System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving audio data; means for converting the audio data to text data; means for analyzing an emotional state from the text data and the audio data; means for generating a response using a generative model; means for outputting the generated response in audio form and text form; and means for connecting to an expert based on the emotional state.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 modern society, the number of people suffering from psychological anxiety and mental health problems is increasing. However, it remains difficult for these people to receive appropriate support, and access to mental health support is limited. Furthermore, the hassle and stigma associated with seeking advice make it difficult for people to casually seek help. Furthermore, while real-time responses are required, there is a serious shortage of counselors, making it difficult to provide prompt and appropriate support.

[0005] Therefore, there is a need for a system that can provide psychological support in a timely manner and allow users to consult with confidence.The purpose of this invention is to solve the above problems and provide professional and prompt support for users' mental health care. [Means for solving the problem]

[0006] The present invention is a system for providing mental care in real time, and includes the following means.

[0007] First, the system includes a means for receiving voice data, capturing the voice uttered by the user as digital data. Next, a means for converting the received voice data into text data is provided, and speech is converted into text using speech recognition technology. Then, a means for analyzing the emotional state from the text data and voice data is provided, and the user's emotional state is evaluated. Furthermore, a means for generating a response to the user using a generative model is incorporated, and an appropriate response is automatically generated based on the analyzed emotional state. This response is output in voice and text formats and provided to the user. The system also includes a means for presenting the generated response to the user in text and voice formats.

[0008] Additionally, we have added a means to connect with experts if an emotional state is deemed urgent, ensuring the safety and security of users by quickly connecting them to professional counselors or helplines as needed.

[0009] This allows users to easily receive mental care anytime and anywhere, which in turn makes it possible to detect and deal with psychological problems early.

[0010] "Voice data" refers to data in which the user's voice is recorded in digital format.

[0011] "Text data" refers to data obtained by converting voice data into character information.

[0012] "Emotional state" refers to the psychological state of the user analyzed from their speech and voice data.

[0013] A "generative model" is a machine learning model that generates appropriate responses based on input text data and emotional state.

[0014] "Experts" are counselors and helpline staff with specialized knowledge and experience in the field of mental health care.

[0015] A "response" is a linguistic answer, including advice and empathy suggestions, generated by a generative model in response to a user's statement.

[0016] A "user" is an individual who receives mental care using this system.

[0017] "Audio format" is a format in which text data is converted into reproducible audio using speech synthesis technology.

[0018] "Speech recognition" is a technology that listens to a user's voice data and converts it into text information.

[0019] "Emergency" refers to a situation in which the user's emotional state is so severe that immediate professional attention is required. [Brief explanation of the drawings]

[0020] [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

[0021] 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.

[0022] First, the terms used in the following description will be explained.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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."

[0041] This invention relates to a system that provides mental health care using generative AI. The main components are a server, a terminal, and a user, and the role of each will be specifically explained.

[0042] Server Processing

[0043] The server plays a central role in the system and operates as follows:

[0044] First, the server receives voice data from the user. This voice data is transmitted through the terminal. The server then converts the received voice data into text data using voice recognition technology. Specifically, the server uses a voice recognition library to analyze the voice and generate corresponding text.

[0045] The server then analyzes the user's emotional state using the text and audio data. It uses an emotion analysis algorithm to evaluate the user's psychological state based on the words and tone of the user's voice. Based on the analysis results, it determines whether the user's emotional state is urgent.

[0046] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and emotional state, which may include an empathetic message or advice tailored to the user's situation.

[0047] The server then generates the generated response in audio and text formats, using speech synthesis technology to convert the text to speech, and prepares both formats for transmission back to the user.

[0048] If the result of the emotion analysis indicates that the user's condition is urgent, the server will execute a procedure to connect to a specialist. If an emergency response is required, the server will immediately attempt to connect to a counselor or helpline based on the user's registered information. This allows the user to receive professional assistance quickly.

[0049] Terminal handling

[0050] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[0051] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[0052] User Experience

[0053] Users can talk to the device to ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this audio to the server. The server converts the audio into text and performs emotion analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[0054] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[0055] This allows users to receive prompt and professional mental health care, enabling early detection and treatment of psychological problems.

[0056] The processing flow will be explained below.

[0057] Server Processing

[0058] Step 1: Receive audio data

[0059] The server receives the voice data sent by the user as an HTTP request, which allows the user's voice input to reach the server.

[0060] Step 2: Convert the audio data to text

[0061] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice into text information.

[0062] Step 3: Analyze the emotional state

[0063] The server uses the text and audio data to run an emotion analysis algorithm to assess the user's emotional state, thereby determining the user's psychological state.

[0064] Step 4: Generate a response

[0065] The server uses the generative model to generate an appropriate response based on the analyzed emotional state and text data, thereby providing feedback to the user.

[0066] Step 5: Generate a response in audio and text format

[0067] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio formats.

[0068] Step 6: Connect with an expert

[0069] Based on the results of the sentiment analysis, the server will connect you to an expert if necessary, or if an emergency response is required, it will immediately connect you to a counselor or helpline.

[0070] Step 7: Send the response back to the user

[0071] The server generates responses and sends them back to the user in both text and audio format, allowing the user to receive these responses on their own device.

[0072] Terminal handling

[0073] Step 1: Capture the user's voice input

[0074] The user speaks to the device to discuss their concerns, and the device uses a microphone to capture this voice.

[0075] Step 2: Send the audio data to the server

[0076] The device sends the captured audio data to the server as an HTTP request, which allows the audio data to arrive at the server.

[0077] Step 3: Receive a response from the server

[0078] Once the response from the server is generated, the device receives the response, which may be in both text and audio format.

[0079] Step 4: Present the response to the user

[0080] The terminal displays the received response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[0081] User Experience

[0082] Step 1: Access the system

[0083] A user accesses the system using a terminal and starts a conversation for mental care.

[0084] Step 2: Use voice input

[0085] The user speaks into the device about their concerns and their condition. For example, they might say, "I've been feeling anxious lately and can't sleep."

[0086] Step 3: Receive a response

[0087] The user receives the response from the server in text and voice format through the terminal, thereby obtaining appropriate feedback on the content of the consultation.

[0088] Step 4: Connect with an expert

[0089] If the user's emotional state is deemed urgent, the device will connect them to a professional counselor or helpline, allowing them to receive the professional assistance they need.

[0090] Example 1

[0091] 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."

[0092] Mental health issues have become a serious problem in modern society, requiring prompt and appropriate responses. However, many people have limited access to mental health care, and access to specialists is difficult, especially in emergencies. Furthermore, conventional systems struggle to accurately analyze a user's emotional state and generate appropriate responses. Therefore, there is a need for a system that allows users to consult with a specialist with peace of mind and receive appropriate advice.

[0093] 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.

[0094] In this invention, the server includes means for receiving voice data using a voice input device, means for converting the voice data into text data using voice recognition technology, means for analyzing an emotional state from the text data and voice data using an emotion analysis algorithm, means for generating a response based on the text data and the emotional state using a generative AI model, means for generating and outputting the generated response in voice and text formats using voice synthesis technology, and means for automatically connecting to an expert if the emotional state satisfies certain conditions, thereby enabling the server to provide a prompt and appropriate response to a user's mental health issues and ensure connection to an expert in an emergency.

[0095] The "voice input device" is a device that a user uses to input voice data, and includes a microphone and the like.

[0096] "Speech recognition technology" is a technology that analyzes input voice data and converts it into corresponding text data.

[0097] An "emotion analysis algorithm" is an algorithm for analyzing text and audio data to assess a user's emotional state.

[0098] A "generative AI model" is a technology that uses a pre-trained artificial intelligence model to generate natural language responses based on input data.

[0099] "Speech synthesis technology" is a technology that converts text data into voice data and is used to provide a voice response to the user.

[0100] The "means for automatically connecting to an expert" is a means for instantly establishing a connection to an expert, counselor, helpline, etc. when the results of the user's emotion analysis satisfy certain conditions.

[0101] This invention relates to a system that uses generative AI to provide mental health care. The main components are a server, a terminal, and a user, and by working together, they provide users with prompt and appropriate mental health care.

[0102] Server Processing

[0103] The server plays a central role in the system and operates as follows: First, the server receives voice data from the user. This voice data is sent through the terminal, and the server uses voice recognition technology to convert the voice data into text data. For this conversion, a voice recognition library (for example, Google Cloud Speech-to-Text API) is used.

[0104] The server then runs a sentiment analysis algorithm using the text and audio data to analyze the user's emotional state, using services such as Microsoft Azure Text Analytics and AWS Comprehend, and determines whether the user's emotional state is urgent.

[0105] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the user's emotional state. This response may include empathetic messages or advice tailored to the user's situation. The generated response is generated in audio and text formats using speech synthesis technologies such as Amazon Polly or the Google Cloud Text-to-Speech API.

[0106] Finally, the server prepares to send the generated response back to the terminal. If the result of the sentiment analysis indicates that the user's condition is urgent, the server executes the procedure to connect to an expert, so that the user can receive professional assistance quickly.

[0107] Terminal handling

[0108] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[0109] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[0110] User Experience

[0111] Users can talk to the device and ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this voice data to the server. The server converts the voice into text and performs emotional analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[0112] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[0113] Specific examples

[0114] As an example of a specific prompt sentence, if a user says to the device, "I've been feeling anxious lately and can't sleep at night," the following prompt will be sent to the generation AI.

[0115] User input: "I've been feeling anxious lately and can't sleep at night."

[0116] Analyze the text for sentiment and generate an empathetic response:

[0117] This allows the server to generate an appropriate response and provide the user with fast, professional assistance.

[0118] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0119] Step 1:

[0120] The user provides voice input

[0121] Users can input their mental health concerns or questions into a device such as a smartphone or PC by voice. For example, they might say, "I've been feeling anxious lately and can't sleep at night." This is input as voice data.

[0122] Step 2:

[0123] The device sends the audio data to the server.

[0124] The device uses a microphone to capture the user's voice and saves it as digital audio data. It then sends this audio data to the server using an HTTP request. The input is the audio data, and the output is the digital audio data sent to the server. Specifically, the device generates a request like this:

[0125] Step 3:

[0126] The server converts the audio data into text data.

[0127] The server converts the received voice data into text format using speech recognition technology. This conversion is performed using a speech recognition library (for example, Google Cloud Speech-to-Text API). The input is voice data, and the output is text data. The server then passes the converted text data to the next processing step.

[0128] Step 4:

[0129] The server performs sentiment analysis

[0130] The server analyzes the emotional state using text and voice data. An emotion analysis algorithm (for example, Microsoft Azure Text Analytics or AWS Comprehend) is used here. The input is text and voice data, and the output is the result of the emotion analysis (emotional state). Based on this data, the server determines whether the user's emotional state is urgent.

[0131] Step 5:

[0132] The server generates a response using the generative AI model

[0133] The server uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the emotional state. The input is the text data and the results of the emotional analysis, and the output is the generated text response. Specifically, the prompt sentence is generated as follows:

[0134] User input: "I've been feeling anxious lately and can't sleep at night."

[0135] Analyze the text for sentiment and generate an empathetic response:”

[0136] Step 6:

[0137] The server generates a generated response in audio and text format.

[0138] The server converts the generated text response into audio format using speech synthesis technology (e.g., Amazon Polly or Google Cloud Text-to-Speech API). The input is the generated text response, and the output is the audio data and the text response.

[0139] Step 7:

[0140] The server sends the response to the device

[0141] The server sends the generated voice data and text response back to the device, using an HTTP response to send the data. The input is the generated voice data and text response, and the output is the response sent to the device.

[0142] Step 8:

[0143] The terminal presents the response to the user

[0144] The terminal analyzes the response received from the server, displays the text response on the screen, and plays the audio response. The input is the response data from the server, and the output is the audio and text response presented to the user, who receives it visually and audibly.

[0145] Step 9:

[0146] Emergency response

[0147] If the result of the sentiment analysis indicates that the user's condition is urgent, the server automatically connects the user to an expert, providing appropriate assistance based on the expert's contact information. The input is the result of the sentiment analysis and the user's registration information, and the output is the connection to the expert.

[0148] (Application example 1)

[0149] 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."

[0150] In conventional security services, potential risks and signs of trouble in the surrounding area are often overlooked through human attention alone. Furthermore, when an emergency occurs, a quick and appropriate response is required, but relying solely on human judgment carries risks. For this reason, there is a need for a system that can accurately analyze surrounding sounds and provide early warning of potential risks.

[0151] 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.

[0152] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing emotional states from the text data and voice data, means for capturing an unspecified number of voice data and performing emotional analysis to detect potential risks, and means for issuing an alarm when the potential risks are detected. This makes it possible to detect potential risks from surrounding voice data early and respond quickly.

[0153] "Speech data" refers to data that includes human speech and surrounding sounds acquired through a voice input device such as a microphone.

[0154] "Text data" refers to data that has been analyzed and converted into text information from voice data.

[0155] "Emotional state" refers to the psychological state or emotion that can be inferred from a person's speech or tone of voice.

[0156] A "generative model" is an algorithm or AI model that generates appropriate responses or predictions based on input data.

[0157] An "expert" is an individual or organization with advanced knowledge and skills in a particular field who provides assistance, especially in emergency situations.

[0158] A "potential risk" is a condition or phenomenon that has not yet manifested itself, but which has the potential to develop into a problem or trouble in the future.

[0159] An "alert" is a voice, text, or alarm that is issued to alert you when a particular condition is detected.

[0160] This invention is a system that uses generative AI to provide mental health care and security services. The main components are a server, a terminal, and a user, and the role of each will be explained in detail.

[0161] Server Processing

[0162] The server plays a central role in the system and operates as follows:

[0163] First, the server receives voice data from the user. This voice data is sent from the terminal via an HTTP request. The server converts the voice data into text data using a speech recognition library (e.g., speech_recognition library).

[0164] The server then analyzes the user's emotional state using the text and audio data. For emotion analysis, it applies the generative AI model's emotion analysis algorithm using Hugging Face's Transformer library, thereby assessing the user's psychological state.

[0165] The server then uses a generative AI model to generate an appropriate response based on the text data and the user's emotional state. This response may include an empathetic message or advice tailored to the user's situation. The generated response is then converted into audio format using speech synthesis technology (e.g., gTTS library) and sent back to the user along with the text format.

[0166] If the sentiment analysis determines that the user's condition is urgent, the server will implement measures to connect them to an expert, including using the system's pre-registered contact information for counselors and helplines to immediately connect them.

[0167] Terminal handling

[0168] The terminal supports the exchange of voice data between the user and the server. When the user speaks into the terminal, this voice is captured by the terminal. The terminal sends this voice data to the server as an HTTP request. When the terminal receives a response from the server, it displays this response in text format and plays it back in audio format. This allows the user to receive messages from the server both visually and audibly.

[0169] User Experience

[0170] Users can talk to their device to ask questions or ask about mental health or security. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this speech to the server. The server converts the speech data into text data and performs emotion analysis.

[0171] If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your life recently?" This response is returned to the user in voice and text format.

[0172] When used as a security service, it can detect potential risks from an unspecified number of voice data. For example, if a heated argument is occurring nearby, the server can analyze the voice data and issue an alert such as, "Caution: A potential risk has been detected. Please check your surroundings."

[0173] An example of a prompt sentence could be, "You are a virtual security guard. Please analyze the text data below and evaluate the emotional state. If there are a lot of negative emotions, please return a warning."

[0174] The system allows users to receive prompt and professional mental health care and security services that can detect potential risks early from surrounding sounds.

[0175] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0176] Step 1:

[0177] The user speaks into the terminal. For example, the user asks a question or asks for advice, such as, "I've been feeling anxious lately and can't sleep at night." The input is the user's voice data, which is received by the terminal. The voice data is obtained as output.

[0178] Step 2:

[0179] The terminal sends the received voice data to the server as an HTTP request. At this time, the terminal converts the voice data into an appropriate data format for sending to the server. The input is the user's voice data, and the voice data converted into the HTTP request format is sent to the server as the output.

[0180] Step 3:

[0181] The server receives the HTTP request and retrieves the audio data. It then converts the audio data into text using a speech recognition library (e.g., speech_recognition). The input is the audio data, and the output is the converted text data.

[0182] Step 4:

[0183] The server uses the converted text and audio data to perform emotion analysis. The model uses a generative AI model based on Hugging Face's Transformer library. The input is text and audio data, and the output is an emotional state.

[0184] Step 5:

[0185] The server uses a generative AI model based on the emotional state to generate an appropriate response. This generated response is based on the user's psychological state and the question content. The input is the emotional state and text data, and the output is a response text.

[0186] Step 6:

[0187] The server converts the generated response text into audio using speech synthesis technology (e.g., gTTS), and also prepares response data in text format. The input is the generated response text, and audio data and text data are generated as output.

