The server performs sentiment analysis and evaluates the emotional tone of the entire text data. The input is text data. The operation is to evaluate the emotional tone of the data using the Sentiment Analysis API. The output is an evaluation result of the emotional tone (negative, positive, neutral).

[0906] Step 6:

[0907] The server generates a feedback report based on the analysis results. The inputs are a list of keywords and the emotional tone evaluation results. In operation, the server integrates these data to create a feedback report. As an output, a feedback report is generated. This report includes the frequency of keywords and the emotional tone.

[0908] Step 7:

[0909] The server sends the generated feedback report to the healthcare professional or caregiver. As input, there is a feedback report. As action, the server sends the report using secure email or a dedicated dashboard. As output, the feedback report has been sent.

[0910] Step 8:

[0911] The device displays or notifies the elderly with messages encouraging regular conversation. The input is an instruction from the system. The action is to display a notification on the screen or play a message by voice. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" is sent. The output is to encourage the elderly to have a conversation.

[0912] Step 9:

[0913] The server suggests suitable social activities for the elderly. As input, it has the feedback report and the profile data of the elderly. As operation, the system selects suitable activities for the elderly and generates a notification. For example, a notification may be generated such as "There is a painting class at your local community center this Saturday. Would you like to join?" As output, the suggested social activities are sent to the elderly.

[0914] Step 10:

[0915] The server re-analyzes the elderly person's conversation data after the suggestions and evaluates whether their feelings of loneliness have been reduced. The input is the new speech data after the suggestions. The operation is to re-apply the method from steps 3 to 5 described above to obtain new analysis results. The output is an evaluation result of whether the signs of loneliness have been reduced.

[0916] Step 11:

[0917] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy. The input is the reanalysis results. The operation involves retraining the machine learning algorithm and introducing a new dataset. The output is a new system with improved algorithm accuracy.

[0918] (Application example 1)

[0919] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0920] Situations in which elderly people are prone to feeling loneliness and stress in their daily lives can have a negative impact on their physical and mental health. Furthermore, if efforts to reduce loneliness are insufficient, there is a higher risk that an emergency at home will go unnoticed. The present invention aims to collect and analyze the everyday conversations of elderly people, detect signs of loneliness early, provide appropriate countermeasures, and reduce the risks associated with loneliness.

[0921] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0922] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and extracting keywords indicating loneliness or stress using a generative AI model, and a means for promptly notifying family members or caregivers if an abnormality is detected. This makes it possible to detect loneliness or stress in the elderly person's everyday conversations early on and take necessary measures promptly.

[0923] "Voice data" refers to data in which voice is electronically recorded, and includes everyday conversations between elderly people.

[0924] "Recording means" refers to devices or technologies for storing audio data, including, for example, smartphones, tablets, and smart speakers.

[0925] A "server" is a computer system that receives, processes, and analyzes voice data.

[0926] "Means for converting to text" refers to technology for converting voice data into text data in sentence format, and includes voice recognition software and APIs.

[0927] "Keyword extraction means" refers to technology for identifying and extracting specific words and phrases related to loneliness and stress from text data.

[0928] A "generative AI model" is an artificial intelligence algorithm that analyzes the meaning and sentiment of text data based on large datasets.

[0929] The "means for generating feedback" is a technology that generates information to suggest measures to medical professionals and caregivers based on the results of text analysis.

[0930] "Means of notification" refers to technologies for quickly transmitting analysis results and feedback information to family members and caregivers, and includes email and messaging services.

[0931] "Means for encouraging regular dialogue" refers to technology that displays or notifies elderly people of messages encouraging dialogue on a regular basis.

[0932] "Means for suggesting social activities" refers to techniques for suggesting events and club activities suitable for the elderly.

[0933] "Means for improving the accuracy of the system" refers to methods or techniques for updating the system's algorithms or databases based on the analysis results, thereby improving the accuracy and efficiency of the analysis.

[0934] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[0935] Data collection

[0936] The device records audio data. For example, an elderly user might use a smartphone to record everyday conversations. The device periodically encrypts the recorded audio data and transmits it to a server in a privacy-protected manner. The hardware used is a standard Android or iOS device, and encryption is performed using the Python "pycryptodome" library.

[0937] Conversation Analysis

[0938] The server processes the received voice data and first converts the speech to text. This conversion is performed using Google's Cloud Speech-to-Text API. Next, the server uses a generative AI model (e.g., OpenAI GPT-4) to extract keywords indicating loneliness or stress from the text data. For example, if keywords such as "lonely" and "solitude" appear frequently, it is determined that the user is feeling lonely. The server then performs sentiment analysis to evaluate the emotional tone (negative, positive, or neutral) of the entire text data.

[0939] Feedback and Advice

[0940] The server generates a feedback report based on the analysis results. This report includes keyword frequency, emotional tone, and the results of emotion analysis using a generative AI model, and is sent to family members or caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently. You should talk to him more often" is provided. If an abnormality is detected, the family or caregiver is promptly notified.

[0941] Promoting dialogue

[0942] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0943] Social activity proposals

[0944] The server suggests social activities suitable for seniors. For example, notifications are sent to seniors recommending them to join hobby clubs or online social gatherings. Specifically, a notification might say, "There's a painting class at the local community center this Saturday. Would you like to join?"

[0945] Evaluation and Improvement

[0946] The server reanalyzes the elderly person's conversation data after the proposal and evaluates whether their feelings of loneliness have been alleviated. For example, it checks whether the frequency of keywords indicating loneliness has decreased. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy.

[0947] Specific examples

[0948] For example, user A might say on his smartphone, "I've been alone a lot lately and I feel lonely." This voice data is sent to the server, converted into text, and analyzed. The keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to the family, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in conversation by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and his quality of life improves.

[0949] Prompt Sentence Examples

[0950] Audio data: "I've been spending a lot of time alone lately and I feel lonely."

[0951] Text data: "I've been spending a lot of time alone lately and I feel lonely."

[0952] Keywords: "Spending a lot of time alone", "lonely"

[0953] Feedback: "User B feels lonely. Please consider visiting or contacting them regularly."

[0954] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[0955] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0956] Step 1:

[0957] A user uses a smartphone to have everyday conversations.

[0958] (Input) User's everyday conversational voice.

[0959] (Output) Audio data recorded on a smartphone.

[0960] (Specific operation) When a user says "I'm lonely" to their smartphone, the voice is recorded by the smartphone's built-in microphone.

[0961] Step 2:

[0962] The device periodically encrypts the recorded audio data and sends it to the server.

[0963] (Input) Recorded audio data.

[0964] (Output) Encrypted audio data.

[0965] (Specific operation) The smartphone encrypts the voice data using AES (Advanced Encryption Standard) and sends it to the server via HTTPS.

[0966] Step 3:

[0967] The server converts the received voice data into text.

[0968] (Input) Encrypted audio data.

[0969] (Output) Text data.

[0970] (Specific operation) The server decrypts the encrypted voice data and converts the voice data into text data using Google's Cloud Speech-to-Text API.

[0971] Step 4:

[0972] The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data.

[0973] (Input) Text data.

[0974] (Output) Extracted keywords.

[0975] (Specific operation) The server inputs the text data into a natural language processing algorithm (e.g., OpenAI GPT-4) and extracts keywords such as "lonely" and "solitude."

[0976] Step 5:

[0977] The server also evaluates the emotional tone of the entire text data.

[0978] (Input) Text data and extracted keywords.

[0979] (Output) Sentiment analysis result (negative, positive, neutral).

[0980] (Specific operation) Using a generative AI model, the emotional tone of the entire text data is analyzed to determine emotional categories such as "negative" or "positive."

[0981] Step 6:

[0982] The server generates a feedback report based on the analysis results.

[0983] (Input) Extracted keywords and sentiment analysis results.

[0984] (Output) Feedback report.

[0985] (Specific operation) Based on the extracted keywords and the results of sentiment analysis, the server generates a feedback report such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[0986] Step 7:

[0987] If an abnormality is detected, the server will promptly notify family members or caregivers.

[0988] (Input) Feedback report.

[0989] (Output) A warning notification.

[0990] (Specific operation) The server uses the contact information of family members or caregivers to send messages via email or SMS such as "User A is feeling lonely, so we recommend regular phone consultations."

[0991] Step 8:

[0992] The device displays or notifies the elderly person of messages encouraging regular dialogue.

[0993] (Input) Instructions from the server.

[0994] (Output) Messages that are displayed on the terminal.

[0995] (Specific action) The smartphone displays a message such as, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[0996] Step 9:

[0997] The server suggests social activities suitable for seniors.

[0998] (Input) Analysis results and social activity information in the database.

[0999] (Output) The proposal message.

[1000] (Specific operation) The server generates a message such as "There's a painting class at the local community center this Saturday. Would you like to join?" and sends it to the terminal.

[1001] Step 10:

[1002] The server reanalyzes the elderly's conversation data after the proposal to improve the accuracy of the system.

[1003] (Input) Re-collected audio data.

[1004] (Output) Updated algorithm to improve system accuracy.

[1005] (Specific operation) The server reanalyzes the elderly's conversation data to see if the frequency of keywords indicating loneliness has decreased, and uses the results to update the algorithm.

[1006] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1007] This invention relates to a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes their emotions. The system aims to reduce the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[1008] The main components of this system are: a means for recording voice data, a means for sending voice data to a server, a means for converting voice data into text, a means for analyzing text data and extracting keywords, an emotion engine, a means for generating feedback and sending it to medical professionals and caregivers, a means for encouraging regular dialogue with the elderly, a means for suggesting social activities, and a means for assessing the condition of the elderly and improving the accuracy of the system.

[1009] An embodiment of this system is described in detail below.

[1010] Data collection

[1011] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[1012] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[1013] Conversation Analysis

[1014] The server processes the received voice data and first converts the voice data into text, then uses voice recognition technology to convert the voice data into text.

[1015] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1016] emotion recognition

[1017] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[1018] Feedback and Advice

[1019] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[1020] For example, specific advice may be provided such as, "User A has been using the word lonely a lot recently, and the emotion engine's evaluation has confirmed that he has strong negative emotions. We recommend that you increase the frequency of conversations with him and suggest some hobby activities."

[1021] Promoting dialogue

[1022] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1023] Social activity proposals

[1024] The server suggests social activities suitable for seniors, using generative AI to select activities based on the user's interests and sending that information to the device.

[1025] Specifically, the device will send a notification to the elderly such as, "There will be a painting class at the local community center this Saturday. Would you like to join?"

[1026] Evaluation and Improvement

[1027] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been reduced. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[1028] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[1029] Specific examples

[1030] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text. The server performs analysis and detects negative emotions through the keyword "lonely" and evaluation by the emotion engine. The server then generates a feedback report and notifies medical professionals that "User B is likely feeling very lonely." The device then sends User B messages such as "Have you talked to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[1031] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[1032] The processing flow will be explained below.

[1033] Step 1:

[1034] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[1035] Step 2:

[1036] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[1037] Step 3:

[1038] The server decrypts the received encrypted voice data and converts the voice data into text data. It uses voice recognition technology to carry out the process of converting the voice data into text information.

[1039] Step 4:

[1040] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1041] Step 5:

[1042] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[1043] Step 6:

[1044] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[1045] Step 7:

[1046] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[1047] Step 8:

[1048] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1049] Step 9:

[1050] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[1051] Step 10:

[1052] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1053] Step 11:

[1054] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[1055] Step 12:

[1056] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[1057] Specific examples

[1058] For example, if user C says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text data.

[1059] The server then performs an analysis and detects that the keyword "lonely" appears frequently, and the emotion engine identifies negative emotions.

[1060] The server then generates a feedback report and notifies the medical professional of the assessment result that "User C is likely to be feeling a strong sense of loneliness."

[1061] The terminal sends a message to user C prompting a conversation, such as "Have you talked to anyone today?"

[1062] Additionally, the server suggests social activities to User C, such as "Would you like to join a local online book club?"

[1063] This reduces User C's sense of loneliness and improves his quality of life.

[1064] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[1065] Example 2

[1066] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1067] There is a lack of systems that can detect elderly people's feelings of loneliness and emotional states early and take appropriate countermeasures. Conventional systems have difficulty analyzing elderly people's emotional states from their everyday conversations and providing feedback. Furthermore, there are limited means to effectively communicate this data to medical professionals and caregivers. The present invention aims to solve these problems, reduce elderly people's feelings of loneliness, and improve their quality of life.

[1068] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1069] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and voice data and evaluating the emotional state, and a means for generating a feedback report based on the analysis results and sending it to a medical professional or caregiver, thereby enabling the emotional state of an elderly person to be analyzed quickly and accurately from their everyday conversations and appropriate countermeasures to be implemented.

[1070] "Audio data" refers to digital audio files that record everyday conversations between elderly people.

[1071] A "server" is an information processing device that processes and analyzes voice data and generates feedback.

[1072] "Text data" refers to character information converted from voice data using voice recognition technology.

[1073] "Keywords" are important words in the text data that indicate feelings of loneliness or stress.

[1074] An "emotion engine" is software that analyzes voice and text data to assess emotional states.

[1075] "Analysis results" are information obtained from analyzing text data and evaluating it using an emotion engine.

[1076] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[1077] "Medical professionals" are professionals such as doctors and nurses who receive the feedback reports and use them to help care for older people.

[1078] A "caregiver" is a person who receives the feedback report and provides support for the elderly person's daily life.

[1079] "Promoting regular dialogue" means displaying or notifying elderly people of messages encouraging dialogue on a regular basis.

[1080] "Social activity suggestions" refers to providing information on social activities that encourage participation based on the interests of the elderly.

[1081] "Evaluating the state" involves reanalyzing the elderly person's conversation data after the proposal to confirm any changes in their feelings of loneliness or emotional state.

[1082] "Improving the accuracy of the system" means updating the analysis algorithm, introducing new keywords, and improving emotion recognition technology to improve the system's performance.

[1083] This invention is a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes emotions. The main purpose of this system is to alleviate the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[1084] Data collection

[1085] Users use devices such as smart speakers and tablets to record voice data. For example, a user may record their everyday conversations using a smart speaker (a common name for such devices). The device periodically encrypts the recorded voice data and transmits it to a server in a privacy-protected state.

[1086] Conversation Analysis

[1087] The server processes the received voice data and first converts it into text using speech recognition technology (e.g., a speech recognition API). The server then analyzes the text using a generative AI model (e.g., a natural language processing tool). The analysis includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1088] emotion recognition

[1089] The server's emotion engine analyzes both voice and text data to assess the elderly person's emotional state. An emotion recognition API (generic name) can be used to infer emotions from tone of voice, speaking style, and text content.

[1090] Feedback and Advice

[1091] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[1092] Promoting dialogue

[1093] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1094] Social activity proposals

[1095] The server suggests social activities suitable for seniors. Using a generative AI model, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior, such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1096] Evaluation and Improvement

[1097] The server re-analyzes the elderly person's conversation data after the proposal and evaluates whether their sense of loneliness has been reduced. It then performs another speech-to-text conversion and keyword extraction to confirm any changes in their condition. It also aims to improve the system's accuracy based on the evaluation results, by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[1098] Specific examples

[1099] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text using a speech recognition API. The server then analyzes the text using a generative AI model and detects negative emotions based on the keyword "lonely" and the evaluation of an emotion engine. The server then generates a feedback report and notifies medical professionals with specific content, such as "User B is likely feeling very lonely." The device then sends User B a message such as "Did you talk to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[1100] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[1101] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1102] Step 1: Recording audio data

[1103] The device records the user's everyday conversations. For example, when a user says "Good morning" using a smart speaker or tablet, the conversation is recorded as audio data. The input is the user's voice, and the output is digital audio data.

[1104] Step 2: Encrypt and transmit the audio data

[1105] The device encrypts the recorded voice data and sends it to the server. For example, voice data recorded at a fixed time every day is encrypted using the AES (Advanced Encryption Standard) algorithm. The encrypted data is then uploaded to the server. The input is voice data, and the output is encrypted voice data.

[1106] Step 3: Convert audio data to text

[1107] The server decrypts the received encrypted voice data and converts it into text data using a speech recognition API (for example, Google Speech-to-Text API). The input is encrypted voice data, and the output is text data. For example, voice data saying "Good morning" is converted into the string "Good morning."