[0188] Step 7:

[0189] The server returns the response data to the terminal as an HTTP response. The input is voice data and text data, and the HTTP response is sent to the terminal as output.

[0190] Step 8:

[0191] The terminal displays and plays back the response data received from the server. Specifically, it displays text data on the screen and plays audio data through the speaker. The input is response data in HTTP response format, and the output presents visual and auditory information to the user.

[0192] 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.

[0193] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[0194] Server Processing

[0195] The server plays a central role in the system and operates as follows:

[0196] First, the server receives the voice data sent by the user. After receiving the voice data, the server uses voice recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information.

[0197] The server then inputs the text and voice data into an emotion engine to perform a detailed analysis of the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to assess the user's psychological state. Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response.

[0198] The generated responses are provided to the user in both text and audio formats. The server utilizes speech synthesis technology to convert text responses into audio. It also has the means to automatically connect to experts in emergencies based on the results of sentiment analysis.

[0199] As a concrete example, suppose a user says, "I've been feeling anxious and can't sleep lately." In this case, the device captures the audio and sends it to the server. The server converts the audio into text and analyzes it using an emotion engine. If the emotion engine determines that the user is in a high state of anxiety, the generative AI model will generate an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?" Furthermore, if the emotional state is determined to be urgent, the server will immediately attempt to connect the user to a counselor or helpline.

[0200] Terminal handling

[0201] The device supports the exchange of voice data between the user and the server. When the user speaks to the device about their concerns, the device captures this voice and sends it to the server. When the device receives a response from the server, it provides this response to the user. Specifically, it displays the response in text format and plays it back in audio format at the same time. This allows the user to receive the response both visually and audibly.

[0202] User Experience

[0203] Users access the system through their device to receive mental health care. When the user speaks to the device about their condition and concerns, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[0204] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to get some rest. Why don't you try doing something to relax?" The response is provided to the user, and if necessary, an automatic connection to an expert is also made.

[0205] This system allows users to receive appropriate mental care anytime, anywhere, and enables early detection and treatment of psychological problems. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[0206] The processing flow will be explained below.

[0207] Server Processing

[0208] Step 1: Receive audio data

[0209] The server receives the voice data sent by the user through the terminal as an HTTP request, and the user's voice input is then incorporated into the system.

[0210] Step 2: Convert the audio data to text

[0211] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice data into a format that can be processed as text information.

[0212] Step 3: Analyze the emotional state using the emotion engine

[0213] The server inputs the converted text and voice data into an emotion engine for detailed analysis of the user's emotional state. The emotion engine evaluates the tone of the voice and the content of the text to determine the user's psychological state.

[0214] Step 4: Generate a response

[0215] The server uses a generative AI model based on the analyzed emotional state and text data to generate an appropriate response, creating feedback tailored to the user's desired assistance.

[0216] Step 5: Generate a response in audio and text format

[0217] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio.

[0218] Step 6: Connect with an expert

[0219] Based on the analysis results of the emotion engine, if the server determines that the user's emotional state is urgent, it will automatically connect the user to a counselor, helpline, or other expert, allowing the user to receive professional assistance quickly.

[0220] Step 7: Send the response back to the user

[0221] The server generates text and audio responses and sends them back to the user, who can then receive the necessary assistance at their terminal.

[0222] Terminal handling

[0223] Step 1: Capture the user's voice input

[0224] When the user speaks into the terminal about the content of the consultation, the terminal uses a microphone to capture the voice, and the user's voice is thus captured by the terminal.

[0225] Step 2: Send the audio data to the server

[0226] The device compiles the captured audio data into an HTTP request and sends it to the server. This process allows the audio data to reach the server.

[0227] Step 3: Receive a response from the server

[0228] The generated response is sent back from the server and received by the device.

[0229] Step 4: Present the response to the user

[0230] The terminal displays the received response in text format and plays it back in audio format, allowing the user to understand the response visually and audibly.

[0231] User Experience

[0232] Step 1: Access the system

[0233] A user accesses the system using a terminal and prepares to receive mental health care.

[0234] Step 2: Use voice input

[0235] The user verbally explains their condition and concerns to the device, for example, "I've been feeling a lot of stress at work lately."

[0236] Step 3: Receive a response

[0237] The user receives responses from the server via their device, which are provided in both text and audio formats, allowing the user to receive feedback in the format that is most convenient for them.

[0238] Step 4: Connect with an expert

[0239] If the emotion engine determines that the user's emotional state is urgent, the user will be automatically connected to a professional counselor or helpline, allowing them to receive direct, professional assistance.

[0240] This system allows users to receive appropriate mental health care in real time, enabling early detection and treatment of psychological problems. In addition, by incorporating an emotion engine, it is possible to grasp the user's emotional state in more detail and provide advanced support based on the results.

[0241] Example 2

[0242] 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."

[0243] Conventional mental health care systems face challenges in accurately analyzing users' emotional states and providing appropriate responses based on those results. They often lack the ability to quickly connect users to specialists in emergencies. Furthermore, they lack a way to provide both visual and auditory responses to improve the user experience.

[0244] 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.

[0245] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and the voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, and means for inputting a prompt sentence to the generative model based on the emotion analysis result. This enables the system to analyze the user's emotional state in detail and provide an appropriate response in real time. Furthermore, in an emergency, it can quickly connect to an expert, improving the quality of support provided to the user.

[0246] The "means for receiving voice data" refers to software and hardware for acquiring voice uttered by the user in digital form and transmitting it to the server.

[0247] "Means for converting voice data into text data" refers to software technology for converting received voice data into text information using voice recognition technology.

[0248] "Means for analyzing emotional states from text data and voice data" refers to an emotion analysis engine that analyzes text and voice data and evaluates the user's emotions and psychological state.

[0249] "Means for generating a response using a generative model" refers to a process that uses an artificial intelligence model to generate an appropriate response based on the results of sentiment analysis.

[0250] "Means for outputting the generated response in audio and text form" refers to speech synthesis and display technologies for converting the generated text response into audio and also displaying it in text form.

[0251] "Means for connecting to an expert based on emotional state" refers to a communication means for automatically connecting to an expert or counselor when the user's emotional state is determined to be urgent.

[0252] "Means for inputting prompt sentences to a generative model based on the results of sentiment analysis" refers to an input means for giving appropriate instructions to a generative model based on the results of sentiment analysis.

[0253] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[0254] Server Processing

[0255] The server plays a central role in the system and has a means of receiving voice data. For example, when a user says, "I've been feeling anxious and can't sleep lately," the voice data is sent from the device to the server. This is done using a microphone as hardware and a real-time voice capture tool as software.

[0256] Next, the server uses speech recognition technology (for example, Google Cloud Speech-to-Text) to convert the voice data into text data. This process generates the text, "I've been feeling anxious lately and can't sleep."

[0257] The server inputs the generated text and voice data into an emotion engine, which uses NVIDIA Clara Guardian and Microsoft Azure Video Indexer to perform a detailed analysis of the user's emotional state. This allows the detection of high anxiety levels.

[0258] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate response, such as, "That's tough. Have there been any changes in your living situation recently?"

[0259] The generated responses are provided to the user in audio and text formats using speech synthesis technologies (e.g., Amazon Polly, Google Cloud Text-to-Speech), with the option to connect with an expert if the emotional state is deemed urgent.

[0260] Terminal handling

[0261] The device supports the exchange of voice data between the user and the server. When the user speaks about their worries into the device, the device captures this voice and sends it to the server. For example, the device captures the user's voice saying, "I've been anxious and can't sleep lately," and sends it digitally to the server.

[0262] When the device receives a response from the server, it provides the response to the user. Specifically, it displays the response "That's tough. Have there been any changes in your living environment recently?" in text format and plays it back in audio format. A text display tool and an audio playback tool are used.

[0263] User Experience

[0264] Users access the system through their device and receive mental health care. When the user says something like, "I've been feeling stressed lately because of a lot of work," this speech is sent from the device to the server. The server converts the speech into text and analyzes it using an emotion engine. Based on the analysis results, the generative AI model generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?"

[0265] This response is provided to the user, and if necessary, an automatic connection to a specialist is also made. This allows users to receive appropriate mental care anytime, anywhere. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[0266] Prompt Sentence Examples

[0267] A specific example of an input prompt sentence for a generative AI model is as follows:

[0268] If a user says, "I've been feeling anxious and can't sleep lately," we convert this speech into text and analyze it with the emotion engine. We want to generate an empathetic response like this:

[0269] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0270] Step 1:

[0271] The user inputs voice into the device. For example, they might say, "I've been feeling anxious lately and can't sleep." This voice data is captured by the device's microphone. The input is the user's voice data, and the output is a digital audio file. Specifically, the device records the voice data and processes it in real time.

[0272] Step 2:

[0273] The device sends audio data to the server. In this process, a digital audio file is sent to the server via the Internet. The input is the audio data captured by the device, and the output is the audio data transferred to the server. Specifically, the device uses a data transmission protocol to upload the audio data to the server.

[0274] Step 3:

[0275] The server receives the audio data. The server stores the audio data in buffer memory and prepares it for the next processing step. The input is the audio data sent from the terminal, and the output is the server's buffer memory where the audio data is stored. Specifically, the server receives an HTTP request and stores the audio data.

[0276] Step 4:

[0277] The server uses speech recognition technology to convert the voice data into text data. For example, Google Cloud Speech-to-Text is used. In this process, the audio file is converted into text in a natural language. The input is the stored voice data, and the output is text data. Specifically, the server makes an API call to convert the voice data into text.

[0278] Step 5:

[0279] The server inputs text and audio data into the emotion engine to analyze the emotional state. NVIDIA Clara Guardian and Microsoft Azure Video Indexer are used. In this step, the emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's emotional state. The input is text and audio data, and the output is the analysis result of the emotional state. Specifically, the server calls the emotion analysis API and obtains the results.

[0280] Step 6:

[0281] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-4) based on the emotion analysis results. The input is the emotion analysis results, and the output is the generated response text. Specifically, the server inputs a prompt sentence into the generative AI model, causing it to generate an empathetic text response. An example of a prompt sentence is, "If a user says, 'I've been unable to sleep recently because I've been anxious,' we converted this speech into text and analyzed it with the emotion engine, which detected high levels of anxiety. We would like you to generate an empathetic response such as the following."

[0282] Step 7:

[0283] The server converts the generated text response into speech. Speech synthesis technology (e.g., Amazon Polly, Google Cloud Text-to-Speech) is used. The input is the generated text response, and the output is the speech response data. Specifically, the server calls an API that converts text data into speech data and generates an audio file.

[0284] Step 8:

[0285] The server sends the generated text and voice responses to the terminal. The input is the generated text and voice response data, and the output is the data sent to the terminal. In specific operations, the server uses a data transmission protocol to upload the text and voice data to the terminal.

[0286] Step 9:

[0287] The terminal receives the response from the server and provides it to the user. Specifically, it displays the generated response in text format and simultaneously plays it in audio format. The input is the text and audio response data sent from the server, and the output is the response presented to the user. Specifically, the terminal uses a text display tool and an audio playback tool to provide the response to the user visually and audibly.

[0288] Step 10:

[0289] In an emergency, the server uses a means to connect to an expert based on the user's emotional state. For example, if the user's emotional state is very unstable and requires an emergency, the server automatically attempts to connect to an expert. The input is the result of the judgment of the urgency of the emotional state, and the output is a connection to an expert. Specifically, the server uses a communication means to contact an expert or counselor.

[0290] This allows users to receive appropriate mental health care anytime, anywhere.

[0291] (Application example 2)

[0292] 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."

[0293] Employees' psychological state has a significant impact on their work, and sustained high levels of stress and anxiety can lead to security incidents and reduced work efficiency. Conventional systems have difficulty monitoring employees' psychological state in real time and detecting abnormalities early, often resulting in delayed implementation of appropriate security measures and responses. This invention aims to prevent security incidents and improve work efficiency by analyzing employees' emotional state using their voice data and immediately notifying the security team when an abnormality is detected, enabling a rapid response.

[0294] 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 receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, means for monitoring the emotional state and detecting an abnormal psychological state, and means for notifying a security team when the abnormal psychological state is detected. This makes it possible to monitor the psychological state of employees in real time and respond immediately when an abnormality is detected.

[0295] "Voice data" refers to data that is a digital recording of a user's spoken words or vocalizations.

[0296] "Text data" refers to data that includes character information converted from audio data.

[0297] "Emotional state" is information that represents the user's psychological state and feelings, and is analyzed from voice data and text data.

[0298] "Means for analyzing" is a technique or device for processing audio and text data and assessing the user's emotional state.

[0299] A "generative model" is a machine learning algorithm or artificial intelligence technique for generating responses based on an analyzed emotional state.

[0300] A "means for generating a response" is a technique or device that uses a generative model to create an appropriate response that corresponds to the user's emotional state.

[0301] "Output means" refers to a technique or device for providing the generated response in audio and text form.

[0302] "Means for connecting" refers to technology or devices that allow a user to connect with an expert or counselor when they are in an emergency psychological state.

[0303] "Monitoring means" refers to technology or devices that continuously observe the user's emotional state and detect abnormalities early on.

[0304] "Means for notifying" is a technology or device that sends an alert to the security team when an abnormal psychological state is detected.

[0305] A specific embodiment of the present invention will be described below. The following system analyzes emotional states using voice data, detects abnormal psychological states, and notifies the security team, thereby preventing security incidents and improving work efficiency.

[0306] Server Processing

[0307] The server plays a central role in this system. First, it receives the voice data sent by the user. Once the voice data is received, the server uses speech recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information. For speech recognition, Google Cloud Speech-to-Text API and Amazon Transcribe are used.

[0308] The server then inputs the text and voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's psychological state. This analysis is performed using IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[0309] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response. The generated response is provided to the user in both text and audio format. Response generation uses natural language generation modules such as OpenAI GPT-4.

[0310] Furthermore, if the emotional state is deemed abnormal, the server has the means to immediately notify the security team, which can be done via email or a chat system (e.g., Slack or Microsoft Teams).

[0311] Terminal handling

[0312] The device supports the exchange of voice data between the user and the server. When the user speaks a report into the device, the device captures this voice and sends it to the server. Specifically, this capture is performed using the smartphone's microphone and an appropriate voice capture module (iOS, Android).

[0313] Upon receiving a response from the server, the terminal provides the response to the user by displaying it in text format and playing it back in audio format, allowing the user to receive the response both visually and audibly.

[0314] User Experience

[0315] Users access the system through their devices and provide feedback on their emotional state. When the user speaks about their state into the device, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[0316] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?" If the emotional state is determined to be abnormal, the security team is immediately notified.

[0317] Example prompt sentence:

[0318] A user says, "I'm feeling very stressed because my project deadline is approaching."

[0319] An example prompt from a generative AI model: "The user's stress level is high. Please select the appropriate response."

[0320] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0321] Step 1:

[0322] Capture and transmit audio data

[0323] The user speaks into the smartphone terminal about their state of mind and emotions. The terminal captures this voice and stores it as digital audio data. The terminal then encrypts this audio data and transmits it to the server using a secure communication method (e.g., HTTPS).

[0324] Input: User's voice

[0325] Output: Encrypted digital audio data

[0326] Step 2:

[0327] Converting audio data to text

[0328] The server receives the voice data sent from the device and converts it into text using speech recognition technology. Specifically, it uses the Google Cloud Speech-to-Text API or Amazon Transcribe to convert the voice data into text.

[0329] Input: Encrypted digital audio data

[0330] Output: Text data

[0331] Step 3:

[0332] Emotional state analysis

[0333] The server inputs the obtained text data and the original voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the voice tone and text content to evaluate the user's psychological state. This analysis uses IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[0334] Input: Text and audio data

[0335] Output: Parsed emotional state

[0336] Step 4:

[0337] Generating a response

[0338] The server uses a generative AI model based on the analyzed emotional state to generate an appropriate response, which is then prepared for delivery in both text and audio formats, using natural language generation modules such as OpenAI GPT-4.

[0339] Input: Parsed emotional state

[0340] Output: Text and audio responses

[0341] Step 5:

[0342] Emergency notifications and connection to experts

[0343] If the emotional state is deemed abnormal or urgent, the server immediately notifies the security team via email or chat systems (e.g., Slack, Microsoft Teams), and automatically connects the individual to a specialist or counselor if necessary.

[0344] Input: Parsed emotional state and generated response

[0345] Output: Notify security team and connect to experts

[0346] Step 6:

[0347] Providing a response to the user

[0348] The server generates a response and sends it back to the terminal, which displays the response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[0349] Input: Text and audio responses

[0350] Output: Feedback to the user (text display and audio playback)

[0351] 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.

[0352] 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.