[1108] Step 4: Analyzing the text data

[1109] The server analyzes the text data using a generative AI model (e.g., a natural language processing tool). The analysis involves extracting keywords that indicate loneliness or stress. The input is the text data, and the output is the analysis results. For example, keywords such as "lonely" and "alone" are extracted, and their frequency is calculated.

[1110] Step 5: Emotion Recognition

[1111] The server's emotion engine analyzes both voice and text data to evaluate the user's emotional state. It uses emotion recognition APIs (such as IBM Watson or Microsoft Azure Emotion API). The input is text and voice data, and the output is an emotional evaluation result. For example, emotions such as "sadness" or "anxiety" are detected.

[1112] Step 6: Generate a feedback report

[1113] The server generates a feedback report based on the analysis results. The report includes the frequency of extracted keywords, emotional tone, and the emotion engine's evaluation results. The input is the analysis results and the emotion evaluation results, and the output is a feedback report. For example, it may contain specific content such as, "User A has been using the word 'lonely' a lot recently, and the emotion evaluation has confirmed a negative state."

[1114] Step 7: Submit your feedback

[1115] The server generates and sends the feedback report to the healthcare professional or caregiver. For example, the report is sent via email or a web application using the notification function of the internal system. The input is the feedback report, and the output is the report being sent to the healthcare professional or caregiver.

[1116] Step 8: Promote regular dialogue

[1117] The terminal displays or notifies the user of messages that prompt regular interactions. For example, a voice or text message such as "Tell me how you're feeling today" is emitted from the terminal at a fixed time every day. The input is an instruction from the server, and the output is a message that is displayed to the user.

[1118] Step 9: Propose a social activity

[1119] The server selects social activities that match the user's interests and sends that information to the device. For example, a notification may be sent saying, "There's an online book club this Saturday. Would you like to join?" The input is the user's interest data and the analysis results of the generative AI, and the output is the suggested content.

[1120] Step 10: Assess the condition and improve the system

[1121] The server reanalyzes the user's conversation data after the proposal and evaluates whether the sense of loneliness has been reduced. Based on the evaluation results, the system's performance can be improved by updating the analysis algorithm or introducing new keywords. The input is the reanalyzed data, and the output is the system improvements. For example, "Since a reduction in the sense of loneliness was observed, we will introduce a new emotion recognition algorithm."

[1122] (Application example 2)

[1123] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1124] The challenge is to provide an effective dialogue system that can quickly detect and alleviate the feelings of loneliness and negative emotions that elderly people may experience in brick-and-mortar stores. It is also necessary to support elderly people in communicating more smoothly with store staff and other customers.

[1125] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for recording voice data, means for transmitting the recorded voice data to the server, means for converting the voice data to text in the server, means for analyzing the text data and extracting keywords indicating loneliness or stress, means for generating feedback based on the analysis results and transmitting the feedback to a medical professional or caregiver, means for displaying or notifying the elderly person of a message encouraging regular conversation, means for suggesting social activities suitable for the elderly, means for evaluating the elderly person's condition after the suggestions and improving the system's accuracy, means for suggesting conversation topics based on the analysis results, and means for providing store staff with information useful for conversations with the elderly. This makes it possible to detect elderly people's loneliness or negative emotions early on and to suggest effective conversations and social activities.

[1126] "Means for recording voice data" refers to a device or program for collecting and recording voices uttered by the elderly person in digital form.

[1127] The "means for transmitting recorded voice data to a server" refers to a device or program for securely transferring recorded voice data to a server via a network.

[1128] The "means for converting voice data into text on the server" is a program that converts voice data into text format using natural language processing technology.

[1129] The "means for analyzing text data and extracting keywords that indicate loneliness and stress" is a program that analyzes text data using machine learning models and natural language processing technology to identify keywords related to loneliness and stress.

[1130] The "means for generating feedback based on the analysis results and sending it to a medical professional or caregiver" is a program that creates a feedback report based on the analysis results and sends it to a medical professional or caregiver.

[1131] "Means for displaying or notifying messages encouraging elderly people to have regular conversations" refers to a device or program that displays or notifies elderly people in voice or text format with messages to encourage them to have regular conversations.

[1132] The "means for suggesting social activities suitable for the elderly" is a program that suggests appropriate social activities based on the interests and emotional state of each elderly person.

[1133] The "means of evaluating the elderly person's condition after the proposal and improving the accuracy of the system" is a program for reanalyzing the elderly person's reaction and condition after the proposal and improving the accuracy of the system's analysis and the effectiveness of the proposal.

[1134] The "means for proposing a topic for conversation based on the analysis result" is a program that provides an appropriate topic for conversation in accordance with the analyzed emotional state.

[1135] "Means for providing store staff with information that will be useful when interacting with the elderly" refers to a program that provides store staff with the information and suggestions they need to smoothly interact with the elderly.

[1136] The present invention relates to a system for elderly people to use in brick-and-mortar stores. This system is designed to detect loneliness and negative emotions in elderly people at an early stage and promote dialogue. Specific embodiments of the system are described below.

[1137] Data collection

[1138] The device records voice data. When an elderly person (user) speaks using smart glasses or a smartphone in a physical store, the voice data is recorded. For example, in a cafe or shopping store, the user may say, "I've been feeling lonely lately because I haven't been able to see my friends."

[1139] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner using a secure communication protocol (e.g., HTTPS).

[1140] Conversation Analysis

[1141] The server processes the received voice data and first converts it into text using software with speech recognition technology (for example, Python's speech_recognition library).

[1142] The server analyzes the text data using a generative AI model. This analysis involves extracting keywords that indicate loneliness or stress and measuring the frequency of these keywords. For example, if keywords such as "lonely" or "alone" appear frequently, it determines that particular attention is needed.

[1143] Emotion Recognition and Feedback

[1144] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, using technology that infers emotions from tone of voice, speaking style, and text content.

[1145] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[1146] Promoting dialogue and proposing social activities

[1147] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1148] The server suggests social activities suitable for seniors. Using generative AI, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1149] Dialogue suggestions

[1150] The server then suggests conversation topics based on the analysis results. This function allows users to have smoother conversations with store staff and other customers. For example, if the emotion analysis results indicate "loneliness," the server might suggest a conversation topic such as, "How about talking about the recent weather or your hobbies?"

[1151] Examples and prompts

[1152] For example, if an elderly person says, "I've been feeling lonely lately because I haven't been able to see my friends," at a brick-and-mortar cafe, this voice data is sent to the server and converted into text. The analysis results indicate the emotion of "loneliness," and the app suggests topics for conversation, such as, "How about talking about the weather recently or your hobbies?"

[1153] Specific prompt examples:

[1154] User text: "I've been missing my friends lately."

[1155] Emotion analysis result: {'emotion': 'sadness'}

[1156] Suggested topic: "Why don't we talk about the weather these days or our hobbies?"

[1157] This will help reduce the sense of loneliness among the elderly and improve their quality of life, while providing store staff with specific support to facilitate smoother conversations with the elderly.

[1158] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1159] Step 1:

[1160] The device records the voice data spoken by elderly people in a physical store. For example, if a user says, "I miss my friends these days," in a cafe, the device digitally records the voice. The input is the user's voice data, and the output is the recorded voice file.

[1161] Step 2:

[1162] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner. A secure communication protocol (HTTPS) is used for data transmission. The input is the recorded audio file, and the output is the transmission of the encrypted audio file.

[1163] Step 3:

[1164] The server converts the received voice data into text using speech recognition technology. Here, we use the Python speech_recognition library. The input is an encrypted audio file, and the output is text data.

[1165] Step 4:

[1166] The server uses a generative AI model to analyze the text data and extract keywords that indicate loneliness and stress. The input is the text data converted from speech, and the output is the extracted keywords and their frequency. For example, keywords such as "lonely" and "alone" are output.

[1167] Step 5:

[1168] The server's emotion engine further analyzes the voice and text data to assess the elderly person's emotional state. The emotion engine uses a generative AI model, whose input is the extracted keywords and their frequency, and whose output is the emotional state (e.g., "loneliness").

[1169] Step 6:

[1170] The server generates a feedback report based on the analysis results and sends it to medical professionals and caregivers. The input is the evaluation result of the emotion engine, and the output is the feedback report. The report includes the frequency of keywords and emotional tone.

[1171] Step 7:

[1172] The device displays or notifies the elderly with messages encouraging regular conversation. The input is a feedback report, and the output is a notification message (e.g., "Talking to someone every day is good for your health. Please tell us how you feel today.").

[1173] Step 8:

[1174] The server suggests social activities suitable for seniors. It uses generative AI to select activities based on the user's interests and sends that information to the device. The input is the analysis results from the generative AI model, and the output is suggested social activity information. For example, a suggestion might be output such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1175] Step 9:

[1176] The server then suggests conversation topics based on the analysis results. The input is emotional state and keywords, and the output is the conversation topic. For example, a suggestion might be, "Why don't we talk about the recent weather and your hobbies?"

[1177] Step 10:

[1178] The server provides store staff with information that is useful when interacting with elderly people. The input is the emotion analysis results and conversation topics, and the output is conversation support information. This allows staff to communicate smoothly with elderly people.

[1179] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1180] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1181] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1182] [Fourth embodiment]

[1183] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1184] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1185] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1186] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1187] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1188] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1189] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1190] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1191] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1192] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1193] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1194] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1195] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1196] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[1197] The main components of this system are: "means for recording voice data," "means for sending voice data to a server," "means for converting voice data into text," "means for analyzing text data and extracting keywords," "means for generating feedback and sending it to medical professionals and caregivers," "means for encouraging regular dialogue with the elderly," "means for suggesting social activities," and "means for assessing the condition of the elderly and improving the accuracy of the system."

[1198] An embodiment of this system is described in detail below.

[1199] Data collection

[1200] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[1201] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[1202] Conversation Analysis

[1203] The server processes the received voice data and first converts the voice into text using voice recognition technology.

[1204] The server uses a generative AI to extract keywords that indicate loneliness or stress from the text data. For example, if keywords such as "lonely" or "solitude" appear frequently, it is determined that the user is feeling lonely.

[1205] The server also performs sentiment analysis, assessing the overall emotional tone (negative, positive, neutral) of the text data.

[1206] Feedback and Advice

[1207] The server generates a feedback report based on the analysis, including keyword frequency and emotional tone, which is sent to medical professionals and caregivers.

[1208] For example, specific advice is provided such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[1209] Promoting dialogue

[1210] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1211] Social activity proposals

[1212] The server will suggest suitable social activities for seniors, such as sending notifications recommending them to join hobby clubs or online meetups.

[1213] Specifically, you might receive a notification such as, "There's a painting class at your local community center this Saturday. Would you like to join?"

[1214] Evaluation and Improvement

[1215] The server then reanalyzes the elderly person's conversation data after the suggestions are made and evaluates whether their sense of loneliness has been alleviated, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[1216] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system.

[1217] Specific examples

[1218] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server, converted into text, and analyzed. The result is that the keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in dialogue by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and their quality of life improves.

[1219] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[1220] The processing flow will be explained below.

[1221] Step 1:

[1222] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[1223] Step 2:

[1224] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[1225] Step 3:

[1226] The server decrypts the received encrypted voice data and converts the voice into text information. It uses voice recognition technology to carry out the process of converting the voice data into text data.

[1227] Step 4:

[1228] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1229] Step 5:

[1230] The server evaluates the emotional tone of the entire text data, analyzing it as negative, positive, or neutral to understand the user's emotional state.

[1231] Step 6:

[1232] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords and an evaluation of emotional tone, and is sent to medical professionals and caregivers.

[1233] Step 7:

[1234] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[1235] Step 8:

[1236] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1237] Step 9:

[1238] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[1239] Step 10:

[1240] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1241] Step 11:

[1242] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[1243] Step 12:

[1244] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate voice recognition technology.

[1245] Example 1

[1246] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1247] Elderly people are prone to feelings of loneliness and stress, which often have a negative impact on their health. Conventional measures make it difficult to detect early signs of loneliness and stress and take appropriate measures. In addition, feedback and responses are often inappropriate, leading to the problem of not being able to adequately support the physical and mental health of elderly people.

[1248] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1249] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data using a generative AI model and extracting keywords indicative of loneliness or stress, a means for generating a feedback report based on the extracted keywords and emotional tone and sending it to a medical professional or caregiver, a means for displaying or notifying a message encouraging the elderly to have regular conversations, a means for suggesting social activities suitable for the elderly, and a means for reanalyzing the elderly's conversation data after the suggestions and evaluating it to improve the accuracy of the system. This makes it possible to detect signs of loneliness or stress early through the elderly's everyday conversations and take appropriate countermeasures.

[1250] "Voice data" is data in digital form obtained from a user's voice.

[1251] A "terminal" is an electronic device for recording and transmitting voice data, including smart speakers and tablets.

[1252] A "server" is a computer system for receiving and processing audio data.

[1253] The "means for converting to text" is a function for converting voice data into text information, and voice recognition technology is used for this.

[1254] A "generative AI model" is an artificial intelligence model that uses machine learning and natural language processing techniques to analyze data and extract specific information.

[1255] "Keywords" are specific words or phrases that indicate loneliness or stress.

[1256] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[1257] "Messages encouraging regular dialogue" are notifications and messages that are displayed to encourage everyday dialogue with elderly people.

[1258] "Means for suggesting social activities" is a suggestion function for providing elderly people with opportunities for community activities and interaction.

[1259] The "means of evaluation" is an evaluation function for reanalyzing the condition of the elderly person and improving the accuracy of the system.

[1260] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[1261] Data collection

[1262] 1. The device records voice data. For example, an elderly person (user) uses a smart speaker or tablet to record their everyday conversations. The recorded voice data is periodically encrypted by the device and sent to a server in a privacy-protected manner. AES (Advanced Encryption Standard) technology is used for encryption, and the data is sent using the HTTPS protocol.

[1263] Conversation Analysis

[1264] 2. The server processes the received voice data and first converts it into text using the Google Cloud Speech-to-Text API. For example, the resulting text is "I was gardening today."

[1265] 3. The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data. Specifically, it uses a machine learning algorithm to extract keywords such as "lonely," "solitude," and "alone." An example of a prompt is, "Please extract keywords indicating loneliness from this text data."

[1266] 4. The server then performs further sentiment analysis to assess the overall emotional tone (negative, positive, neutral) of the text data using the Sentiment Analysis API.

[1267] Feedback and Advice

[1268] 5. The server generates a feedback report based on the analysis results. The report includes the frequency of keywords and emotional tone, and is sent to medical professionals and caregivers. For example, a report might say, "User A has been using the word lonely a lot recently. The emotional tone is negative."

[1269] Promoting dialogue

[1270] 6. The device will display or notify the elderly with messages encouraging regular conversations. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" will be sent.

[1271] Social activity proposals

[1272] 7. The server will suggest suitable social activities for the elderly. For example, a notification will be sent to the elderly saying, "There will be an art class at the local community center this Saturday. Would you like to join?"

[1273] Evaluation and Improvement

[1274] 8. The server reanalyzes the elderly person's conversation data after the suggestion and evaluates whether the sense of loneliness has been reduced, for example, by checking whether the frequency of keywords indicating loneliness has decreased.

[1275] 9. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the accuracy of the system, which involves retraining the machine learning algorithm and introducing new datasets.

[1276] Specific examples

[1277] For example, user A might say through a smart speaker, "I've been alone a lot lately and I feel lonely." This voice data is sent to a server and converted into text using the Google Cloud Speech-to-Text API. A generative AI model then analyzes the text data, finding that the keyword "lonely" appears frequently. The server then sends feedback to a medical professional, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A with a message asking, "Have you been able to talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases and his quality of life improves.

[1278] In this way, the present invention is a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[1279] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1280] Step 1:

[1281] The device records voice data. As input, it acquires the elderly person's (user's) everyday conversation. As operation, when the user speaks to the smart speaker or tablet, the device records the voice in digital format. As output, it generates the recorded voice data.

[1282] Step 2:

[1283] The terminal encrypts the recorded voice data and sends it to the server. As input, there is the recorded voice data. As operation, the terminal encrypts the voice data using AES technology to ensure the security of the data. Then, it sends the encrypted data to the server via HTTPS protocol. As output, the encrypted voice data is sent to the server.