[0353] 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.

[0354] [Second embodiment]

[0355] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0356] 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.

[0357] 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).

[0358] 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.

[0359] 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.

[0360] 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).

[0361] 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.

[0362] 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.

[0363] 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.

[0364] 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.

[0365] 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.

[0366] 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."

[0367] This invention relates to a system that provides mental health care using generative AI. The main components are a server, a terminal, and a user, and the role of each will be specifically explained.

[0368] Server Processing

[0369] The server plays a central role in the system and operates as follows:

[0370] First, the server receives voice data from the user. This voice data is transmitted through the terminal. The server then converts the received voice data into text data using voice recognition technology. Specifically, the server uses a voice recognition library to analyze the voice and generate corresponding text.

[0371] The server then analyzes the user's emotional state using the text and audio data. It uses an emotion analysis algorithm to evaluate the user's psychological state based on the words and tone of the user's voice. Based on the analysis results, it determines whether the user's emotional state is urgent.

[0372] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and emotional state, which may include an empathetic message or advice tailored to the user's situation.

[0373] The server then generates the generated response in audio and text formats, using speech synthesis technology to convert the text to speech, and prepares both formats for transmission back to the user.

[0374] If the result of the emotion analysis indicates that the user's condition is urgent, the server will execute a procedure to connect to a specialist. If an emergency response is required, the server will immediately attempt to connect to a counselor or helpline based on the user's registered information. This allows the user to receive professional assistance quickly.

[0375] Terminal handling

[0376] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[0377] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[0378] User Experience

[0379] Users can talk to the device to ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this audio to the server. The server converts the audio into text and performs emotion analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[0380] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[0381] This allows users to receive prompt and professional mental health care, enabling early detection and treatment of psychological problems.

[0382] The processing flow will be explained below.

[0383] Server Processing

[0384] Step 1: Receive audio data

[0385] The server receives the voice data sent by the user as an HTTP request, which allows the user's voice input to reach the server.

[0386] Step 2: Convert the audio data to text

[0387] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice into text information.

[0388] Step 3: Analyze the emotional state

[0389] The server uses the text and audio data to run an emotion analysis algorithm to assess the user's emotional state, thereby determining the user's psychological state.

[0390] Step 4: Generate a response

[0391] The server uses the generative model to generate an appropriate response based on the analyzed emotional state and text data, thereby providing feedback to the user.

[0392] Step 5: Generate a response in audio and text format

[0393] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio formats.

[0394] Step 6: Connect with an expert

[0395] Based on the results of the sentiment analysis, the server will connect you to an expert if necessary, or if an emergency response is required, it will immediately connect you to a counselor or helpline.

[0396] Step 7: Send the response back to the user

[0397] The server generates responses and sends them back to the user in both text and audio format, allowing the user to receive these responses on their own device.

[0398] Terminal handling

[0399] Step 1: Capture the user's voice input

[0400] The user speaks to the device to discuss their concerns, and the device uses a microphone to capture this voice.

[0401] Step 2: Send the audio data to the server

[0402] The device sends the captured audio data to the server as an HTTP request, which allows the audio data to arrive at the server.

[0403] Step 3: Receive a response from the server

[0404] Once the response from the server is generated, the device receives the response, which may be in both text and audio format.

[0405] Step 4: Present the response to the user

[0406] The terminal displays the received response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[0407] User Experience

[0408] Step 1: Access the system

[0409] A user accesses the system using a terminal and starts a conversation for mental care.

[0410] Step 2: Use voice input

[0411] The user speaks into the device about their concerns and their condition. For example, they might say, "I've been feeling anxious lately and can't sleep."

[0412] Step 3: Receive a response

[0413] The user receives the response from the server in text and voice format through the terminal, thereby obtaining appropriate feedback on the content of the consultation.

[0414] Step 4: Connect with an expert

[0415] If the user's emotional state is deemed urgent, the device will connect them to a professional counselor or helpline, allowing them to receive the professional assistance they need.

[0416] Example 1

[0417] 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."

[0418] Mental health issues have become a serious problem in modern society, requiring prompt and appropriate responses. However, many people have limited access to mental health care, and access to specialists is difficult, especially in emergencies. Furthermore, conventional systems struggle to accurately analyze a user's emotional state and generate appropriate responses. Therefore, there is a need for a system that allows users to consult with a specialist with peace of mind and receive appropriate advice.

[0419] 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.

[0420] In this invention, the server includes means for receiving voice data using a voice input device, means for converting the voice data into text data using voice recognition technology, means for analyzing an emotional state from the text data and voice data using an emotion analysis algorithm, means for generating a response based on the text data and the emotional state using a generative AI model, means for generating and outputting the generated response in voice and text formats using voice synthesis technology, and means for automatically connecting to an expert if the emotional state satisfies certain conditions, thereby enabling the server to provide a prompt and appropriate response to a user's mental health issues and ensure connection to an expert in an emergency.

[0421] The "voice input device" is a device that a user uses to input voice data, and includes a microphone and the like.

[0422] "Speech recognition technology" is a technology that analyzes input voice data and converts it into corresponding text data.

[0423] An "emotion analysis algorithm" is an algorithm for analyzing text and audio data to assess a user's emotional state.

[0424] A "generative AI model" is a technology that uses a pre-trained artificial intelligence model to generate natural language responses based on input data.

[0425] "Speech synthesis technology" is a technology that converts text data into voice data and is used to provide a voice response to the user.

[0426] The "means for automatically connecting to an expert" is a means for instantly establishing a connection to an expert, counselor, helpline, etc. when the results of the user's emotion analysis satisfy certain conditions.

[0427] This invention relates to a system that uses generative AI to provide mental health care. The main components are a server, a terminal, and a user, and by working together, they provide users with prompt and appropriate mental health care.

[0428] Server Processing

[0429] The server plays a central role in the system and operates as follows: First, the server receives voice data from the user. This voice data is sent through the terminal, and the server uses voice recognition technology to convert the voice data into text data. For this conversion, a voice recognition library (for example, Google Cloud Speech-to-Text API) is used.

[0430] The server then runs a sentiment analysis algorithm using the text and audio data to analyze the user's emotional state, using services such as Microsoft Azure Text Analytics and AWS Comprehend, and determines whether the user's emotional state is urgent.

[0431] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the user's emotional state. This response may include empathetic messages or advice tailored to the user's situation. The generated response is generated in audio and text formats using speech synthesis technologies such as Amazon Polly or the Google Cloud Text-to-Speech API.

[0432] Finally, the server prepares to send the generated response back to the terminal. If the result of the sentiment analysis indicates that the user's condition is urgent, the server executes the procedure to connect to an expert, so that the user can receive professional assistance quickly.

[0433] Terminal handling

[0434] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[0435] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[0436] User Experience

[0437] Users can talk to the device and ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this voice data to the server. The server converts the voice into text and performs emotional analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[0438] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[0439] Specific examples

[0440] As an example of a specific prompt sentence, if a user says to the device, "I've been feeling anxious lately and can't sleep at night," the following prompt will be sent to the generation AI.

[0441] User input: "I've been feeling anxious lately and can't sleep at night."

[0442] Analyze the text for sentiment and generate an empathetic response:

[0443] This allows the server to generate an appropriate response and provide the user with fast, professional assistance.

[0444] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0445] Step 1:

[0446] The user provides voice input

[0447] Users can input their mental health concerns or questions into a device such as a smartphone or PC by voice. For example, they might say, "I've been feeling anxious lately and can't sleep at night." This is input as voice data.

[0448] Step 2:

[0449] The device sends the audio data to the server.

[0450] The device uses a microphone to capture the user's voice and saves it as digital audio data. It then sends this audio data to the server using an HTTP request. The input is the audio data, and the output is the digital audio data sent to the server. Specifically, the device generates a request like this:

[0451] Step 3:

[0452] The server converts the audio data into text data.

[0453] The server converts the received voice data into text format using speech recognition technology. This conversion is performed using a speech recognition library (for example, Google Cloud Speech-to-Text API). The input is voice data, and the output is text data. The server then passes the converted text data to the next processing step.

[0454] Step 4:

[0455] The server performs sentiment analysis

[0456] The server analyzes the emotional state using text and voice data. An emotion analysis algorithm (for example, Microsoft Azure Text Analytics or AWS Comprehend) is used here. The input is text and voice data, and the output is the result of the emotion analysis (emotional state). Based on this data, the server determines whether the user's emotional state is urgent.

[0457] Step 5:

[0458] The server generates a response using the generative AI model

[0459] The server uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the emotional state. The input is the text data and the results of the emotional analysis, and the output is the generated text response. Specifically, the prompt sentence is generated as follows:

[0460] User input: "I've been feeling anxious lately and can't sleep at night."

[0461] Analyze the text for sentiment and generate an empathetic response:”

[0462] Step 6:

[0463] The server generates a generated response in audio and text format.

[0464] The server converts the generated text response into audio format using speech synthesis technology (e.g., Amazon Polly or Google Cloud Text-to-Speech API). The input is the generated text response, and the output is the audio data and the text response.

[0465] Step 7:

[0466] The server sends the response to the device

[0467] The server sends the generated voice data and text response back to the device, using an HTTP response to send the data. The input is the generated voice data and text response, and the output is the response sent to the device.

[0468] Step 8:

[0469] The terminal presents the response to the user

[0470] The terminal analyzes the response received from the server, displays the text response on the screen, and plays the audio response. The input is the response data from the server, and the output is the audio and text response presented to the user, who receives it visually and audibly.

[0471] Step 9:

[0472] Emergency response

[0473] If the result of the sentiment analysis indicates that the user's condition is urgent, the server automatically connects the user to an expert, providing appropriate assistance based on the expert's contact information. The input is the result of the sentiment analysis and the user's registration information, and the output is the connection to the expert.

[0474] (Application example 1)

[0475] 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."

[0476] In conventional security services, potential risks and signs of trouble in the surrounding area are often overlooked through human attention alone. Furthermore, when an emergency occurs, a quick and appropriate response is required, but relying solely on human judgment carries risks. For this reason, there is a need for a system that can accurately analyze surrounding sounds and provide early warning of potential risks.

[0477] 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.

[0478] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing emotional states from the text data and voice data, means for capturing an unspecified number of voice data and performing emotional analysis to detect potential risks, and means for issuing an alarm when the potential risks are detected. This makes it possible to detect potential risks from surrounding voice data early and respond quickly.

[0479] "Speech data" refers to data that includes human speech and surrounding sounds acquired through a voice input device such as a microphone.

[0480] "Text data" refers to data that has been analyzed and converted into text information from voice data.

[0481] "Emotional state" refers to the psychological state or emotion that can be inferred from a person's speech or tone of voice.

[0482] A "generative model" is an algorithm or AI model that generates appropriate responses or predictions based on input data.

[0483] An "expert" is an individual or organization with advanced knowledge and skills in a particular field who provides assistance, especially in emergency situations.

[0484] A "potential risk" is a condition or phenomenon that has not yet manifested itself, but which has the potential to develop into a problem or trouble in the future.

[0485] An "alert" is a voice, text, or alarm that is issued to alert you when a particular condition is detected.

[0486] This invention is a system that uses generative AI to provide mental health care and security services. The main components are a server, a terminal, and a user, and the role of each will be explained in detail.

[0487] Server Processing

[0488] The server plays a central role in the system and operates as follows:

[0489] First, the server receives voice data from the user. This voice data is sent from the terminal via an HTTP request. The server converts the voice data into text data using a speech recognition library (e.g., speech_recognition library).

[0490] The server then analyzes the user's emotional state using the text and audio data. For emotion analysis, it applies the generative AI model's emotion analysis algorithm using Hugging Face's Transformer library, thereby assessing the user's psychological state.

[0491] The server then uses a generative AI model to generate an appropriate response based on the text data and the user's emotional state. This response may include an empathetic message or advice tailored to the user's situation. The generated response is then converted into audio format using speech synthesis technology (e.g., gTTS library) and sent back to the user along with the text format.

[0492] If the sentiment analysis determines that the user's condition is urgent, the server will implement measures to connect them to an expert, including using the system's pre-registered contact information for counselors and helplines to immediately connect them.

[0493] Terminal handling

[0494] The terminal supports the exchange of voice data between the user and the server. When the user speaks into the terminal, this voice is captured by the terminal. The terminal sends this voice data to the server as an HTTP request. When the terminal receives a response from the server, it displays this response in text format and plays it back in audio format. This allows the user to receive messages from the server both visually and audibly.

[0495] User Experience

[0496] Users can talk to their device to ask questions or ask about mental health or security. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this speech to the server. The server converts the speech data into text data and performs emotion analysis.

[0497] If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your life recently?" This response is returned to the user in voice and text format.

[0498] When used as a security service, it can detect potential risks from an unspecified number of voice data. For example, if a heated argument is occurring nearby, the server can analyze the voice data and issue an alert such as, "Caution: A potential risk has been detected. Please check your surroundings."

[0499] An example of a prompt sentence could be, "You are a virtual security guard. Please analyze the text data below and evaluate the emotional state. If there are a lot of negative emotions, please return a warning."

[0500] The system allows users to receive prompt and professional mental health care and security services that can detect potential risks early from surrounding sounds.

[0501] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0502] Step 1:

[0503] The user speaks into the terminal. For example, the user asks a question or asks for advice, such as, "I've been feeling anxious lately and can't sleep at night." The input is the user's voice data, which is received by the terminal. The voice data is obtained as output.

[0504] Step 2:

[0505] The terminal sends the received voice data to the server as an HTTP request. At this time, the terminal converts the voice data into an appropriate data format for sending to the server. The input is the user's voice data, and the voice data converted into the HTTP request format is sent to the server as the output.

[0506] Step 3:

[0507] The server receives the HTTP request and retrieves the audio data. It then converts the audio data into text using a speech recognition library (e.g., speech_recognition). The input is the audio data, and the output is the converted text data.

[0508] Step 4:

[0509] The server uses the converted text and audio data to perform emotion analysis. The model uses a generative AI model based on Hugging Face's Transformer library. The input is text and audio data, and the output is an emotional state.

[0510] Step 5:

[0511] The server uses a generative AI model based on the emotional state to generate an appropriate response. This generated response is based on the user's psychological state and the question content. The input is the emotional state and text data, and the output is a response text.

[0512] Step 6:

[0513] The server converts the generated response text into audio using speech synthesis technology (e.g., gTTS), and also prepares response data in text format. The input is the generated response text, and audio data and text data are generated as output.

[0514] Step 7:

[0515] The server returns the response data to the terminal as an HTTP response. The input is voice data and text data, and the HTTP response is sent to the terminal as output.

[0516] Step 8:

[0517] The terminal displays and plays back the response data received from the server. Specifically, it displays text data on the screen and plays audio data through the speaker. The input is response data in HTTP response format, and the output presents visual and auditory information to the user.

[0518] 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.

[0519] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[0520] Server Processing

[0521] The server plays a central role in the system and operates as follows:

[0522] First, the server receives the voice data sent by the user. After receiving the voice data, the server uses voice recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information.

[0523] The server then inputs the text and voice data into an emotion engine to perform a detailed analysis of the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to assess the user's psychological state. Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response.

[0524] The generated responses are provided to the user in both text and audio formats. The server utilizes speech synthesis technology to convert text responses into audio. It also has the means to automatically connect to experts in emergencies based on the results of sentiment analysis.

[0525] As a concrete example, suppose a user says, "I've been feeling anxious and can't sleep lately." In this case, the device captures the audio and sends it to the server. The server converts the audio into text and analyzes it using an emotion engine. If the emotion engine determines that the user is in a high state of anxiety, the generative AI model will generate an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?" Furthermore, if the emotional state is determined to be urgent, the server will immediately attempt to connect the user to a counselor or helpline.

[0526] Terminal handling

[0527] The device supports the exchange of voice data between the user and the server. When the user speaks to the device about their concerns, the device captures this voice and sends it to the server. When the device receives a response from the server, it provides this response to the user. Specifically, it displays the response in text format and plays it back in audio format at the same time. This allows the user to receive the response both visually and audibly.

[0528] User Experience

[0529] Users access the system through their device to receive mental health care. When the user speaks to the device about their condition and concerns, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[0530] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to get some rest. Why don't you try doing something to relax?" The response is provided to the user, and if necessary, an automatic connection to an expert is also made.

[0531] This system allows users to receive appropriate mental care anytime, anywhere, and enables early detection and treatment of psychological problems. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[0532] The processing flow will be explained below.

[0533] Server Processing

[0534] Step 1: Receive audio data

[0535] The server receives the voice data sent by the user through the terminal as an HTTP request, and the user's voice input is then incorporated into the system.

[0536] Step 2: Convert the audio data to text

[0537] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice data into a format that can be processed as text information.