[1284] Step 3:

[1285] The server converts the received voice data into text. As input, it receives encrypted voice data. In operation, the server uses speech recognition technology to convert the data into text format. Specifically, it uses the Google Cloud Speech-to-Text API. As output, it generates text data.

[1286] Step 4:

[1287] The server uses a generative AI model to extract keywords indicating loneliness and stress from text data. The input is text data. The operation is to apply a machine learning algorithm to detect specific keywords. An example of a prompt is "Please extract keywords indicating loneliness from this text data." The output is a list of extracted keywords.

[1288] Step 5:

[1289] The server performs sentiment analysis and evaluates the emotional tone of the entire text data. The input is text data. The operation is to evaluate the emotional tone of the data using the Sentiment Analysis API. The output is an evaluation result of the emotional tone (negative, positive, neutral).

[1290] Step 6:

[1291] The server generates a feedback report based on the analysis results. The inputs are a list of keywords and the emotional tone evaluation results. In operation, the server integrates these data to create a feedback report. As an output, a feedback report is generated. This report includes the frequency of keywords and the emotional tone.

[1292] Step 7:

[1293] The server sends the generated feedback report to the healthcare professional or caregiver. As input, there is a feedback report. As action, the server sends the report using secure email or a dedicated dashboard. As output, the feedback report has been sent.

[1294] Step 8:

[1295] The device displays or notifies the elderly with messages encouraging regular conversation. The input is an instruction from the system. The action is to display a notification on the screen or play a message by voice. For example, a notification such as "Talking to someone every day is good for your health. Tell us how you're feeling today" is sent. The output is to encourage the elderly to have a conversation.

[1296] Step 9:

[1297] The server suggests suitable social activities for the elderly. As input, it has the feedback report and the profile data of the elderly. As operation, the system selects suitable activities for the elderly and generates a notification. For example, a notification may be generated such as "There is a painting class at your local community center this Saturday. Would you like to join?" As output, the suggested social activities are sent to the elderly.

[1298] Step 10:

[1299] The server re-analyzes the elderly person's conversation data after the suggestions and evaluates whether their feelings of loneliness have been reduced. The input is the new speech data after the suggestions. The operation is to re-apply the method from steps 3 to 5 described above to obtain new analysis results. The output is an evaluation result of whether the signs of loneliness have been reduced.

[1300] Step 11:

[1301] Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy. The input is the reanalysis results. The operation involves retraining the machine learning algorithm and introducing a new dataset. The output is a new system with improved algorithm accuracy.

[1302] (Application example 1)

[1303] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1304] Situations in which elderly people are prone to feeling loneliness and stress in their daily lives can have a negative impact on their physical and mental health. Furthermore, if efforts to reduce loneliness are insufficient, there is a higher risk that an emergency at home will go unnoticed. The present invention aims to collect and analyze the everyday conversations of elderly people, detect signs of loneliness early, provide appropriate countermeasures, and reduce the risks associated with loneliness.

[1305] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1306] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and extracting keywords indicating loneliness or stress using a generative AI model, and a means for promptly notifying family members or caregivers if an abnormality is detected. This makes it possible to detect loneliness or stress in the elderly person's everyday conversations early on and take necessary measures promptly.

[1307] "Voice data" refers to data in which voice is electronically recorded, and includes everyday conversations between elderly people.

[1308] "Recording means" refers to devices or technologies for storing audio data, including, for example, smartphones, tablets, and smart speakers.

[1309] A "server" is a computer system that receives, processes, and analyzes voice data.

[1310] "Means for converting to text" refers to technology for converting voice data into text data in sentence format, and includes voice recognition software and APIs.

[1311] "Keyword extraction means" refers to technology for identifying and extracting specific words and phrases related to loneliness and stress from text data.

[1312] A "generative AI model" is an artificial intelligence algorithm that analyzes the meaning and sentiment of text data based on large datasets.

[1313] The "means for generating feedback" is a technology that generates information to suggest measures to medical professionals and caregivers based on the results of text analysis.

[1314] "Means of notification" refers to technologies for quickly transmitting analysis results and feedback information to family members and caregivers, and includes email and messaging services.

[1315] "Means for encouraging regular dialogue" refers to technology that displays or notifies elderly people of messages encouraging dialogue on a regular basis.

[1316] "Means for suggesting social activities" refers to techniques for suggesting events and club activities suitable for the elderly.

[1317] "Means for improving the accuracy of the system" refers to methods or techniques for updating the system's algorithms or databases based on the analysis results, thereby improving the accuracy and efficiency of the analysis.

[1318] This invention relates to a system that collects and analyzes the everyday conversations of the elderly and detects elderly people who are prone to feeling lonely at an early stage. The system aims to reduce the elderly's sense of loneliness by detecting signs of loneliness based on the collected voice data and suggesting appropriate conversations and social activities.

[1319] Data collection

[1320] The device records audio data. For example, an elderly user might use a smartphone to record everyday conversations. The device periodically encrypts the recorded audio data and transmits it to a server in a privacy-protected manner. The hardware used is a standard Android or iOS device, and encryption is performed using the Python "pycryptodome" library.

[1321] Conversation Analysis

[1322] The server processes the received voice data and first converts the speech to text. This conversion is performed using Google's Cloud Speech-to-Text API. Next, the server uses a generative AI model (e.g., OpenAI GPT-4) to extract keywords indicating loneliness or stress from the text data. For example, if keywords such as "lonely" and "solitude" appear frequently, it is determined that the user is feeling lonely. The server then performs sentiment analysis to evaluate the emotional tone (negative, positive, or neutral) of the entire text data.

[1323] Feedback and Advice

[1324] The server generates a feedback report based on the analysis results. This report includes keyword frequency, emotional tone, and the results of emotion analysis using a generative AI model, and is sent to family members or caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently. You should talk to him more often" is provided. If an abnormality is detected, the family or caregiver is promptly notified.

[1325] Promoting dialogue

[1326] The device will display or notify the elderly with messages encouraging regular conversations, such as "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1327] Social activity proposals

[1328] The server suggests social activities suitable for seniors. For example, notifications are sent to seniors recommending them to join hobby clubs or online social gatherings. Specifically, a notification might say, "There's a painting class at the local community center this Saturday. Would you like to join?"

[1329] Evaluation and Improvement

[1330] The server reanalyzes the elderly person's conversation data after the proposal and evaluates whether their feelings of loneliness have been alleviated. For example, it checks whether the frequency of keywords indicating loneliness has decreased. Based on the evaluation results, the server updates the algorithm and introduces new keywords to improve the system's accuracy.

[1331] Specific examples

[1332] For example, user A might say on his smartphone, "I've been alone a lot lately and I feel lonely." This voice data is sent to the server, converted into text, and analyzed. The keyword "lonely," which indicates feelings of loneliness, appears frequently. The server then sends feedback to the family, offering specific advice such as, "User A seems lonely, so we recommend regular phone consultations." At the same time, the device prompts user A to engage in conversation by notifying them, "Did you talk to your family or friends today?" The server also suggests social activities to user A, such as, "Why not join a local hobby club?" As a result, user A's sense of loneliness decreases, and his quality of life improves.

[1333] Prompt Sentence Examples

[1334] Audio data: "I've been spending a lot of time alone lately and I feel lonely."

[1335] Text data: "I've been spending a lot of time alone lately and I feel lonely."

[1336] Keywords: "Spending a lot of time alone", "lonely"

[1337] Feedback: "User B feels lonely. Please consider visiting or contacting them regularly."

[1338] In this way, the present invention provides a system that supports the physical and mental health of elderly people by detecting feelings of loneliness early through their everyday conversations and taking appropriate countermeasures.

[1339] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1340] Step 1:

[1341] A user uses a smartphone to have everyday conversations.

[1342] (Input) User's everyday conversational voice.

[1343] (Output) Audio data recorded on a smartphone.

[1344] (Specific operation) When a user says "I'm lonely" to their smartphone, the voice is recorded by the smartphone's built-in microphone.

[1345] Step 2:

[1346] The device periodically encrypts the recorded audio data and sends it to the server.

[1347] (Input) Recorded audio data.

[1348] (Output) Encrypted audio data.