[0538] Step 3: Analyze the emotional state using the emotion engine

[0539] The server inputs the converted text and voice data into an emotion engine for detailed analysis of the user's emotional state. The emotion engine evaluates the tone of the voice and the content of the text to determine the user's psychological state.

[0540] Step 4: Generate a response

[0541] The server uses a generative AI model based on the analyzed emotional state and text data to generate an appropriate response, creating feedback tailored to the user's desired assistance.

[0542] Step 5: Generate a response in audio and text format

[0543] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio.

[0544] Step 6: Connect with an expert

[0545] Based on the analysis results of the emotion engine, if the server determines that the user's emotional state is urgent, it will automatically connect the user to a counselor, helpline, or other expert, allowing the user to receive professional assistance quickly.

[0546] Step 7: Send the response back to the user

[0547] The server generates text and audio responses and sends them back to the user, who can then receive the necessary assistance at their terminal.

[0548] Terminal handling

[0549] Step 1: Capture the user's voice input

[0550] When the user speaks into the terminal about the content of the consultation, the terminal uses a microphone to capture the voice, and the user's voice is thus captured by the terminal.

[0551] Step 2: Send the audio data to the server

[0552] The device compiles the captured audio data into an HTTP request and sends it to the server. This process allows the audio data to reach the server.

[0553] Step 3: Receive a response from the server

[0554] The generated response is sent back from the server and received by the device.

[0555] Step 4: Present the response to the user

[0556] The terminal displays the received response in text format and plays it back in audio format, allowing the user to understand the response visually and audibly.

[0557] User Experience

[0558] Step 1: Access the system

[0559] A user accesses the system using a terminal and prepares to receive mental health care.

[0560] Step 2: Use voice input

[0561] The user verbally explains their condition and concerns to the device, for example, "I've been feeling a lot of stress at work lately."

[0562] Step 3: Receive a response

[0563] The user receives responses from the server via their device, which are provided in both text and audio formats, allowing the user to receive feedback in the format that is most convenient for them.

[0564] Step 4: Connect with an expert

[0565] If the emotion engine determines that the user's emotional state is urgent, the user will be automatically connected to a professional counselor or helpline, allowing them to receive direct, professional assistance.

[0566] This system allows users to receive appropriate mental health care in real time, enabling early detection and treatment of psychological problems. In addition, by incorporating an emotion engine, it is possible to grasp the user's emotional state in more detail and provide advanced support based on the results.

[0567] Example 2

[0568] 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."

[0569] Conventional mental health care systems face challenges in accurately analyzing users' emotional states and providing appropriate responses based on those results. They often lack the ability to quickly connect users to specialists in emergencies. Furthermore, they lack a way to provide both visual and auditory responses to improve the user experience.

[0570] 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.

[0571] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and the voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, and means for inputting a prompt sentence to the generative model based on the emotion analysis result. This enables the system to analyze the user's emotional state in detail and provide an appropriate response in real time. Furthermore, in an emergency, it can quickly connect to an expert, improving the quality of support provided to the user.

[0572] The "means for receiving voice data" refers to software and hardware for acquiring voice uttered by the user in digital form and transmitting it to the server.

[0573] "Means for converting voice data into text data" refers to software technology for converting received voice data into text information using voice recognition technology.

[0574] "Means for analyzing emotional states from text data and voice data" refers to an emotion analysis engine that analyzes text and voice data and evaluates the user's emotions and psychological state.

[0575] "Means for generating a response using a generative model" refers to a process that uses an artificial intelligence model to generate an appropriate response based on the results of sentiment analysis.

[0576] "Means for outputting the generated response in audio and text form" refers to speech synthesis and display technologies for converting the generated text response into audio and also displaying it in text form.

[0577] "Means for connecting to an expert based on emotional state" refers to a communication means for automatically connecting to an expert or counselor when the user's emotional state is determined to be urgent.

[0578] "Means for inputting prompt sentences to a generative model based on the results of sentiment analysis" refers to an input means for giving appropriate instructions to a generative model based on the results of sentiment analysis.

[0579] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[0580] Server Processing

[0581] The server plays a central role in the system and has a means of receiving voice data. For example, when a user says, "I've been feeling anxious and can't sleep lately," the voice data is sent from the device to the server. This is done using a microphone as hardware and a real-time voice capture tool as software.

[0582] Next, the server uses speech recognition technology (for example, Google Cloud Speech-to-Text) to convert the voice data into text data. This process generates the text, "I've been feeling anxious lately and can't sleep."

[0583] The server inputs the generated text and voice data into an emotion engine, which uses NVIDIA Clara Guardian and Microsoft Azure Video Indexer to perform a detailed analysis of the user's emotional state. This allows the detection of high anxiety levels.

[0584] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate response, such as, "That's tough. Have there been any changes in your living situation recently?"

[0585] The generated responses are provided to the user in audio and text formats using speech synthesis technologies (e.g., Amazon Polly, Google Cloud Text-to-Speech), with the option to connect with an expert if the emotional state is deemed urgent.

[0586] Terminal handling

[0587] The device supports the exchange of voice data between the user and the server. When the user speaks about their worries into the device, the device captures this voice and sends it to the server. For example, the device captures the user's voice saying, "I've been anxious and can't sleep lately," and sends it digitally to the server.

[0588] When the device receives a response from the server, it provides the response to the user. Specifically, it displays the response "That's tough. Have there been any changes in your living environment recently?" in text format and plays it back in audio format. A text display tool and an audio playback tool are used.

[0589] User Experience

[0590] Users access the system through their device and receive mental health care. When the user says something like, "I've been feeling stressed lately because of a lot of work," this speech is sent from the device to the server. The server converts the speech into text and analyzes it using an emotion engine. Based on the analysis results, the generative AI model generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?"

[0591] This response is provided to the user, and if necessary, an automatic connection to a specialist is also made. This allows users to receive appropriate mental care anytime, anywhere. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[0592] Prompt Sentence Examples

[0593] A specific example of an input prompt sentence for a generative AI model is as follows:

[0594] If a user says, "I've been feeling anxious and can't sleep lately," we convert this speech into text and analyze it with the emotion engine. We want to generate an empathetic response like this:

[0595] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0596] Step 1:

[0597] The user inputs voice into the device. For example, they might say, "I've been feeling anxious lately and can't sleep." This voice data is captured by the device's microphone. The input is the user's voice data, and the output is a digital audio file. Specifically, the device records the voice data and processes it in real time.

[0598] Step 2:

[0599] The device sends audio data to the server. In this process, a digital audio file is sent to the server via the Internet. The input is the audio data captured by the device, and the output is the audio data transferred to the server. Specifically, the device uses a data transmission protocol to upload the audio data to the server.

[0600] Step 3:

[0601] The server receives the audio data. The server stores the audio data in buffer memory and prepares it for the next processing step. The input is the audio data sent from the terminal, and the output is the server's buffer memory where the audio data is stored. Specifically, the server receives an HTTP request and stores the audio data.

[0602] Step 4:

[0603] The server uses speech recognition technology to convert the voice data into text data. For example, Google Cloud Speech-to-Text is used. In this process, the audio file is converted into text in a natural language. The input is the stored voice data, and the output is text data. Specifically, the server makes an API call to convert the voice data into text.

[0604] Step 5:

[0605] The server inputs text and audio data into the emotion engine to analyze the emotional state. NVIDIA Clara Guardian and Microsoft Azure Video Indexer are used. In this step, the emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's emotional state. The input is text and audio data, and the output is the analysis result of the emotional state. Specifically, the server calls the emotion analysis API and obtains the results.

[0606] Step 6:

[0607] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-4) based on the emotion analysis results. The input is the emotion analysis results, and the output is the generated response text. Specifically, the server inputs a prompt sentence into the generative AI model, causing it to generate an empathetic text response. An example of a prompt sentence is, "If a user says, 'I've been unable to sleep recently because I've been anxious,' we converted this speech into text and analyzed it with the emotion engine, which detected high levels of anxiety. We would like you to generate an empathetic response such as the following."

[0608] Step 7:

[0609] The server converts the generated text response into speech. Speech synthesis technology (e.g., Amazon Polly, Google Cloud Text-to-Speech) is used. The input is the generated text response, and the output is the speech response data. Specifically, the server calls an API that converts text data into speech data and generates an audio file.

[0610] Step 8:

[0611] The server sends the generated text and voice responses to the terminal. The input is the generated text and voice response data, and the output is the data sent to the terminal. In specific operations, the server uses a data transmission protocol to upload the text and voice data to the terminal.

[0612] Step 9:

[0613] The terminal receives the response from the server and provides it to the user. Specifically, it displays the generated response in text format and simultaneously plays it in audio format. The input is the text and audio response data sent from the server, and the output is the response presented to the user. Specifically, the terminal uses a text display tool and an audio playback tool to provide the response to the user visually and audibly.

[0614] Step 10:

[0615] In an emergency, the server uses a means to connect to an expert based on the user's emotional state. For example, if the user's emotional state is very unstable and requires an emergency, the server automatically attempts to connect to an expert. The input is the result of the judgment of the urgency of the emotional state, and the output is a connection to an expert. Specifically, the server uses a communication means to contact an expert or counselor.

[0616] This allows users to receive appropriate mental health care anytime, anywhere.

[0617] (Application example 2)

[0618] 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."

[0619] Employees' psychological state has a significant impact on their work, and sustained high levels of stress and anxiety can lead to security incidents and reduced work efficiency. Conventional systems have difficulty monitoring employees' psychological state in real time and detecting abnormalities early, often resulting in delayed implementation of appropriate security measures and responses. This invention aims to prevent security incidents and improve work efficiency by analyzing employees' emotional state using their voice data and immediately notifying the security team when an abnormality is detected, enabling a rapid response.

[0620] 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 receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, means for monitoring the emotional state and detecting an abnormal psychological state, and means for notifying a security team when the abnormal psychological state is detected. This makes it possible to monitor the psychological state of employees in real time and respond immediately when an abnormality is detected.

[0621] "Voice data" refers to data that is a digital recording of a user's spoken words or vocalizations.

[0622] "Text data" refers to data that includes character information converted from audio data.

[0623] "Emotional state" is information that represents the user's psychological state and feelings, and is analyzed from voice data and text data.

[0624] "Means for analyzing" is a technique or device for processing audio and text data and assessing the user's emotional state.

[0625] A "generative model" is a machine learning algorithm or artificial intelligence technique for generating responses based on an analyzed emotional state.

[0626] A "means for generating a response" is a technique or device that uses a generative model to create an appropriate response that corresponds to the user's emotional state.

[0627] "Output means" refers to a technique or device for providing the generated response in audio and text form.

[0628] "Means for connecting" refers to technology or devices that allow a user to connect with an expert or counselor when they are in an emergency psychological state.

[0629] "Monitoring means" refers to technology or devices that continuously observe the user's emotional state and detect abnormalities early on.

[0630] "Means for notifying" is a technology or device that sends an alert to the security team when an abnormal psychological state is detected.

[0631] A specific embodiment of the present invention will be described below. The following system analyzes emotional states using voice data, detects abnormal psychological states, and notifies the security team, thereby preventing security incidents and improving work efficiency.

[0632] Server Processing

[0633] The server plays a central role in this system. First, it receives the voice data sent by the user. Once the voice data is received, the server uses speech recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information. For speech recognition, Google Cloud Speech-to-Text API and Amazon Transcribe are used.

[0634] The server then inputs the text and voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's psychological state. This analysis is performed using IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[0635] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response. The generated response is provided to the user in both text and audio format. Response generation uses natural language generation modules such as OpenAI GPT-4.

[0636] Furthermore, if the emotional state is deemed abnormal, the server has the means to immediately notify the security team, which can be done via email or a chat system (e.g., Slack or Microsoft Teams).

[0637] Terminal handling

[0638] The device supports the exchange of voice data between the user and the server. When the user speaks a report into the device, the device captures this voice and sends it to the server. Specifically, this capture is performed using the smartphone's microphone and an appropriate voice capture module (iOS, Android).

[0639] Upon receiving a response from the server, the terminal provides the response to the user by displaying it in text format and playing it back in audio format, allowing the user to receive the response both visually and audibly.

[0640] User Experience

[0641] Users access the system through their devices and provide feedback on their emotional state. When the user speaks about their state into the device, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[0642] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?" If the emotional state is determined to be abnormal, the security team is immediately notified.

[0643] Example prompt sentence:

[0644] A user says, "I'm feeling very stressed because my project deadline is approaching."

[0645] An example prompt from a generative AI model: "The user's stress level is high. Please select the appropriate response."

[0646] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0647] Step 1:

[0648] Capture and transmit audio data

[0649] The user speaks into the smartphone terminal about their state of mind and emotions. The terminal captures this voice and stores it as digital audio data. The terminal then encrypts this audio data and transmits it to the server using a secure communication method (e.g., HTTPS).

[0650] Input: User's voice

[0651] Output: Encrypted digital audio data

[0652] Step 2:

[0653] Converting audio data to text

[0654] The server receives the voice data sent from the device and converts it into text using speech recognition technology. Specifically, it uses the Google Cloud Speech-to-Text API or Amazon Transcribe to convert the voice data into text.

[0655] Input: Encrypted digital audio data

[0656] Output: Text data

[0657] Step 3:

[0658] Emotional state analysis

[0659] The server inputs the obtained text data and the original voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the voice tone and text content to evaluate the user's psychological state. This analysis uses IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[0660] Input: Text and audio data

[0661] Output: Parsed emotional state

[0662] Step 4:

[0663] Generating a response

[0664] The server uses a generative AI model based on the analyzed emotional state to generate an appropriate response, which is then prepared for delivery in both text and audio formats, using natural language generation modules such as OpenAI GPT-4.

[0665] Input: Parsed emotional state

[0666] Output: Text and audio responses

[0667] Step 5:

[0668] Emergency notifications and connection to experts

[0669] If the emotional state is deemed abnormal or urgent, the server immediately notifies the security team via email or chat systems (e.g., Slack, Microsoft Teams), and automatically connects the individual to a specialist or counselor if necessary.

[0670] Input: Parsed emotional state and generated response

[0671] Output: Notify security team and connect to experts

[0672] Step 6:

[0673] Providing a response to the user

[0674] The server generates a response and sends it back to the terminal, which displays the response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[0675] Input: Text and audio responses

[0676] Output: Feedback to the user (text display and audio playback)

[0677] 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.

[0678] 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.

[0679] 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.

[0680] [Third embodiment]

[0681] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0682] 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.

[0683] 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).

[0684] 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.

[0685] 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.

[0686] 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).

[0687] 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.

[0688] 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.

[0689] 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.

[0690] 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.

[0691] 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.

[0692] 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."

[0693] This invention relates to a system that provides mental health care using generative AI. The main components are a server, a terminal, and a user, and the role of each will be specifically explained.

[0694] Server Processing

[0695] The server plays a central role in the system and operates as follows:

[0696] First, the server receives voice data from the user. This voice data is transmitted through the terminal. The server then converts the received voice data into text data using voice recognition technology. Specifically, the server uses a voice recognition library to analyze the voice and generate corresponding text.

[0697] The server then analyzes the user's emotional state using the text and audio data. It uses an emotion analysis algorithm to evaluate the user's psychological state based on the words and tone of the user's voice. Based on the analysis results, it determines whether the user's emotional state is urgent.

[0698] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and emotional state, which may include an empathetic message or advice tailored to the user's situation.

[0699] The server then generates the generated response in audio and text formats, using speech synthesis technology to convert the text to speech, and prepares both formats for transmission back to the user.

[0700] If the result of the emotion analysis indicates that the user's condition is urgent, the server will execute a procedure to connect to a specialist. If an emergency response is required, the server will immediately attempt to connect to a counselor or helpline based on the user's registered information. This allows the user to receive professional assistance quickly.

[0701] Terminal handling

[0702] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[0703] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[0704] User Experience

[0705] Users can talk to the device to ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this audio to the server. The server converts the audio into text and performs emotion analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[0706] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[0707] This allows users to receive prompt and professional mental health care, enabling early detection and treatment of psychological problems.

[0708] The processing flow will be explained below.

[0709] Server Processing

[0710] Step 1: Receive audio data

[0711] The server receives the voice data sent by the user as an HTTP request, which allows the user's voice input to reach the server.

[0712] Step 2: Convert the audio data to text

[0713] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice into text information.

[0714] Step 3: Analyze the emotional state

[0715] The server uses the text and audio data to run an emotion analysis algorithm to assess the user's emotional state, thereby determining the user's psychological state.

[0716] Step 4: Generate a response

[0717] The server uses the generative model to generate an appropriate response based on the analyzed emotional state and text data, thereby providing feedback to the user.

[0718] Step 5: Generate a response in audio and text format

[0719] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio formats.

[0720] Step 6: Connect with an expert

[0721] Based on the results of the sentiment analysis, the server will connect you to an expert if necessary, or if an emergency response is required, it will immediately connect you to a counselor or helpline.

[0722] Step 7: Send the response back to the user

[0723] The server generates responses and sends them back to the user in both text and audio format, allowing the user to receive these responses on their own device.

[0724] Terminal handling

[0725] Step 1: Capture the user's voice input

[0726] The user speaks to the device to discuss their concerns, and the device uses a microphone to capture this voice.

[0727] Step 2: Send the audio data to the server

[0728] The device sends the captured audio data to the server as an HTTP request, which allows the audio data to arrive at the server.

[0729] Step 3: Receive a response from the server

[0730] Once the response from the server is generated, the device receives the response, which may be in both text and audio format.

[0731] Step 4: Present the response to the user

[0732] The terminal displays the received response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[0733] User Experience

[0734] Step 1: Access the system

[0735] A user accesses the system using a terminal and starts a conversation for mental care.

[0736] Step 2: Use voice input

[0737] The user speaks into the device about their concerns and their condition. For example, they might say, "I've been feeling anxious lately and can't sleep."

[0738] Step 3: Receive a response

[0739] The user receives the response from the server in text and voice format through the terminal, thereby obtaining appropriate feedback on the content of the consultation.

[0740] Step 4: Connect with an expert

[0741] If the user's emotional state is deemed urgent, the device will connect them to a professional counselor or helpline, allowing them to receive the professional assistance they need.

[0742] Example 1

[0743] 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."

[0744] Mental health issues have become a serious problem in modern society, requiring prompt and appropriate responses. However, many people have limited access to mental health care, and access to specialists is difficult, especially in emergencies. Furthermore, conventional systems struggle to accurately analyze a user's emotional state and generate appropriate responses. Therefore, there is a need for a system that allows users to consult with a specialist with peace of mind and receive appropriate advice.

[0745] 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.

[0746] In this invention, the server includes means for receiving voice data using a voice input device, means for converting the voice data into text data using voice recognition technology, means for analyzing an emotional state from the text data and voice data using an emotion analysis algorithm, means for generating a response based on the text data and the emotional state using a generative AI model, means for generating and outputting the generated response in voice and text formats using voice synthesis technology, and means for automatically connecting to an expert if the emotional state satisfies certain conditions, thereby enabling the server to provide a prompt and appropriate response to a user's mental health issues and ensure connection to an expert in an emergency.

[0747] The "voice input device" is a device that a user uses to input voice data, and includes a microphone and the like.

[0748] "Speech recognition technology" is a technology that analyzes input voice data and converts it into corresponding text data.

[0749] An "emotion analysis algorithm" is an algorithm for analyzing text and audio data to assess a user's emotional state.

[0750] A "generative AI model" is a technology that uses a pre-trained artificial intelligence model to generate natural language responses based on input data.

[0751] "Speech synthesis technology" is a technology that converts text data into voice data and is used to provide a voice response to the user.

[0752] The "means for automatically connecting to an expert" is a means for instantly establishing a connection to an expert, counselor, helpline, etc. when the results of the user's emotion analysis satisfy certain conditions.

[0753] This invention relates to a system that uses generative AI to provide mental health care. The main components are a server, a terminal, and a user, and by working together, they provide users with prompt and appropriate mental health care.

[0754] Server Processing

[0755] The server plays a central role in the system and operates as follows: First, the server receives voice data from the user. This voice data is sent through the terminal, and the server uses voice recognition technology to convert the voice data into text data. For this conversion, a voice recognition library (for example, Google Cloud Speech-to-Text API) is used.

[0756] The server then runs a sentiment analysis algorithm using the text and audio data to analyze the user's emotional state, using services such as Microsoft Azure Text Analytics and AWS Comprehend, and determines whether the user's emotional state is urgent.

[0757] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the user's emotional state. This response may include empathetic messages or advice tailored to the user's situation. The generated response is generated in audio and text formats using speech synthesis technologies such as Amazon Polly or the Google Cloud Text-to-Speech API.

[0758] Finally, the server prepares to send the generated response back to the terminal. If the result of the sentiment analysis indicates that the user's condition is urgent, the server executes the procedure to connect to an expert, so that the user can receive professional assistance quickly.

[0759] Terminal handling

[0760] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[0761] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[0762] User Experience

[0763] Users can talk to the device and ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this voice data to the server. The server converts the voice into text and performs emotional analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[0764] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[0765] Specific examples

[0766] As an example of a specific prompt sentence, if a user says to the device, "I've been feeling anxious lately and can't sleep at night," the following prompt will be sent to the generation AI.

[0767] User input: "I've been feeling anxious lately and can't sleep at night."

[0768] Analyze the text for sentiment and generate an empathetic response:

[0769] This allows the server to generate an appropriate response and provide the user with fast, professional assistance.

[0770] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0771] Step 1:

[0772] The user provides voice input

[0773] Users can input their mental health concerns or questions into a device such as a smartphone or PC by voice. For example, they might say, "I've been feeling anxious lately and can't sleep at night." This is input as voice data.

[0774] Step 2:

[0775] The device sends the audio data to the server.

[0776] The device uses a microphone to capture the user's voice and saves it as digital audio data. It then sends this audio data to the server using an HTTP request. The input is the audio data, and the output is the digital audio data sent to the server. Specifically, the device generates a request like this:

[0777] Step 3:

[0778] The server converts the audio data into text data.

[0779] The server converts the received voice data into text format using speech recognition technology. This conversion is performed using a speech recognition library (for example, Google Cloud Speech-to-Text API). The input is voice data, and the output is text data. The server then passes the converted text data to the next processing step.

[0780] Step 4:

[0781] The server performs sentiment analysis

[0782] The server analyzes the emotional state using text and voice data. An emotion analysis algorithm (for example, Microsoft Azure Text Analytics or AWS Comprehend) is used here. The input is text and voice data, and the output is the result of the emotion analysis (emotional state). Based on this data, the server determines whether the user's emotional state is urgent.

[0783] Step 5:

[0784] The server generates a response using the generative AI model

[0785] The server uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the emotional state. The input is the text data and the results of the emotional analysis, and the output is the generated text response. Specifically, the prompt sentence is generated as follows:

[0786] User input: "I've been feeling anxious lately and can't sleep at night."

[0787] Analyze the text for sentiment and generate an empathetic response:”

[0788] Step 6:

[0789] The server generates a generated response in audio and text format.

[0790] The server converts the generated text response into audio format using speech synthesis technology (e.g., Amazon Polly or Google Cloud Text-to-Speech API). The input is the generated text response, and the output is the audio data and the text response.

[0791] Step 7:

[0792] The server sends the response to the device

[0793] The server sends the generated voice data and text response back to the device, using an HTTP response to send the data. The input is the generated voice data and text response, and the output is the response sent to the device.

[0794] Step 8:

[0795] The terminal presents the response to the user

[0796] The terminal analyzes the response received from the server, displays the text response on the screen, and plays the audio response. The input is the response data from the server, and the output is the audio and text response presented to the user, who receives it visually and audibly.

[0797] Step 9:

[0798] Emergency response

[0799] If the result of the sentiment analysis indicates that the user's condition is urgent, the server automatically connects the user to an expert, providing appropriate assistance based on the expert's contact information. The input is the result of the sentiment analysis and the user's registration information, and the output is the connection to the expert.

[0800] (Application example 1)

[0801] 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."

[0802] In conventional security services, potential risks and signs of trouble in the surrounding area are often overlooked through human attention alone. Furthermore, when an emergency occurs, a quick and appropriate response is required, but relying solely on human judgment carries risks. For this reason, there is a need for a system that can accurately analyze surrounding sounds and provide early warning of potential risks.

[0803] 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.

[0804] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing emotional states from the text data and voice data, means for capturing an unspecified number of voice data and performing emotional analysis to detect potential risks, and means for issuing an alarm when the potential risks are detected. This makes it possible to detect potential risks from surrounding voice data early and respond quickly.

[0805] "Speech data" refers to data that includes human speech and surrounding sounds acquired through a voice input device such as a microphone.

[0806] "Text data" refers to data that has been analyzed and converted into text information from voice data.

[0807] "Emotional state" refers to the psychological state or emotion that can be inferred from a person's speech or tone of voice.

[0808] A "generative model" is an algorithm or AI model that generates appropriate responses or predictions based on input data.

[0809] An "expert" is an individual or organization with advanced knowledge and skills in a particular field who provides assistance, especially in emergency situations.

[0810] A "potential risk" is a condition or phenomenon that has not yet manifested itself, but which has the potential to develop into a problem or trouble in the future.

[0811] An "alert" is a voice, text, or alarm that is issued to alert you when a particular condition is detected.

[0812] This invention is a system that uses generative AI to provide mental health care and security services. The main components are a server, a terminal, and a user, and the role of each will be explained in detail.

[0813] Server Processing

[0814] The server plays a central role in the system and operates as follows:

[0815] First, the server receives voice data from the user. This voice data is sent from the terminal via an HTTP request. The server converts the voice data into text data using a speech recognition library (e.g., speech_recognition library).

[0816] The server then analyzes the user's emotional state using the text and audio data. For emotion analysis, it applies the generative AI model's emotion analysis algorithm using Hugging Face's Transformer library, thereby assessing the user's psychological state.

[0817] The server then uses a generative AI model to generate an appropriate response based on the text data and the user's emotional state. This response may include an empathetic message or advice tailored to the user's situation. The generated response is then converted into audio format using speech synthesis technology (e.g., gTTS library) and sent back to the user along with the text format.

[0818] If the sentiment analysis determines that the user's condition is urgent, the server will implement measures to connect them to an expert, including using the system's pre-registered contact information for counselors and helplines to immediately connect them.

[0819] Terminal handling

[0820] The terminal supports the exchange of voice data between the user and the server. When the user speaks into the terminal, this voice is captured by the terminal. The terminal sends this voice data to the server as an HTTP request. When the terminal receives a response from the server, it displays this response in text format and plays it back in audio format. This allows the user to receive messages from the server both visually and audibly.

[0821] User Experience

[0822] Users can talk to their device to ask questions or ask about mental health or security. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this speech to the server. The server converts the speech data into text data and performs emotion analysis.

[0823] If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your life recently?" This response is returned to the user in voice and text format.

[0824] When used as a security service, it can detect potential risks from an unspecified number of voice data. For example, if a heated argument is occurring nearby, the server can analyze the voice data and issue an alert such as, "Caution: A potential risk has been detected. Please check your surroundings."

[0825] An example of a prompt sentence could be, "You are a virtual security guard. Please analyze the text data below and evaluate the emotional state. If there are a lot of negative emotions, please return a warning."

[0826] The system allows users to receive prompt and professional mental health care and security services that can detect potential risks early from surrounding sounds.

[0827] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0828] Step 1:

[0829] The user speaks into the terminal. For example, the user asks a question or asks for advice, such as, "I've been feeling anxious lately and can't sleep at night." The input is the user's voice data, which is received by the terminal. The voice data is obtained as output.

[0830] Step 2:

[0831] The terminal sends the received voice data to the server as an HTTP request. At this time, the terminal converts the voice data into an appropriate data format for sending to the server. The input is the user's voice data, and the voice data converted into the HTTP request format is sent to the server as the output.

[0832] Step 3:

[0833] The server receives the HTTP request and retrieves the audio data. It then converts the audio data into text using a speech recognition library (e.g., speech_recognition). The input is the audio data, and the output is the converted text data.

[0834] Step 4:

[0835] The server uses the converted text and audio data to perform emotion analysis. The model uses a generative AI model based on Hugging Face's Transformer library. The input is text and audio data, and the output is an emotional state.

[0836] Step 5:

[0837] The server uses a generative AI model based on the emotional state to generate an appropriate response. This generated response is based on the user's psychological state and the question content. The input is the emotional state and text data, and the output is a response text.

[0838] Step 6:

[0839] The server converts the generated response text into audio using speech synthesis technology (e.g., gTTS), and also prepares response data in text format. The input is the generated response text, and audio data and text data are generated as output.

[0840] Step 7:

[0841] The server returns the response data to the terminal as an HTTP response. The input is voice data and text data, and the HTTP response is sent to the terminal as output.

[0842] Step 8:

[0843] The terminal displays and plays back the response data received from the server. Specifically, it displays text data on the screen and plays audio data through the speaker. The input is response data in HTTP response format, and the output presents visual and auditory information to the user.

[0844] 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.

[0845] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[0846] Server Processing

[0847] The server plays a central role in the system and operates as follows:

[0848] First, the server receives the voice data sent by the user. After receiving the voice data, the server uses voice recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information.

[0849] The server then inputs the text and voice data into an emotion engine to perform a detailed analysis of the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to assess the user's psychological state. Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response.

[0850] The generated responses are provided to the user in both text and audio formats. The server utilizes speech synthesis technology to convert text responses into audio. It also has the means to automatically connect to experts in emergencies based on the results of sentiment analysis.

[0851] As a concrete example, suppose a user says, "I've been feeling anxious and can't sleep lately." In this case, the device captures the audio and sends it to the server. The server converts the audio into text and analyzes it using an emotion engine. If the emotion engine determines that the user is in a high state of anxiety, the generative AI model will generate an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?" Furthermore, if the emotional state is determined to be urgent, the server will immediately attempt to connect the user to a counselor or helpline.

[0852] Terminal handling

[0853] The device supports the exchange of voice data between the user and the server. When the user speaks to the device about their concerns, the device captures this voice and sends it to the server. When the device receives a response from the server, it provides this response to the user. Specifically, it displays the response in text format and plays it back in audio format at the same time. This allows the user to receive the response both visually and audibly.

[0854] User Experience

[0855] Users access the system through their device to receive mental health care. When the user speaks to the device about their condition and concerns, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[0856] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to get some rest. Why don't you try doing something to relax?" The response is provided to the user, and if necessary, an automatic connection to an expert is also made.

[0857] This system allows users to receive appropriate mental care anytime, anywhere, and enables early detection and treatment of psychological problems. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[0858] The processing flow will be explained below.

[0859] Server Processing

[0860] Step 1: Receive audio data

[0861] The server receives the voice data sent by the user through the terminal as an HTTP request, and the user's voice input is then incorporated into the system.

[0862] Step 2: Convert the audio data to text

[0863] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice data into a format that can be processed as text information.

[0864] Step 3: Analyze the emotional state using the emotion engine

[0865] The server inputs the converted text and voice data into an emotion engine for detailed analysis of the user's emotional state. The emotion engine evaluates the tone of the voice and the content of the text to determine the user's psychological state.

[0866] Step 4: Generate a response

[0867] The server uses a generative AI model based on the analyzed emotional state and text data to generate an appropriate response, creating feedback tailored to the user's desired assistance.

[0868] Step 5: Generate a response in audio and text format

[0869] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio.

[0870] Step 6: Connect with an expert

[0871] Based on the analysis results of the emotion engine, if the server determines that the user's emotional state is urgent, it will automatically connect the user to a counselor, helpline, or other expert, allowing the user to receive professional assistance quickly.

[0872] Step 7: Send the response back to the user

[0873] The server generates text and audio responses and sends them back to the user, who can then receive the necessary assistance at their terminal.

[0874] Terminal handling

[0875] Step 1: Capture the user's voice input

[0876] When the user speaks into the terminal about the content of the consultation, the terminal uses a microphone to capture the voice, and the user's voice is thus captured by the terminal.

[0877] Step 2: Send the audio data to the server

[0878] The device compiles the captured audio data into an HTTP request and sends it to the server. This process allows the audio data to reach the server.

[0879] Step 3: Receive a response from the server

[0880] The generated response is sent back from the server and received by the device.

[0881] Step 4: Present the response to the user

[0882] The terminal displays the received response in text format and plays it back in audio format, allowing the user to understand the response visually and audibly.

[0883] User Experience

[0884] Step 1: Access the system

[0885] A user accesses the system using a terminal and prepares to receive mental health care.

[0886] Step 2: Use voice input

[0887] The user verbally explains their condition and concerns to the device, for example, "I've been feeling a lot of stress at work lately."

[0888] Step 3: Receive a response

[0889] The user receives responses from the server via their device, which are provided in both text and audio formats, allowing the user to receive feedback in the format that is most convenient for them.

[0890] Step 4: Connect with an expert

[0891] If the emotion engine determines that the user's emotional state is urgent, the user will be automatically connected to a professional counselor or helpline, allowing them to receive direct, professional assistance.

[0892] This system allows users to receive appropriate mental health care in real time, enabling early detection and treatment of psychological problems. In addition, by incorporating an emotion engine, it is possible to grasp the user's emotional state in more detail and provide advanced support based on the results.

[0893] Example 2

[0894] 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."