[1349] (Specific operation) The smartphone encrypts the voice data using AES (Advanced Encryption Standard) and sends it to the server via HTTPS.

[1350] Step 3:

[1351] The server converts the received voice data into text.

[1352] (Input) Encrypted audio data.

[1353] (Output) Text data.

[1354] (Specific operation) The server decrypts the encrypted voice data and converts the voice data into text data using Google's Cloud Speech-to-Text API.

[1355] Step 4:

[1356] The server uses a generative AI model to extract keywords indicating loneliness and stress from the text data.

[1357] (Input) Text data.

[1358] (Output) Extracted keywords.

[1359] (Specific operation) The server inputs the text data into a natural language processing algorithm (e.g., OpenAI GPT-4) and extracts keywords such as "lonely" and "solitude."

[1360] Step 5:

[1361] The server also evaluates the emotional tone of the entire text data.

[1362] (Input) Text data and extracted keywords.

[1363] (Output) Sentiment analysis result (negative, positive, neutral).

[1364] (Specific operation) Using a generative AI model, the emotional tone of the entire text data is analyzed to determine emotional categories such as "negative" or "positive."

[1365] Step 6:

[1366] The server generates a feedback report based on the analysis results.

[1367] (Input) Extracted keywords and sentiment analysis results.

[1368] (Output) Feedback report.

[1369] (Specific operation) Based on the extracted keywords and the results of sentiment analysis, the server generates a feedback report such as, "User A has been using the word lonely a lot recently. You should increase the frequency of conversations with him."

[1370] Step 7:

[1371] If an abnormality is detected, the server will promptly notify family members or caregivers.

[1372] (Input) Feedback report.

[1373] (Output) A warning notification.

[1374] (Specific operation) The server uses the contact information of family members or caregivers to send messages via email or SMS such as "User A is feeling lonely, so we recommend regular phone consultations."

[1375] Step 8:

[1376] The device displays or notifies the elderly person of messages encouraging regular dialogue.

[1377] (Input) Instructions from the server.

[1378] (Output) Messages that are displayed on the terminal.

[1379] (Specific action) The smartphone displays a message such as, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1380] Step 9:

[1381] The server suggests social activities suitable for seniors.

[1382] (Input) Analysis results and social activity information in the database.

[1383] (Output) The proposal message.

[1384] (Specific operation) The server generates a message such as "There's a painting class at the local community center this Saturday. Would you like to join?" and sends it to the terminal.

[1385] Step 10:

[1386] The server reanalyzes the elderly's conversation data after the proposal to improve the accuracy of the system.

[1387] (Input) Re-collected audio data.

[1388] (Output) Updated algorithm to improve system accuracy.

[1389] (Specific operation) The server reanalyzes the elderly's conversation data to see if the frequency of keywords indicating loneliness has decreased, and uses the results to update the algorithm.

[1390] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1391] This invention relates to a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes their emotions. The system aims to reduce the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[1392] The main components of this system are: a means for recording voice data, a means for sending voice data to a server, a means for converting voice data into text, a means for analyzing text data and extracting keywords, an emotion engine, a means for generating feedback and sending it to medical professionals and caregivers, a means for encouraging regular dialogue with the elderly, a means for suggesting social activities, and a means for assessing the condition of the elderly and improving the accuracy of the system.

[1393] An embodiment of this system is described in detail below.

[1394] Data collection

[1395] The device records voice data. For example, elderly users can use smart speakers or tablets to record their everyday conversations.

[1396] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner.

[1397] Conversation Analysis

[1398] The server processes the received voice data and first converts the voice data into text, then uses voice recognition technology to convert the voice data into text.

[1399] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1400] emotion recognition

[1401] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[1402] Feedback and Advice

[1403] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[1404] For example, specific advice may be provided such as, "User A has been using the word lonely a lot recently, and the emotion engine's evaluation has confirmed that he has strong negative emotions. We recommend that you increase the frequency of conversations with him and suggest some hobby activities."

[1405] Promoting dialogue

[1406] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1407] Social activity proposals

[1408] The server suggests social activities suitable for seniors, using generative AI to select activities based on the user's interests and sending that information to the device.

[1409] Specifically, the device will send a notification to the elderly such as, "There will be a painting class at the local community center this Saturday. Would you like to join?"

[1410] Evaluation and Improvement

[1411] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been reduced. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[1412] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[1413] Specific examples

[1414] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text. The server performs analysis and detects negative emotions through the keyword "lonely" and evaluation by the emotion engine. The server then generates a feedback report and notifies medical professionals that "User B is likely feeling very lonely." The device then sends User B messages such as "Have you talked to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[1415] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[1416] The processing flow will be explained below.

[1417] Step 1:

[1418] The device records the voice of the user (elderly person). The user uses a smart speaker or tablet to naturally record everyday conversations.

[1419] Step 2:

[1420] The device encrypts the recorded audio data and periodically transmits it to a server. The encryption is for privacy reasons, and the data transfer occurs, for example, at a fixed time every night.

[1421] Step 3:

[1422] The server decrypts the received encrypted voice data and converts the voice data into text data. It uses voice recognition technology to carry out the process of converting the voice data into text information.

[1423] Step 4:

[1424] The server uses generative AI to analyze the text data, which includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1425] Step 5:

[1426] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, for example, by inferring emotion from the tone of voice, speaking style, and text content.

[1427] Step 6:

[1428] The server generates a feedback report based on the analysis results, which includes the frequency of extracted keywords, emotional tone, and the evaluation results of the emotion engine, and is sent to medical professionals and caregivers.

[1429] Step 7:

[1430] Healthcare professionals and caregivers receive the generated feedback report and consider appropriate measures, such as increasing the frequency of interactions with the elderly or suggesting specific activities.

[1431] Step 8:

[1432] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1433] Step 9:

[1434] The server proposes social activities suitable for each elderly person, using generative AI to select activities based on the user's interests and sending that information to the device.

[1435] Step 10:

[1436] The device will then notify the elderly of suggested social activities, such as sending a message saying, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1437] Step 11:

[1438] The server re-analyzes the elderly person's conversation data after the proposal to evaluate whether their sense of loneliness has been alleviated. It then performs speech-to-text conversion and keyword extraction again to confirm any changes in their condition.

[1439] Step 12:

[1440] The server will use the evaluation results to improve the accuracy of the system by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[1441] Specific examples

[1442] For example, if user C says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text data.

[1443] The server then performs an analysis and detects that the keyword "lonely" appears frequently, and the emotion engine identifies negative emotions.

[1444] The server then generates a feedback report and notifies the medical professional of the assessment result that "User C is likely to be feeling a strong sense of loneliness."

[1445] The terminal sends a message to user C prompting a conversation, such as "Have you talked to anyone today?"

[1446] Additionally, the server suggests social activities to User C, such as "Would you like to join a local online book club?"

[1447] This reduces User C's sense of loneliness and improves his quality of life.

[1448] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[1449] Example 2

[1450] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1451] There is a lack of systems that can detect elderly people's feelings of loneliness and emotional states early and take appropriate countermeasures. Conventional systems have difficulty analyzing elderly people's emotional states from their everyday conversations and providing feedback. Furthermore, there are limited means to effectively communicate this data to medical professionals and caregivers. The present invention aims to solve these problems, reduce elderly people's feelings of loneliness, and improve their quality of life.

[1452] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1453] In this invention, the server includes a means for converting voice data into text, a means for analyzing the text data and voice data and evaluating the emotional state, and a means for generating a feedback report based on the analysis results and sending it to a medical professional or caregiver, thereby enabling the emotional state of an elderly person to be analyzed quickly and accurately from their everyday conversations and appropriate countermeasures to be implemented.

[1454] "Audio data" refers to digital audio files that record everyday conversations between elderly people.

[1455] A "server" is an information processing device that processes and analyzes voice data and generates feedback.

[1456] "Text data" refers to character information converted from voice data using voice recognition technology.

[1457] "Keywords" are important words in the text data that indicate feelings of loneliness or stress.

[1458] An "emotion engine" is software that analyzes voice and text data to assess emotional states.

[1459] "Analysis results" are information obtained from analyzing text data and evaluating it using an emotion engine.

[1460] A "feedback report" is a report generated based on the analysis results, including keyword frequency and emotional tone.