[0895] Conventional mental health care systems face challenges in accurately analyzing users' emotional states and providing appropriate responses based on those results. They often lack the ability to quickly connect users to specialists in emergencies. Furthermore, they lack a way to provide both visual and auditory responses to improve the user experience.

[0896] 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.

[0897] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and the voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, and means for inputting a prompt sentence to the generative model based on the emotion analysis result. This enables the system to analyze the user's emotional state in detail and provide an appropriate response in real time. Furthermore, in an emergency, it can quickly connect to an expert, improving the quality of support provided to the user.

[0898] The "means for receiving voice data" refers to software and hardware for acquiring voice uttered by the user in digital form and transmitting it to the server.

[0899] "Means for converting voice data into text data" refers to software technology for converting received voice data into text information using voice recognition technology.

[0900] "Means for analyzing emotional states from text data and voice data" refers to an emotion analysis engine that analyzes text and voice data and evaluates the user's emotions and psychological state.

[0901] "Means for generating a response using a generative model" refers to a process that uses an artificial intelligence model to generate an appropriate response based on the results of sentiment analysis.

[0902] "Means for outputting the generated response in audio and text form" refers to speech synthesis and display technologies for converting the generated text response into audio and also displaying it in text form.

[0903] "Means for connecting to an expert based on emotional state" refers to a communication means for automatically connecting to an expert or counselor when the user's emotional state is determined to be urgent.

[0904] "Means for inputting prompt sentences to a generative model based on the results of sentiment analysis" refers to an input means for giving appropriate instructions to a generative model based on the results of sentiment analysis.

[0905] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[0906] Server Processing

[0907] The server plays a central role in the system and has a means of receiving voice data. For example, when a user says, "I've been feeling anxious and can't sleep lately," the voice data is sent from the device to the server. This is done using a microphone as hardware and a real-time voice capture tool as software.

[0908] Next, the server uses speech recognition technology (for example, Google Cloud Speech-to-Text) to convert the voice data into text data. This process generates the text, "I've been feeling anxious lately and can't sleep."

[0909] The server inputs the generated text and voice data into an emotion engine, which uses NVIDIA Clara Guardian and Microsoft Azure Video Indexer to perform a detailed analysis of the user's emotional state. This allows the detection of high anxiety levels.

[0910] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate response, such as, "That's tough. Have there been any changes in your living situation recently?"

[0911] The generated responses are provided to the user in audio and text formats using speech synthesis technologies (e.g., Amazon Polly, Google Cloud Text-to-Speech), with the option to connect with an expert if the emotional state is deemed urgent.

[0912] Terminal handling

[0913] The device supports the exchange of voice data between the user and the server. When the user speaks about their worries into the device, the device captures this voice and sends it to the server. For example, the device captures the user's voice saying, "I've been anxious and can't sleep lately," and sends it digitally to the server.

[0914] When the device receives a response from the server, it provides the response to the user. Specifically, it displays the response "That's tough. Have there been any changes in your living environment recently?" in text format and plays it back in audio format. A text display tool and an audio playback tool are used.

[0915] User Experience

[0916] Users access the system through their device and receive mental health care. When the user says something like, "I've been feeling stressed lately because of a lot of work," this speech is sent from the device to the server. The server converts the speech into text and analyzes it using an emotion engine. Based on the analysis results, the generative AI model generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?"

[0917] This response is provided to the user, and if necessary, an automatic connection to a specialist is also made. This allows users to receive appropriate mental care anytime, anywhere. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[0918] Prompt Sentence Examples

[0919] A specific example of an input prompt sentence for a generative AI model is as follows:

[0920] If a user says, "I've been feeling anxious and can't sleep lately," we convert this speech into text and analyze it with the emotion engine. We want to generate an empathetic response like this:

[0921] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0922] Step 1:

[0923] The user inputs voice into the device. For example, they might say, "I've been feeling anxious lately and can't sleep." This voice data is captured by the device's microphone. The input is the user's voice data, and the output is a digital audio file. Specifically, the device records the voice data and processes it in real time.

[0924] Step 2:

[0925] The device sends audio data to the server. In this process, a digital audio file is sent to the server via the Internet. The input is the audio data captured by the device, and the output is the audio data transferred to the server. Specifically, the device uses a data transmission protocol to upload the audio data to the server.

[0926] Step 3:

[0927] The server receives the audio data. The server stores the audio data in buffer memory and prepares it for the next processing step. The input is the audio data sent from the terminal, and the output is the server's buffer memory where the audio data is stored. Specifically, the server receives an HTTP request and stores the audio data.

[0928] Step 4:

[0929] The server uses speech recognition technology to convert the voice data into text data. For example, Google Cloud Speech-to-Text is used. In this process, the audio file is converted into text in a natural language. The input is the stored voice data, and the output is text data. Specifically, the server makes an API call to convert the voice data into text.

[0930] Step 5:

[0931] The server inputs text and audio data into the emotion engine to analyze the emotional state. NVIDIA Clara Guardian and Microsoft Azure Video Indexer are used. In this step, the emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's emotional state. The input is text and audio data, and the output is the analysis result of the emotional state. Specifically, the server calls the emotion analysis API and obtains the results.

[0932] Step 6:

[0933] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-4) based on the emotion analysis results. The input is the emotion analysis results, and the output is the generated response text. Specifically, the server inputs a prompt sentence into the generative AI model, causing it to generate an empathetic text response. An example of a prompt sentence is, "If a user says, 'I've been unable to sleep recently because I've been anxious,' we converted this speech into text and analyzed it with the emotion engine, which detected high levels of anxiety. We would like you to generate an empathetic response such as the following."

[0934] Step 7:

[0935] The server converts the generated text response into speech. Speech synthesis technology (e.g., Amazon Polly, Google Cloud Text-to-Speech) is used. The input is the generated text response, and the output is the speech response data. Specifically, the server calls an API that converts text data into speech data and generates an audio file.

[0936] Step 8:

[0937] The server sends the generated text and voice responses to the terminal. The input is the generated text and voice response data, and the output is the data sent to the terminal. In specific operations, the server uses a data transmission protocol to upload the text and voice data to the terminal.

[0938] Step 9:

[0939] The terminal receives the response from the server and provides it to the user. Specifically, it displays the generated response in text format and simultaneously plays it in audio format. The input is the text and audio response data sent from the server, and the output is the response presented to the user. Specifically, the terminal uses a text display tool and an audio playback tool to provide the response to the user visually and audibly.

[0940] Step 10:

[0941] In an emergency, the server uses a means to connect to an expert based on the user's emotional state. For example, if the user's emotional state is very unstable and requires an emergency, the server automatically attempts to connect to an expert. The input is the result of the judgment of the urgency of the emotional state, and the output is a connection to an expert. Specifically, the server uses a communication means to contact an expert or counselor.

[0942] This allows users to receive appropriate mental health care anytime, anywhere.

[0943] (Application example 2)

[0944] 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."

[0945] Employees' psychological state has a significant impact on their work, and sustained high levels of stress and anxiety can lead to security incidents and reduced work efficiency. Conventional systems have difficulty monitoring employees' psychological state in real time and detecting abnormalities early, often resulting in delayed implementation of appropriate security measures and responses. This invention aims to prevent security incidents and improve work efficiency by analyzing employees' emotional state using their voice data and immediately notifying the security team when an abnormality is detected, enabling a rapid response.

[0946] 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 receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, means for monitoring the emotional state and detecting an abnormal psychological state, and means for notifying a security team when the abnormal psychological state is detected. This makes it possible to monitor the psychological state of employees in real time and respond immediately when an abnormality is detected.

[0947] "Voice data" refers to data that is a digital recording of a user's spoken words or vocalizations.

[0948] "Text data" refers to data that includes character information converted from audio data.

[0949] "Emotional state" is information that represents the user's psychological state and feelings, and is analyzed from voice data and text data.

[0950] "Means for analyzing" is a technique or device for processing audio and text data and assessing the user's emotional state.

[0951] A "generative model" is a machine learning algorithm or artificial intelligence technique for generating responses based on an analyzed emotional state.

[0952] A "means for generating a response" is a technique or device that uses a generative model to create an appropriate response that corresponds to the user's emotional state.

[0953] "Output means" refers to a technique or device for providing the generated response in audio and text form.

[0954] "Means for connecting" refers to technology or devices that allow a user to connect with an expert or counselor when they are in an emergency psychological state.

[0955] "Monitoring means" refers to technology or devices that continuously observe the user's emotional state and detect abnormalities early on.

[0956] "Means for notifying" is a technology or device that sends an alert to the security team when an abnormal psychological state is detected.

[0957] A specific embodiment of the present invention will be described below. The following system analyzes emotional states using voice data, detects abnormal psychological states, and notifies the security team, thereby preventing security incidents and improving work efficiency.

[0958] Server Processing

[0959] The server plays a central role in this system. First, it receives the voice data sent by the user. Once the voice data is received, the server uses speech recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information. For speech recognition, Google Cloud Speech-to-Text API and Amazon Transcribe are used.

[0960] The server then inputs the text and voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's psychological state. This analysis is performed using IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[0961] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response. The generated response is provided to the user in both text and audio format. Response generation uses natural language generation modules such as OpenAI GPT-4.

[0962] Furthermore, if the emotional state is deemed abnormal, the server has the means to immediately notify the security team, which can be done via email or a chat system (e.g., Slack or Microsoft Teams).

[0963] Terminal handling

[0964] The device supports the exchange of voice data between the user and the server. When the user speaks a report into the device, the device captures this voice and sends it to the server. Specifically, this capture is performed using the smartphone's microphone and an appropriate voice capture module (iOS, Android).

[0965] Upon receiving a response from the server, the terminal provides the response to the user by displaying it in text format and playing it back in audio format, allowing the user to receive the response both visually and audibly.

[0966] User Experience

[0967] Users access the system through their devices and provide feedback on their emotional state. When the user speaks about their state into the device, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[0968] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?" If the emotional state is determined to be abnormal, the security team is immediately notified.

[0969] Example prompt sentence:

[0970] A user says, "I'm feeling very stressed because my project deadline is approaching."

[0971] An example prompt from a generative AI model: "The user's stress level is high. Please select the appropriate response."

[0972] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0973] Step 1:

[0974] Capture and transmit audio data

[0975] The user speaks into the smartphone terminal about their state of mind and emotions. The terminal captures this voice and stores it as digital audio data. The terminal then encrypts this audio data and transmits it to the server using a secure communication method (e.g., HTTPS).

[0976] Input: User's voice

[0977] Output: Encrypted digital audio data

[0978] Step 2:

[0979] Converting audio data to text

[0980] The server receives the voice data sent from the device and converts it into text using speech recognition technology. Specifically, it uses the Google Cloud Speech-to-Text API or Amazon Transcribe to convert the voice data into text.

[0981] Input: Encrypted digital audio data

[0982] Output: Text data

[0983] Step 3:

[0984] Emotional state analysis

[0985] The server inputs the obtained text data and the original voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the voice tone and text content to evaluate the user's psychological state. This analysis uses IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[0986] Input: Text and audio data

[0987] Output: Parsed emotional state

[0988] Step 4:

[0989] Generating a response

[0990] The server uses a generative AI model based on the analyzed emotional state to generate an appropriate response, which is then prepared for delivery in both text and audio formats, using natural language generation modules such as OpenAI GPT-4.

[0991] Input: Parsed emotional state

[0992] Output: Text and audio responses

[0993] Step 5:

[0994] Emergency notifications and connection to experts

[0995] If the emotional state is deemed abnormal or urgent, the server immediately notifies the security team via email or chat systems (e.g., Slack, Microsoft Teams), and automatically connects the individual to a specialist or counselor if necessary.

[0996] Input: Parsed emotional state and generated response

[0997] Output: Notify security team and connect to experts

[0998] Step 6:

[0999] Providing a response to the user

[1000] The server generates a response and sends it back to the terminal, which displays the response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[1001] Input: Text and audio responses

[1002] Output: Feedback to the user (text display and audio playback)

[1003] 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.

[1004] 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.

[1005] 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.

[1006] [Fourth embodiment]

[1007] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1008] 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.

[1009] 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).

[1010] 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.

[1011] 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.

[1012] 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).

[1013] 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.

[1014] 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.

[1015] 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.

[1016] 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.

[1017] 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.

[1018] 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.

[1019] 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."

[1020] This invention relates to a system that provides mental health care using generative AI. The main components are a server, a terminal, and a user, and the role of each will be specifically explained.

[1021] Server Processing

[1022] The server plays a central role in the system and operates as follows:

[1023] First, the server receives voice data from the user. This voice data is transmitted through the terminal. The server then converts the received voice data into text data using voice recognition technology. Specifically, the server uses a voice recognition library to analyze the voice and generate corresponding text.

[1024] The server then analyzes the user's emotional state using the text and audio data. It uses an emotion analysis algorithm to evaluate the user's psychological state based on the words and tone of the user's voice. Based on the analysis results, it determines whether the user's emotional state is urgent.

[1025] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and emotional state, which may include an empathetic message or advice tailored to the user's situation.

[1026] The server then generates the generated response in audio and text formats, using speech synthesis technology to convert the text to speech, and prepares both formats for transmission back to the user.

[1027] If the result of the emotion analysis indicates that the user's condition is urgent, the server will execute a procedure to connect to a specialist. If an emergency response is required, the server will immediately attempt to connect to a counselor or helpline based on the user's registered information. This allows the user to receive professional assistance quickly.

[1028] Terminal handling

[1029] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[1030] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[1031] User Experience

[1032] Users can talk to the device to ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this audio to the server. The server converts the audio into text and performs emotion analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[1033] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[1034] This allows users to receive prompt and professional mental health care, enabling early detection and treatment of psychological problems.

[1035] The processing flow will be explained below.

[1036] Server Processing

[1037] Step 1: Receive audio data

[1038] The server receives the voice data sent by the user as an HTTP request, which allows the user's voice input to reach the server.

[1039] Step 2: Convert the audio data to text

[1040] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice into text information.

[1041] Step 3: Analyze the emotional state

[1042] The server uses the text and audio data to run an emotion analysis algorithm to assess the user's emotional state, thereby determining the user's psychological state.

[1043] Step 4: Generate a response

[1044] The server uses the generative model to generate an appropriate response based on the analyzed emotional state and text data, thereby providing feedback to the user.

[1045] Step 5: Generate a response in audio and text format

[1046] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio formats.

[1047] Step 6: Connect with an expert

[1048] Based on the results of the sentiment analysis, the server will connect you to an expert if necessary, or if an emergency response is required, it will immediately connect you to a counselor or helpline.

[1049] Step 7: Send the response back to the user

[1050] The server generates responses and sends them back to the user in both text and audio format, allowing the user to receive these responses on their own device.

[1051] Terminal handling

[1052] Step 1: Capture the user's voice input

[1053] The user speaks to the device to discuss their concerns, and the device uses a microphone to capture this voice.

[1054] Step 2: Send the audio data to the server

[1055] The device sends the captured audio data to the server as an HTTP request, which allows the audio data to arrive at the server.

[1056] Step 3: Receive a response from the server

[1057] Once the response from the server is generated, the device receives the response, which may be in both text and audio format.

[1058] Step 4: Present the response to the user

[1059] The terminal displays the received response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[1060] User Experience

[1061] Step 1: Access the system

[1062] A user accesses the system using a terminal and starts a conversation for mental care.

[1063] Step 2: Use voice input

[1064] The user speaks into the device about their concerns and their condition. For example, they might say, "I've been feeling anxious lately and can't sleep."

[1065] Step 3: Receive a response

[1066] The user receives the response from the server in text and voice format through the terminal, thereby obtaining appropriate feedback on the content of the consultation.

[1067] Step 4: Connect with an expert

[1068] If the user's emotional state is deemed urgent, the device will connect them to a professional counselor or helpline, allowing them to receive the professional assistance they need.

[1069] Example 1

[1070] 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."

[1071] Mental health issues have become a serious problem in modern society, requiring prompt and appropriate responses. However, many people have limited access to mental health care, and access to specialists is difficult, especially in emergencies. Furthermore, conventional systems struggle to accurately analyze a user's emotional state and generate appropriate responses. Therefore, there is a need for a system that allows users to consult with a specialist with peace of mind and receive appropriate advice.

[1072] 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.

[1073] In this invention, the server includes means for receiving voice data using a voice input device, means for converting the voice data into text data using voice recognition technology, means for analyzing an emotional state from the text data and voice data using an emotion analysis algorithm, means for generating a response based on the text data and the emotional state using a generative AI model, means for generating and outputting the generated response in voice and text formats using voice synthesis technology, and means for automatically connecting to an expert if the emotional state satisfies certain conditions, thereby enabling the server to provide a prompt and appropriate response to a user's mental health issues and ensure connection to an expert in an emergency.