[1461] "Medical professionals" are professionals such as doctors and nurses who receive the feedback reports and use them to help care for older people.

[1462] A "caregiver" is a person who receives the feedback report and provides support for the elderly person's daily life.

[1463] "Promoting regular dialogue" means displaying or notifying elderly people of messages encouraging dialogue on a regular basis.

[1464] "Social activity suggestions" refers to providing information on social activities that encourage participation based on the interests of the elderly.

[1465] "Evaluating the state" involves reanalyzing the elderly person's conversation data after the proposal to confirm any changes in their feelings of loneliness or emotional state.

[1466] "Improving the accuracy of the system" means updating the analysis algorithm, introducing new keywords, and improving emotion recognition technology to improve the system's performance.

[1467] This invention is a system that collects and analyzes the everyday conversations of elderly people, detects elderly people who are prone to feeling lonely at an early stage, and combines it with an emotion engine that recognizes emotions. The main purpose of this system is to alleviate the sense of loneliness of elderly people by detecting feelings of loneliness and emotional states based on collected voice and text data and suggesting appropriate conversations and social activities.

[1468] Data collection

[1469] Users use devices such as smart speakers and tablets to record voice data. For example, a user may record their everyday conversations using a smart speaker (a common name for such devices). The device periodically encrypts the recorded voice data and transmits it to a server in a privacy-protected state.

[1470] Conversation Analysis

[1471] The server processes the received voice data and first converts it into text using speech recognition technology (e.g., a speech recognition API). The server then analyzes the text using a generative AI model (e.g., a natural language processing tool). The analysis includes steps to extract keywords that indicate loneliness and stress and measure the frequency of these keywords.

[1472] emotion recognition

[1473] The server's emotion engine analyzes both voice and text data to assess the elderly person's emotional state. An emotion recognition API (generic name) can be used to infer emotions from tone of voice, speaking style, and text content.

[1474] Feedback and Advice

[1475] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[1476] Promoting dialogue

[1477] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1478] Social activity proposals

[1479] The server suggests social activities suitable for seniors. Using a generative AI model, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior, such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1480] Evaluation and Improvement

[1481] The server re-analyzes the elderly person's conversation data after the proposal and evaluates whether their sense of loneliness has been reduced. It then performs another speech-to-text conversion and keyword extraction to confirm any changes in their condition. It also aims to improve the system's accuracy based on the evaluation results, by updating the analysis algorithm, introducing new keywords, and applying more accurate emotion recognition technology.

[1482] Specific examples

[1483] For example, if User B says through a smart speaker, "I've been alone a lot lately and I feel lonely," this voice data is sent to the server and converted into text using a speech recognition API. The server then analyzes the text using a generative AI model and detects negative emotions based on the keyword "lonely" and the evaluation of an emotion engine. The server then generates a feedback report and notifies medical professionals with specific content, such as "User B is likely feeling very lonely." The device then sends User B a message such as "Did you talk to anyone today?" to encourage dialogue. The server also suggests social activities to User B, such as "Why not join a local online book club?" This reduces User B's sense of loneliness and improves their quality of life.

[1484] In this way, by combining emotion recognition technology, the present invention provides a system that can detect feelings of loneliness early through everyday conversations among elderly people and take appropriate countermeasures to support their physical and mental health.

[1485] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1486] Step 1: Recording audio data

[1487] The device records the user's everyday conversations. For example, when a user says "Good morning" using a smart speaker or tablet, the conversation is recorded as audio data. The input is the user's voice, and the output is digital audio data.

[1488] Step 2: Encrypt and transmit the audio data

[1489] The device encrypts the recorded voice data and sends it to the server. For example, voice data recorded at a fixed time every day is encrypted using the AES (Advanced Encryption Standard) algorithm. The encrypted data is then uploaded to the server. The input is voice data, and the output is encrypted voice data.

[1490] Step 3: Convert audio data to text

[1491] The server decrypts the received encrypted voice data and converts it into text data using a speech recognition API (for example, Google Speech-to-Text API). The input is encrypted voice data, and the output is text data. For example, voice data saying "Good morning" is converted into the string "Good morning."

[1492] Step 4: Analyzing the text data

[1493] The server analyzes the text data using a generative AI model (e.g., a natural language processing tool). The analysis involves extracting keywords that indicate loneliness or stress. The input is the text data, and the output is the analysis results. For example, keywords such as "lonely" and "alone" are extracted, and their frequency is calculated.

[1494] Step 5: Emotion Recognition

[1495] The server's emotion engine analyzes both voice and text data to evaluate the user's emotional state. It uses emotion recognition APIs (such as IBM Watson or Microsoft Azure Emotion API). The input is text and voice data, and the output is an emotional evaluation result. For example, emotions such as "sadness" or "anxiety" are detected.

[1496] Step 6: Generate a feedback report

[1497] The server generates a feedback report based on the analysis results. The report includes the frequency of extracted keywords, emotional tone, and the emotion engine's evaluation results. The input is the analysis results and the emotion evaluation results, and the output is a feedback report. For example, it may contain specific content such as, "User A has been using the word 'lonely' a lot recently, and the emotion evaluation has confirmed a negative state."

[1498] Step 7: Submit your feedback

[1499] The server generates and sends the feedback report to the healthcare professional or caregiver. For example, the report is sent via email or a web application using the notification function of the internal system. The input is the feedback report, and the output is the report being sent to the healthcare professional or caregiver.

[1500] Step 8: Promote regular dialogue

[1501] The terminal displays or notifies the user of messages that prompt regular interactions. For example, a voice or text message such as "Tell me how you're feeling today" is emitted from the terminal at a fixed time every day. The input is an instruction from the server, and the output is a message that is displayed to the user.

[1502] Step 9: Propose a social activity

[1503] The server selects social activities that match the user's interests and sends that information to the device. For example, a notification may be sent saying, "There's an online book club this Saturday. Would you like to join?" The input is the user's interest data and the analysis results of the generative AI, and the output is the suggested content.

[1504] Step 10: Assess the condition and improve the system

[1505] The server reanalyzes the user's conversation data after the proposal and evaluates whether the sense of loneliness has been reduced. Based on the evaluation results, the system's performance can be improved by updating the analysis algorithm or introducing new keywords. The input is the reanalyzed data, and the output is the system improvements. For example, "Since a reduction in the sense of loneliness was observed, we will introduce a new emotion recognition algorithm."

[1506] (Application example 2)

[1507] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1508] The challenge is to provide an effective dialogue system that can quickly detect and alleviate the feelings of loneliness and negative emotions that elderly people may experience in brick-and-mortar stores. It is also necessary to support elderly people in communicating more smoothly with store staff and other customers.

[1509] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for recording voice data, means for transmitting the recorded voice data to the server, means for converting the voice data to text in the server, means for analyzing the text data and extracting keywords indicating loneliness or stress, means for generating feedback based on the analysis results and transmitting the feedback to a medical professional or caregiver, means for displaying or notifying the elderly person of a message encouraging regular conversation, means for suggesting social activities suitable for the elderly, means for evaluating the elderly person's condition after the suggestions and improving the system's accuracy, means for suggesting conversation topics based on the analysis results, and means for providing store staff with information useful for conversations with the elderly. This makes it possible to detect elderly people's loneliness or negative emotions early on and to suggest effective conversations and social activities.

[1510] "Means for recording voice data" refers to a device or program for collecting and recording voices uttered by the elderly person in digital form.

[1511] The "means for transmitting recorded voice data to a server" refers to a device or program for securely transferring recorded voice data to a server via a network.

[1512] The "means for converting voice data into text on the server" is a program that converts voice data into text format using natural language processing technology.

[1513] The "means for analyzing text data and extracting keywords that indicate loneliness and stress" is a program that analyzes text data using machine learning models and natural language processing technology to identify keywords related to loneliness and stress.

[1514] The "means for generating feedback based on the analysis results and sending it to a medical professional or caregiver" is a program that creates a feedback report based on the analysis results and sends it to a medical professional or caregiver.

[1515] "Means for displaying or notifying messages encouraging elderly people to have regular conversations" refers to a device or program that displays or notifies elderly people in voice or text format with messages to encourage them to have regular conversations.

[1516] The "means for suggesting social activities suitable for the elderly" is a program that suggests appropriate social activities based on the interests and emotional state of each elderly person.