[1074] The "voice input device" is a device that a user uses to input voice data, and includes a microphone and the like.

[1075] "Speech recognition technology" is a technology that analyzes input voice data and converts it into corresponding text data.

[1076] An "emotion analysis algorithm" is an algorithm for analyzing text and audio data to assess a user's emotional state.

[1077] A "generative AI model" is a technology that uses a pre-trained artificial intelligence model to generate natural language responses based on input data.

[1078] "Speech synthesis technology" is a technology that converts text data into voice data and is used to provide a voice response to the user.

[1079] The "means for automatically connecting to an expert" is a means for instantly establishing a connection to an expert, counselor, helpline, etc. when the results of the user's emotion analysis satisfy certain conditions.

[1080] This invention relates to a system that uses generative AI to provide mental health care. The main components are a server, a terminal, and a user, and by working together, they provide users with prompt and appropriate mental health care.

[1081] Server Processing

[1082] The server plays a central role in the system and operates as follows: First, the server receives voice data from the user. This voice data is sent through the terminal, and the server uses voice recognition technology to convert the voice data into text data. For this conversion, a voice recognition library (for example, Google Cloud Speech-to-Text API) is used.

[1083] The server then runs a sentiment analysis algorithm using the text and audio data to analyze the user's emotional state, using services such as Microsoft Azure Text Analytics and AWS Comprehend, and determines whether the user's emotional state is urgent.

[1084] The server then uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the user's emotional state. This response may include empathetic messages or advice tailored to the user's situation. The generated response is generated in audio and text formats using speech synthesis technologies such as Amazon Polly or the Google Cloud Text-to-Speech API.

[1085] Finally, the server prepares to send the generated response back to the terminal. If the result of the sentiment analysis indicates that the user's condition is urgent, the server executes the procedure to connect to an expert, so that the user can receive professional assistance quickly.

[1086] Terminal handling

[1087] The device supports the exchange of voice data between the user and the server. When the user speaks into the device, the voice is captured by the device. The device then makes an HTTP request to send the voice data to the server, and sends the audio file to the server.

[1088] When the terminal receives a response from the server, it presents the response to the user. Specifically, the terminal displays the response in text format and plays it back in audio format, allowing the user to receive the message from the server both visually and audibly.

[1089] User Experience

[1090] Users can talk to the device and ask questions or seek advice about their mental health. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this voice data to the server. The server converts the voice into text and performs emotional analysis. If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?"

[1091] The response is returned to the user in audio and text formats, and the user can listen to it in real time via their device. If the user's condition is severe, the server can automatically connect them to a counselor or helpline to receive direct professional assistance.

[1092] Specific examples

[1093] As an example of a specific prompt sentence, if a user says to the device, "I've been feeling anxious lately and can't sleep at night," the following prompt will be sent to the generation AI.

[1094] User input: "I've been feeling anxious lately and can't sleep at night."

[1095] Analyze the text for sentiment and generate an empathetic response:

[1096] This allows the server to generate an appropriate response and provide the user with fast, professional assistance.

[1097] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1098] Step 1:

[1099] The user provides voice input

[1100] Users can input their mental health concerns or questions into a device such as a smartphone or PC by voice. For example, they might say, "I've been feeling anxious lately and can't sleep at night." This is input as voice data.

[1101] Step 2:

[1102] The device sends the audio data to the server.

[1103] The device uses a microphone to capture the user's voice and saves it as digital audio data. It then sends this audio data to the server using an HTTP request. The input is the audio data, and the output is the digital audio data sent to the server. Specifically, the device generates a request like this:

[1104] Step 3:

[1105] The server converts the audio data into text data.

[1106] The server converts the received voice data into text format using speech recognition technology. This conversion is performed using a speech recognition library (for example, Google Cloud Speech-to-Text API). The input is voice data, and the output is text data. The server then passes the converted text data to the next processing step.

[1107] Step 4:

[1108] The server performs sentiment analysis

[1109] The server analyzes the emotional state using text and voice data. An emotion analysis algorithm (for example, Microsoft Azure Text Analytics or AWS Comprehend) is used here. The input is text and voice data, and the output is the result of the emotion analysis (emotional state). Based on this data, the server determines whether the user's emotional state is urgent.

[1110] Step 5:

[1111] The server generates a response using the generative AI model

[1112] The server uses a generative AI model (e.g., a generative language model) to generate an appropriate response based on the text data and the emotional state. The input is the text data and the results of the emotional analysis, and the output is the generated text response. Specifically, the prompt sentence is generated as follows:

[1113] User input: "I've been feeling anxious lately and can't sleep at night."

[1114] Analyze the text for sentiment and generate an empathetic response:”

[1115] Step 6:

[1116] The server generates a generated response in audio and text format.

[1117] The server converts the generated text response into audio format using speech synthesis technology (e.g., Amazon Polly or Google Cloud Text-to-Speech API). The input is the generated text response, and the output is the audio data and the text response.

[1118] Step 7:

[1119] The server sends the response to the device

[1120] The server sends the generated voice data and text response back to the device, using an HTTP response to send the data. The input is the generated voice data and text response, and the output is the response sent to the device.

[1121] Step 8:

[1122] The terminal presents the response to the user

[1123] The terminal analyzes the response received from the server, displays the text response on the screen, and plays the audio response. The input is the response data from the server, and the output is the audio and text response presented to the user, who receives it visually and audibly.

[1124] Step 9:

[1125] Emergency response

[1126] If the result of the sentiment analysis indicates that the user's condition is urgent, the server automatically connects the user to an expert, providing appropriate assistance based on the expert's contact information. The input is the result of the sentiment analysis and the user's registration information, and the output is the connection to the expert.

[1127] (Application example 1)

[1128] 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."

[1129] In conventional security services, potential risks and signs of trouble in the surrounding area are often overlooked through human attention alone. Furthermore, when an emergency occurs, a quick and appropriate response is required, but relying solely on human judgment carries risks. For this reason, there is a need for a system that can accurately analyze surrounding sounds and provide early warning of potential risks.

[1130] 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.

[1131] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing emotional states from the text data and voice data, means for capturing an unspecified number of voice data and performing emotional analysis to detect potential risks, and means for issuing an alarm when the potential risks are detected. This makes it possible to detect potential risks from surrounding voice data early and respond quickly.

[1132] "Speech data" refers to data that includes human speech and surrounding sounds acquired through a voice input device such as a microphone.

[1133] "Text data" refers to data that has been analyzed and converted into text information from voice data.

[1134] "Emotional state" refers to the psychological state or emotion that can be inferred from a person's speech or tone of voice.

[1135] A "generative model" is an algorithm or AI model that generates appropriate responses or predictions based on input data.

[1136] An "expert" is an individual or organization with advanced knowledge and skills in a particular field who provides assistance, especially in emergency situations.

[1137] A "potential risk" is a condition or phenomenon that has not yet manifested itself, but which has the potential to develop into a problem or trouble in the future.

[1138] An "alert" is a voice, text, or alarm that is issued to alert you when a particular condition is detected.

[1139] This invention is a system that uses generative AI to provide mental health care and security services. The main components are a server, a terminal, and a user, and the role of each will be explained in detail.

[1140] Server Processing

[1141] The server plays a central role in the system and operates as follows:

[1142] First, the server receives voice data from the user. This voice data is sent from the terminal via an HTTP request. The server converts the voice data into text data using a speech recognition library (e.g., speech_recognition library).

[1143] The server then analyzes the user's emotional state using the text and audio data. For emotion analysis, it applies the generative AI model's emotion analysis algorithm using Hugging Face's Transformer library, thereby assessing the user's psychological state.

[1144] The server then uses a generative AI model to generate an appropriate response based on the text data and the user's emotional state. This response may include an empathetic message or advice tailored to the user's situation. The generated response is then converted into audio format using speech synthesis technology (e.g., gTTS library) and sent back to the user along with the text format.

[1145] If the sentiment analysis determines that the user's condition is urgent, the server will implement measures to connect them to an expert, including using the system's pre-registered contact information for counselors and helplines to immediately connect them.

[1146] Terminal handling

[1147] The terminal supports the exchange of voice data between the user and the server. When the user speaks into the terminal, this voice is captured by the terminal. The terminal sends this voice data to the server as an HTTP request. When the terminal receives a response from the server, it displays this response in text format and plays it back in audio format. This allows the user to receive messages from the server both visually and audibly.

[1148] User Experience

[1149] Users can talk to their device to ask questions or ask about mental health or security. For example, if a user says, "I've been feeling anxious lately and can't sleep at night," the device sends this speech to the server. The server converts the speech data into text data and performs emotion analysis.

[1150] If the server determines that the user is anxious, it generates an empathetic response such as, "That's tough. Have there been any changes in your life recently?" This response is returned to the user in voice and text format.

[1151] When used as a security service, it can detect potential risks from an unspecified number of voice data. For example, if a heated argument is occurring nearby, the server can analyze the voice data and issue an alert such as, "Caution: A potential risk has been detected. Please check your surroundings."

[1152] An example of a prompt sentence could be, "You are a virtual security guard. Please analyze the text data below and evaluate the emotional state. If there are a lot of negative emotions, please return a warning."

[1153] The system allows users to receive prompt and professional mental health care and security services that can detect potential risks early from surrounding sounds.

[1154] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1155] Step 1:

[1156] The user speaks into the terminal. For example, the user asks a question or asks for advice, such as, "I've been feeling anxious lately and can't sleep at night." The input is the user's voice data, which is received by the terminal. The voice data is obtained as output.

[1157] Step 2:

[1158] The terminal sends the received voice data to the server as an HTTP request. At this time, the terminal converts the voice data into an appropriate data format for sending to the server. The input is the user's voice data, and the voice data converted into the HTTP request format is sent to the server as the output.

[1159] Step 3:

[1160] The server receives the HTTP request and retrieves the audio data. It then converts the audio data into text using a speech recognition library (e.g., speech_recognition). The input is the audio data, and the output is the converted text data.

[1161] Step 4:

[1162] The server uses the converted text and audio data to perform emotion analysis. The model uses a generative AI model based on Hugging Face's Transformer library. The input is text and audio data, and the output is an emotional state.

[1163] Step 5:

[1164] The server uses a generative AI model based on the emotional state to generate an appropriate response. This generated response is based on the user's psychological state and the question content. The input is the emotional state and text data, and the output is a response text.

[1165] Step 6:

[1166] The server converts the generated response text into audio using speech synthesis technology (e.g., gTTS), and also prepares response data in text format. The input is the generated response text, and audio data and text data are generated as output.

[1167] Step 7:

[1168] The server returns the response data to the terminal as an HTTP response. The input is voice data and text data, and the HTTP response is sent to the terminal as output.

[1169] Step 8:

[1170] The terminal displays and plays back the response data received from the server. Specifically, it displays text data on the screen and plays audio data through the speaker. The input is response data in HTTP response format, and the output presents visual and auditory information to the user.

[1171] 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.

[1172] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[1173] Server Processing

[1174] The server plays a central role in the system and operates as follows:

[1175] First, the server receives the voice data sent by the user. After receiving the voice data, the server uses voice recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information.

[1176] The server then inputs the text and voice data into an emotion engine to perform a detailed analysis of the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to assess the user's psychological state. Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response.

[1177] The generated responses are provided to the user in both text and audio formats. The server utilizes speech synthesis technology to convert text responses into audio. It also has the means to automatically connect to experts in emergencies based on the results of sentiment analysis.

[1178] As a concrete example, suppose a user says, "I've been feeling anxious and can't sleep lately." In this case, the device captures the audio and sends it to the server. The server converts the audio into text and analyzes it using an emotion engine. If the emotion engine determines that the user is in a high state of anxiety, the generative AI model will generate an empathetic response such as, "That's tough. Have there been any changes in your living environment recently?" Furthermore, if the emotional state is determined to be urgent, the server will immediately attempt to connect the user to a counselor or helpline.

[1179] Terminal handling

[1180] The device supports the exchange of voice data between the user and the server. When the user speaks to the device about their concerns, the device captures this voice and sends it to the server. When the device receives a response from the server, it provides this response to the user. Specifically, it displays the response in text format and plays it back in audio format at the same time. This allows the user to receive the response both visually and audibly.

[1181] User Experience

[1182] Users access the system through their device to receive mental health care. When the user speaks to the device about their condition and concerns, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[1183] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to get some rest. Why don't you try doing something to relax?" The response is provided to the user, and if necessary, an automatic connection to an expert is also made.

[1184] This system allows users to receive appropriate mental care anytime, anywhere, and enables early detection and treatment of psychological problems. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[1185] The processing flow will be explained below.

[1186] Server Processing

[1187] Step 1: Receive audio data

[1188] The server receives the voice data sent by the user through the terminal as an HTTP request, and the user's voice input is then incorporated into the system.

[1189] Step 2: Convert the audio data to text

[1190] The server uses a speech recognition library to convert the received voice data into text data, which converts the voice data into a format that can be processed as text information.

[1191] Step 3: Analyze the emotional state using the emotion engine

[1192] The server inputs the converted text and voice data into an emotion engine for detailed analysis of the user's emotional state. The emotion engine evaluates the tone of the voice and the content of the text to determine the user's psychological state.

[1193] Step 4: Generate a response

[1194] The server uses a generative AI model based on the analyzed emotional state and text data to generate an appropriate response, creating feedback tailored to the user's desired assistance.

[1195] Step 5: Generate a response in audio and text format

[1196] The server stores the generated response in text format and also generates it in audio format using speech synthesis technology, allowing the user to receive the response in both text and audio.

[1197] Step 6: Connect with an expert

[1198] Based on the analysis results of the emotion engine, if the server determines that the user's emotional state is urgent, it will automatically connect the user to a counselor, helpline, or other expert, allowing the user to receive professional assistance quickly.

[1199] Step 7: Send the response back to the user

[1200] The server generates text and audio responses and sends them back to the user, who can then receive the necessary assistance at their terminal.

[1201] Terminal handling

[1202] Step 1: Capture the user's voice input

[1203] When the user speaks into the terminal about the content of the consultation, the terminal uses a microphone to capture the voice, and the user's voice is thus captured by the terminal.

[1204] Step 2: Send the audio data to the server

[1205] The device compiles the captured audio data into an HTTP request and sends it to the server. This process allows the audio data to reach the server.

[1206] Step 3: Receive a response from the server

[1207] The generated response is sent back from the server and received by the device.

[1208] Step 4: Present the response to the user

[1209] The terminal displays the received response in text format and plays it back in audio format, allowing the user to understand the response visually and audibly.

[1210] User Experience

[1211] Step 1: Access the system

[1212] A user accesses the system using a terminal and prepares to receive mental health care.

[1213] Step 2: Use voice input

[1214] The user verbally explains their condition and concerns to the device, for example, "I've been feeling a lot of stress at work lately."

[1215] Step 3: Receive a response

[1216] The user receives responses from the server via their device, which are provided in both text and audio formats, allowing the user to receive feedback in the format that is most convenient for them.

[1217] Step 4: Connect with an expert

[1218] If the emotion engine determines that the user's emotional state is urgent, the user will be automatically connected to a professional counselor or helpline, allowing them to receive direct, professional assistance.

[1219] This system allows users to receive appropriate mental health care in real time, enabling early detection and treatment of psychological problems. In addition, by incorporating an emotion engine, it is possible to grasp the user's emotional state in more detail and provide advanced support based on the results.

[1220] Example 2

[1221] 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."

[1222] Conventional mental health care systems face challenges in accurately analyzing users' emotional states and providing appropriate responses based on those results. They often lack the ability to quickly connect users to specialists in emergencies. Furthermore, they lack a way to provide both visual and auditory responses to improve the user experience.

[1223] 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.

[1224] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and the voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, and means for inputting a prompt sentence to the generative model based on the emotion analysis result. This enables the system to analyze the user's emotional state in detail and provide an appropriate response in real time. Furthermore, in an emergency, it can quickly connect to an expert, improving the quality of support provided to the user.

[1225] The "means for receiving voice data" refers to software and hardware for acquiring voice uttered by the user in digital form and transmitting it to the server.

[1226] "Means for converting voice data into text data" refers to software technology for converting received voice data into text information using voice recognition technology.

[1227] "Means for analyzing emotional states from text data and voice data" refers to an emotion analysis engine that analyzes text and voice data and evaluates the user's emotions and psychological state.

[1228] "Means for generating a response using a generative model" refers to a process that uses an artificial intelligence model to generate an appropriate response based on the results of sentiment analysis.

[1229] "Means for outputting the generated response in audio and text form" refers to speech synthesis and display technologies for converting the generated text response into audio and also displaying it in text form.