[1517] The "means of evaluating the elderly person's condition after the proposal and improving the accuracy of the system" is a program for reanalyzing the elderly person's reaction and condition after the proposal and improving the accuracy of the system's analysis and the effectiveness of the proposal.

[1518] The "means for proposing a topic for conversation based on the analysis result" is a program that provides an appropriate topic for conversation in accordance with the analyzed emotional state.

[1519] "Means for providing store staff with information that will be useful when interacting with the elderly" refers to a program that provides store staff with the information and suggestions they need to smoothly interact with the elderly.

[1520] The present invention relates to a system for elderly people to use in brick-and-mortar stores. This system is designed to detect loneliness and negative emotions in elderly people at an early stage and promote dialogue. Specific embodiments of the system are described below.

[1521] Data collection

[1522] The device records voice data. When an elderly person (user) speaks using smart glasses or a smartphone in a physical store, the voice data is recorded. For example, in a cafe or shopping store, the user may say, "I've been feeling lonely lately because I haven't been able to see my friends."

[1523] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner using a secure communication protocol (e.g., HTTPS).

[1524] Conversation Analysis

[1525] The server processes the received voice data and first converts it into text using software with speech recognition technology (for example, Python's speech_recognition library).

[1526] The server analyzes the text data using a generative AI model. This analysis involves extracting keywords that indicate loneliness or stress and measuring the frequency of these keywords. For example, if keywords such as "lonely" or "alone" appear frequently, it determines that particular attention is needed.

[1527] Emotion Recognition and Feedback

[1528] The server's emotion engine analyzes both the voice and text data to assess the elderly person's emotional state, using technology that infers emotions from tone of voice, speaking style, and text content.

[1529] The server generates a feedback report based on the analysis results. This report includes the frequency of the extracted keywords, the emotional tone, and the evaluation results of the emotion engine. The feedback report is sent to medical professionals and caregivers. For example, specific advice such as "User A has been using the word lonely a lot recently, and the emotion engine's evaluation confirmed that he has a strong negative emotion. It is recommended that you increase the frequency of conversations with him and suggest some hobby activities" is provided.

[1530] Promoting dialogue and proposing social activities

[1531] The device displays or notifies the elderly with a message encouraging regular conversation, for example, "Talking to someone every day is good for your health. Tell us how you're feeling today."

[1532] The server suggests social activities suitable for seniors. Using generative AI, it selects activities based on the user's interests and sends that information to the device. Specifically, the device sends a notification to the senior such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1533] Dialogue suggestions

[1534] The server then suggests conversation topics based on the analysis results. This function allows users to have smoother conversations with store staff and other customers. For example, if the emotion analysis results indicate "loneliness," the server might suggest a conversation topic such as, "How about talking about the recent weather or your hobbies?"

[1535] Examples and prompts

[1536] For example, if an elderly person says, "I've been feeling lonely lately because I haven't been able to see my friends," at a brick-and-mortar cafe, this voice data is sent to the server and converted into text. The analysis results indicate the emotion of "loneliness," and the app suggests topics for conversation, such as, "How about talking about the weather recently or your hobbies?"

[1537] Specific prompt examples:

[1538] User text: "I've been missing my friends lately."

[1539] Emotion analysis result: {'emotion': 'sadness'}

[1540] Suggested topic: "Why don't we talk about the weather these days or our hobbies?"

[1541] This will help reduce the sense of loneliness among the elderly and improve their quality of life, while providing store staff with specific support to facilitate smoother conversations with the elderly.

[1542] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1543] Step 1:

[1544] The device records the voice data spoken by elderly people in a physical store. For example, if a user says, "I miss my friends these days," in a cafe, the device digitally records the voice. The input is the user's voice data, and the output is the recorded voice file.

[1545] Step 2:

[1546] The device periodically encrypts the recorded audio data and transmits it to the server in a privacy-protected manner. A secure communication protocol (HTTPS) is used for data transmission. The input is the recorded audio file, and the output is the transmission of the encrypted audio file.

[1547] Step 3:

[1548] The server converts the received voice data into text using speech recognition technology. Here, we use the Python speech_recognition library. The input is an encrypted audio file, and the output is text data.

[1549] Step 4:

[1550] The server uses a generative AI model to analyze the text data and extract keywords that indicate loneliness and stress. The input is the text data converted from speech, and the output is the extracted keywords and their frequency. For example, keywords such as "lonely" and "alone" are output.

[1551] Step 5:

[1552] The server's emotion engine further analyzes the voice and text data to assess the elderly person's emotional state. The emotion engine uses a generative AI model, whose input is the extracted keywords and their frequency, and whose output is the emotional state (e.g., "loneliness").

[1553] Step 6:

[1554] The server generates a feedback report based on the analysis results and sends it to medical professionals and caregivers. The input is the evaluation result of the emotion engine, and the output is the feedback report. The report includes the frequency of keywords and emotional tone.

[1555] Step 7:

[1556] The device displays or notifies the elderly with messages encouraging regular conversation. The input is a feedback report, and the output is a notification message (e.g., "Talking to someone every day is good for your health. Please tell us how you feel today.").

[1557] Step 8:

[1558] The server suggests social activities suitable for seniors. It uses generative AI to select activities based on the user's interests and sends that information to the device. The input is the analysis results from the generative AI model, and the output is suggested social activity information. For example, a suggestion might be output such as, "There's a painting class this Saturday at the local community center. Would you like to join?"

[1559] Step 9:

[1560] The server then suggests conversation topics based on the analysis results. The input is emotional state and keywords, and the output is the conversation topic. For example, a suggestion might be, "Why don't we talk about the recent weather and your hobbies?"

[1561] Step 10:

[1562] The server provides store staff with information that is useful when interacting with elderly people. The input is the emotion analysis results and conversation topics, and the output is conversation support information. This allows staff to communicate smoothly with elderly people.

[1563] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1564] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1565] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1566] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1567] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1568] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1569] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1570] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1571] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1572] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1573] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1574] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1575] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1576] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1577] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1578] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1579] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1580] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1581] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1582] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1583] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1584] The following is further disclosed regarding the above embodiment.

[1585] (Claim 1)

[1586] means for recording audio data;

[1587] means for transmitting the recorded voice data to a server;

[1588] means for converting voice data into text at the server;

[1589] A means of analyzing text data and extracting keywords that indicate loneliness and stress;

[1590] a means for generating and transmitting feedback based on the analysis results to healthcare professionals and caregivers;

[1591] A means for displaying or notifying a message to encourage regular dialogue to the elderly person;

[1592] A means of suggesting suitable social activities for older people;

[1593] A means to evaluate the condition of the elderly person after the proposal and improve the accuracy of the system;

[1594] A system including:

[1595] (Claim 2)

[1596] 10. The system of claim 1, wherein the feedback generated based on the analysis results is a report including keyword frequencies and emotional tones.

[1597] (Claim 3)

[1598] 10. The system of claim 1, wherein the voice data is recorded with the senior's permission and is encrypted when transmitted.

[1599] "Example 1"

[1600] (Claim 1)

[1601] a terminal for recording audio data;

[1602] a terminal that encrypts the recorded voice data and transmits it to a server;

[1603] means for converting voice data into text at the server;

[1604] A method for analyzing text data using a generative AI model to extract keywords that indicate loneliness and stress;

[1605] A means for generating a feedback report based on the extracted keywords and emotional tone and sending it to medical professionals and caregivers;

[1606] a terminal that displays or notifies a message to encourage regular conversations with the elderly; ...

Claims

1. means for recording audio data; means for transmitting the recorded voice data to a server; means for converting voice data into text at the server; A means of analyzing text data and extracting keywords that indicate loneliness and stress; a means for generating and transmitting feedback based on the analysis results to healthcare professionals and caregivers; A means for displaying or notifying a message to encourage regular dialogue to the elderly person; A means of suggesting suitable social activities for older people; A means to evaluate the condition of the elderly person after the proposal and improve the accuracy of the system; A system including:

2. The system of claim 1 , wherein the feedback generated based on the analysis results is a report including keyword frequencies and emotional tones.

3. 2. The system of claim 1, wherein the voice data is recorded with the senior's permission and is encrypted when transmitted.

Images