[1230] "Means for connecting to an expert based on emotional state" refers to a communication means for automatically connecting to an expert or counselor when the user's emotional state is determined to be urgent.

[1231] "Means for inputting prompt sentences to a generative model based on the results of sentiment analysis" refers to an input means for giving appropriate instructions to a generative model based on the results of sentiment analysis.

[1232] This invention relates to a system that provides mental health care using generative AI and an emotion engine. It uses three main components: a server, a terminal, and a user, to provide appropriate mental health care to users in real time.

[1233] Server Processing

[1234] The server plays a central role in the system and has a means of receiving voice data. For example, when a user says, "I've been feeling anxious and can't sleep lately," the voice data is sent from the device to the server. This is done using a microphone as hardware and a real-time voice capture tool as software.

[1235] Next, the server uses speech recognition technology (for example, Google Cloud Speech-to-Text) to convert the voice data into text data. This process generates the text, "I've been feeling anxious lately and can't sleep."

[1236] The server inputs the generated text and voice data into an emotion engine, which uses NVIDIA Clara Guardian and Microsoft Azure Video Indexer to perform a detailed analysis of the user's emotional state. This allows the detection of high anxiety levels.

[1237] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate response, such as, "That's tough. Have there been any changes in your living situation recently?"

[1238] The generated responses are provided to the user in audio and text formats using speech synthesis technologies (e.g., Amazon Polly, Google Cloud Text-to-Speech), with the option to connect with an expert if the emotional state is deemed urgent.

[1239] Terminal handling

[1240] The device supports the exchange of voice data between the user and the server. When the user speaks about their worries into the device, the device captures this voice and sends it to the server. For example, the device captures the user's voice saying, "I've been anxious and can't sleep lately," and sends it digitally to the server.

[1241] When the device receives a response from the server, it provides the response to the user. Specifically, it displays the response "That's tough. Have there been any changes in your living environment recently?" in text format and plays it back in audio format. A text display tool and an audio playback tool are used.

[1242] User Experience

[1243] Users access the system through their device and receive mental health care. When the user says something like, "I've been feeling stressed lately because of a lot of work," this speech is sent from the device to the server. The server converts the speech into text and analyzes it using an emotion engine. Based on the analysis results, the generative AI model generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?"

[1244] This response is provided to the user, and if necessary, an automatic connection to a specialist is also made. This allows users to receive appropriate mental care anytime, anywhere. By combining it with an emotion engine, more accurate emotion analysis and response generation based on that analysis can be achieved, improving the quality of support for users.

[1245] Prompt Sentence Examples

[1246] A specific example of an input prompt sentence for a generative AI model is as follows:

[1247] If a user says, "I've been feeling anxious and can't sleep lately," we convert this speech into text and analyze it with the emotion engine. We want to generate an empathetic response like this:

[1248] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1249] Step 1:

[1250] The user inputs voice into the device. For example, they might say, "I've been feeling anxious lately and can't sleep." This voice data is captured by the device's microphone. The input is the user's voice data, and the output is a digital audio file. Specifically, the device records the voice data and processes it in real time.

[1251] Step 2:

[1252] The device sends audio data to the server. In this process, a digital audio file is sent to the server via the Internet. The input is the audio data captured by the device, and the output is the audio data transferred to the server. Specifically, the device uses a data transmission protocol to upload the audio data to the server.

[1253] Step 3:

[1254] The server receives the audio data. The server stores the audio data in buffer memory and prepares it for the next processing step. The input is the audio data sent from the terminal, and the output is the server's buffer memory where the audio data is stored. Specifically, the server receives an HTTP request and stores the audio data.

[1255] Step 4:

[1256] The server uses speech recognition technology to convert the voice data into text data. For example, Google Cloud Speech-to-Text is used. In this process, the audio file is converted into text in a natural language. The input is the stored voice data, and the output is text data. Specifically, the server makes an API call to convert the voice data into text.

[1257] Step 5:

[1258] The server inputs text and audio data into the emotion engine to analyze the emotional state. NVIDIA Clara Guardian and Microsoft Azure Video Indexer are used. In this step, the emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's emotional state. The input is text and audio data, and the output is the analysis result of the emotional state. Specifically, the server calls the emotion analysis API and obtains the results.

[1259] Step 6:

[1260] The server generates an appropriate response using a generative AI model (e.g., OpenAI GPT-4) based on the emotion analysis results. The input is the emotion analysis results, and the output is the generated response text. Specifically, the server inputs a prompt sentence into the generative AI model, causing it to generate an empathetic text response. An example of a prompt sentence is, "If a user says, 'I've been unable to sleep recently because I've been anxious,' we converted this speech into text and analyzed it with the emotion engine, which detected high levels of anxiety. We would like you to generate an empathetic response such as the following."

[1261] Step 7:

[1262] The server converts the generated text response into speech. Speech synthesis technology (e.g., Amazon Polly, Google Cloud Text-to-Speech) is used. The input is the generated text response, and the output is the speech response data. Specifically, the server calls an API that converts text data into speech data and generates an audio file.

[1263] Step 8:

[1264] The server sends the generated text and voice responses to the terminal. The input is the generated text and voice response data, and the output is the data sent to the terminal. In specific operations, the server uses a data transmission protocol to upload the text and voice data to the terminal.

[1265] Step 9:

[1266] The terminal receives the response from the server and provides it to the user. Specifically, it displays the generated response in text format and simultaneously plays it in audio format. The input is the text and audio response data sent from the server, and the output is the response presented to the user. Specifically, the terminal uses a text display tool and an audio playback tool to provide the response to the user visually and audibly.

[1267] Step 10:

[1268] In an emergency, the server uses a means to connect to an expert based on the user's emotional state. For example, if the user's emotional state is very unstable and requires an emergency, the server automatically attempts to connect to an expert. The input is the result of the judgment of the urgency of the emotional state, and the output is a connection to an expert. Specifically, the server uses a communication means to contact an expert or counselor.

[1269] This allows users to receive appropriate mental health care anytime, anywhere.

[1270] (Application example 2)

[1271] 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."

[1272] Employees' psychological state has a significant impact on their work, and sustained high levels of stress and anxiety can lead to security incidents and reduced work efficiency. Conventional systems have difficulty monitoring employees' psychological state in real time and detecting abnormalities early, often resulting in delayed implementation of appropriate security measures and responses. This invention aims to prevent security incidents and improve work efficiency by analyzing employees' emotional state using their voice data and immediately notifying the security team when an abnormality is detected, enabling a rapid response.

[1273] 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 receiving voice data, means for converting the voice data into text data, means for analyzing an emotional state from the text data and voice data, means for generating a response using a generative model, means for outputting the generated response in voice and text formats, means for connecting to an expert based on the emotional state, means for monitoring the emotional state and detecting an abnormal psychological state, and means for notifying a security team when the abnormal psychological state is detected. This makes it possible to monitor the psychological state of employees in real time and respond immediately when an abnormality is detected.

[1274] "Voice data" refers to data that is a digital recording of a user's spoken words or vocalizations.

[1275] "Text data" refers to data that includes character information converted from audio data.

[1276] "Emotional state" is information that represents the user's psychological state and feelings, and is analyzed from voice data and text data.

[1277] "Means for analyzing" is a technique or device for processing audio and text data and assessing the user's emotional state.

[1278] A "generative model" is a machine learning algorithm or artificial intelligence technique for generating responses based on an analyzed emotional state.

[1279] A "means for generating a response" is a technique or device that uses a generative model to create an appropriate response that corresponds to the user's emotional state.

[1280] "Output means" refers to a technique or device for providing the generated response in audio and text form.

[1281] "Means for connecting" refers to technology or devices that allow a user to connect with an expert or counselor when they are in an emergency psychological state.

[1282] "Monitoring means" refers to technology or devices that continuously observe the user's emotional state and detect abnormalities early on.

[1283] "Means for notifying" is a technology or device that sends an alert to the security team when an abnormal psychological state is detected.

[1284] A specific embodiment of the present invention will be described below. The following system analyzes emotional states using voice data, detects abnormal psychological states, and notifies the security team, thereby preventing security incidents and improving work efficiency.

[1285] Server Processing

[1286] The server plays a central role in this system. First, it receives the voice data sent by the user. Once the voice data is received, the server uses speech recognition technology to convert the voice data into text data, which makes it possible to process the voice data as text information. For speech recognition, Google Cloud Speech-to-Text API and Amazon Transcribe are used.

[1287] The server then inputs the text and voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the tone of the voice and the content of the text to evaluate the user's psychological state. This analysis is performed using IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[1288] Based on the emotional state obtained from the emotion engine, the server uses a generative AI model to generate an appropriate response. The generated response is provided to the user in both text and audio format. Response generation uses natural language generation modules such as OpenAI GPT-4.

[1289] Furthermore, if the emotional state is deemed abnormal, the server has the means to immediately notify the security team, which can be done via email or a chat system (e.g., Slack or Microsoft Teams).

[1290] Terminal handling

[1291] The device supports the exchange of voice data between the user and the server. When the user speaks a report into the device, the device captures this voice and sends it to the server. Specifically, this capture is performed using the smartphone's microphone and an appropriate voice capture module (iOS, Android).

[1292] Upon receiving a response from the server, the terminal provides the response to the user by displaying it in text format and playing it back in audio format, allowing the user to receive the response both visually and audibly.

[1293] User Experience

[1294] Users access the system through their devices and provide feedback on their emotional state. When the user speaks about their state into the device, this speech is sent to the server. The server converts the speech into text and analyzes it with an emotion engine to evaluate the user's emotional state in detail. The generative AI model then generates an appropriate response, which is sent back to the user in text and voice.

[1295] For example, if a user says, "I've been feeling stressed lately because of a lot of work," the server converts this speech into text and uses an emotion engine to analyze the level of stress. The generative AI model then generates advice such as, "That's tough. It's important to take a rest. Why don't you try doing something to relax?" If the emotional state is determined to be abnormal, the security team is immediately notified.

[1296] Example prompt sentence:

[1297] A user says, "I'm feeling very stressed because my project deadline is approaching."

[1298] An example prompt from a generative AI model: "The user's stress level is high. Please select the appropriate response."

[1299] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1300] Step 1:

[1301] Capture and transmit audio data

[1302] The user speaks into the smartphone terminal about their state of mind and emotions. The terminal captures this voice and stores it as digital audio data. The terminal then encrypts this audio data and transmits it to the server using a secure communication method (e.g., HTTPS).

[1303] Input: User's voice

[1304] Output: Encrypted digital audio data

[1305] Step 2:

[1306] Converting audio data to text

[1307] The server receives the voice data sent from the device and converts it into text using speech recognition technology. Specifically, it uses the Google Cloud Speech-to-Text API or Amazon Transcribe to convert the voice data into text.

[1308] Input: Encrypted digital audio data

[1309] Output: Text data

[1310] Step 3:

[1311] Emotional state analysis

[1312] The server inputs the obtained text data and the original voice data into an emotion engine to analyze the user's emotional state. The emotion engine analyzes the voice tone and text content to evaluate the user's psychological state. This analysis uses IBM Watson Tone Analyzer and Microsoft Azure Text Analytics.

[1313] Input: Text and audio data

[1314] Output: Parsed emotional state

[1315] Step 4:

[1316] Generating a response

[1317] The server uses a generative AI model based on the analyzed emotional state to generate an appropriate response, which is then prepared for delivery in both text and audio formats, using natural language generation modules such as OpenAI GPT-4.

[1318] Input: Parsed emotional state

[1319] Output: Text and audio responses

[1320] Step 5:

[1321] Emergency notifications and connection to experts

[1322] If the emotional state is deemed abnormal or urgent, the server immediately notifies the security team via email or chat systems (e.g., Slack, Microsoft Teams), and automatically connects the individual to a specialist or counselor if necessary.

[1323] Input: Parsed emotional state and generated response

[1324] Output: Notify security team and connect to experts

[1325] Step 6:

[1326] Providing a response to the user

[1327] The server generates a response and sends it back to the terminal, which displays the response in text format and plays it back in audio format, allowing the user to receive the response both visually and audibly.

[1328] Input: Text and audio responses

[1329] Output: Feedback to the user (text display and audio playback)

[1330] 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.

[1331] 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.

[1332] 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.

[1333] 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.

[1334] FIG. 9 is a diagram illustrating 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 actions 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.

[1335] 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.

[1336] 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).

[1337] 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.

[1338] 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."

[1339] 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.

[1340] 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).

[1341] 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.

[1342] 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.

[1343] 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.

[1344] 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.

[1345] 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.

[1346] 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.

[1347] 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.

[1348] 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.

[1349] 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.

[1350] 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.

[1351] The following is further disclosed regarding the above embodiment.

[1352] (Claim 1)

[1353] means for receiving audio data;

[1354] means for converting the voice data into text data;

[1355] means for analyzing an emotional state from the text data and the voice data;

[1356] means for generating a response using the generative model;

[1357] means for outputting the generated response in audio and text formats;

[1358] means for connecting with an expert based on said emotional state;

[1359] A system including:

[1360] (Claim 2)

[1361] 10. The system of claim 1, wherein if the emotional state is determined to be an emergency, an expert is automatically connected.

[1362] (Claim 3)

[1363] 10. The system of claim 1, further comprising means for presenting the generated response to the user in text and audio form.

[1364] "Example 1"

[1365] (Claim 1)

[1366] means for receiving audio data using an audio input device;

[1367] means for converting the voice data into text data using voice recognition technology;

[1368] means for analyzing an emotional state from the text data and the voice data using an emotion analysis algorithm;

[1369] means for generating a response based on the text data and the emotional state using a generative AI model;

[1370] means for generating and outputting the generated response in voice and text formats using speech synthesis technology;

[1371] means for automatically connecting to an expert if said emotional state satisfies certain conditions;

[1372] A system including:

[1373] (Claim 2)

[1374] 10. The system of claim 1, wherein if the emotional state is determined to be an emergency, an expert is automatically connected.

[1375] (Claim 3)

[1376] 10. The system of claim 1, further comprising means for presenting the generated response to the user in text and audio form.

[1377] "Application Example 1"

[1378] (Claim 1)

[1379] means for receiving audio data;

[1380] means for converting the voice data into text data;

[1381] means for analyzing an emotional state from the text data and the voice data;

[1382] means for generating a response using the generative model;

[1383] means for outputting the generated response in audio and text formats;

[1384] means for connecting with an expert based on said emotional state;

[1385] A means of capturing an unspecified number of voice data and performing emotion analysis to detect potential risks;

[1386] means for issuing an alarm when the potential risk is detected;

[1387] A system including:

[1388] (Claim 2)

[1389] 10. The system of claim 1, wherein if the emotional state is determined to be an emergency, an expert is automatically connected.

[1390] (Claim 3)

[1391] 10. The system of claim 1, further comprising means for presenting the generated response to the user in text and audio form.

[1392] "Example 2: Combining Emotion Engines"

[1393] (Claim 1)

[1394] means for receiving audio data;

[1395] means for converting the voice data into text data;

[1396] means for analyzing an emotional state from the text data and the voice data;

[1397] means for generating a response using the generative model;

[1398] means for outputting the generated response in audio and text formats;

[1399] means for connecting with an expert based on said emotional state;

[1400] a means for inputting a prompt sentence into the generative model based on the sentiment analysis result;

[1401] A system including:

[1402] (Claim 2)

[1403] 10. The system of claim 1, wherein if the emotional state is determined to be an emergency, an expert is automatically connected.

[1404] (Claim 3)

[1405] 10. The system of claim 1, further comprising means for presenting the generated response to the user in text and audio form.

[1406] "Application example 2 when combining emotion engines"

[1407] (Claim 1)

[1408] means for receiving audio data;

[1409] means for converting the voice data into text data;

[1410] means for analyzing an emotional state from the text data and the voice data;

[1411] means for generating a response using the generative model;

[1412] means for outputting the generated response in audio and text formats;

[1413] means for connecting with an expert based on said emotional state;

[1414] means for monitoring said emotional state and detecting abnormal psychological states;

[1415] means for notifying a security team when said abnormal psychological state is detected;

[1416] A system including:

[1417] (Claim 2)

[1418] 10. The system of claim 1, wherein if the emotional state is determined to be an emergency, an expert is automatically connected.

[1419] (Claim 3)

[1420] 10. The system of claim 1, further comprising means for presenting the generated response to the user in text and audio form. [Explanation of symbols]

[1421] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving audio data; means for converting the voice data into text data; means for analyzing an emotional state from the text data and voice data; means for generating a response using the generative model; means for outputting the generated response in audio and text formats; means for connecting with an expert based on said emotional state; A system including:

2. 10. The system of claim 1, wherein if the emotional state is determined to be an emergency, an expert is automatically connected.

3. 10. The system of claim 1, further comprising means for presenting the generated response to a user in text and audio form.

Images