System

Abstract

A system is provided.SOLUTION: A system comprising: means for capturing information input by a user by voice; means for converting the captured voice data into text data; means for analyzing an intention of the user based on the text data; means for acquiring related information from a database based on the analysis result; means for generating an appropriate advice for the user based on the acquired information; means for converting the generated advice from text data into voice data; and means for providing the converted voice data to the user.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, while technological innovation continues to advance, an increasing number of people are unable to effectively utilize information and are becoming isolated. Furthermore, the current situation is one in which inequality in information is widening the gap between rich and poor. In particular, there are many situations in which it is difficult to make appropriate decisions or obtain appropriate advice in daily life. To solve these issues, there is a growing need for systems that provide useful information for daily life through self-interaction and support optimal decision-making. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including: means for capturing information input by voice from a user; means for converting the captured voice data into text data; means for analyzing the user's intent based on the text data; means for retrieving related information from a database based on the analysis results; means for generating appropriate advice for the user based on the retrieved information; means for converting the generated advice from text data into voice data; and means for providing the converted voice data to the user. Furthermore, the means for converting voice data into text data utilizes a voice recognition engine accessible via a network, thereby improving processing performance. Furthermore, the means for analyzing the user's intent utilizes a natural language processing engine to extract keywords and associations, thereby enabling more accurate advice to be provided.

[0006] A "user" is a person who uses the system to input voice and receive advice.

[0007] "Information entered by voice" refers to questions or inquiries that a user communicates to the system through voice.

[0008] "Capturing means" refers to technology or devices that record the user's voice and store it as digital data.

[0009] "Audio data" means a digital representation of a captured audio signal.

[0010] "Text data" refers to textual information converted from voice data using voice recognition technology.

[0011] "Means for converting into text data" refers to technology or software that analyzes voice data and generates corresponding text data.

[0012] "Means for analyzing user intent" refers to technology or software that analyzes text data and extracts key keywords and intent from it.

[0013] "Relevant information" refers to data necessary to provide appropriate answers or advice to the user's intent or question.

[0014] A "database" is a system that stores information and data in an organized manner and allows for quick search and retrieval.

[0015] "Means of acquisition" refers to the technology or system that searches for and extracts the necessary information from the database.

[0016] "Means for generating appropriate advice" refers to technology or software that generates optimal advice for users in written form based on acquired information.

[0017] "Means for converting into audio data" refers to the technology or software that converts the generated text data into audio format.

[0018] "Means for providing" refers to the technology or system that transmits and plays back the generated audio data to the user.

[0019] A "voice recognition engine" is an algorithm or software that analyzes voice data and converts it into text data.

[0020] A "natural language processing engine" is an algorithm or software that analyzes text data, understands its meaning, and extracts keywords and intent. [Brief explanation of the drawings]

[0021] [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

[0022] 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.

[0023] First, the terms used in the following description will be explained.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 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.

[0032] 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).

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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."

[0042] This invention is a system that provides appropriate advice based on information input by voice by the user. The system is designed to be easy to use by users through a smartphone app.

[0043] Basic system configuration

[0044] The system consists of the following main components:

[0045] 1. Audio input capture means (terminal)

[0046] 2. Speech recognition means (server)

[0047] 3. Natural language processing means (server)

[0048] 4. Database access method (server)

[0049] 5. Dialogue Generation Means (Server)

[0050] 6. Audio output means (terminal)

[0051] Program processing overview

[0052] The specific processing flow of the system will be explained in natural language below.

[0053] Acquiring voice input

[0054] The user starts the smartphone app and inputs a question or request by voice. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This voice is captured by the device and saved as digital audio data.

[0055] Converting audio data to text

[0056] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[0057] Intention analysis using natural language processing

[0058] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[0059] Retrieving information from a database

[0060] The server retrieves relevant information from a database based on the result of the intent analysis, for example, by issuing a query to retrieve information about fatigue reduction, and then retrieves the result.

[0061] Dialogue generation

[0062] The server generates appropriate advice for the user based on the acquired information. For example, it might say, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0063] Text-to-speech conversion

[0064] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0065] Audio output

[0066] Finally, the device plays back the audio data and provides advice to the user, who can receive appropriate advice through audio and use it in their daily lives.

[0067] Specific examples

[0068] For example, if a user asks "Inner Self F" a question like, "I've been so busy at work lately that I'm feeling stressed. What should I do?", the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "stress" and "ways to relieve stress," and retrieves useful ways to relieve stress from a database. Advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest" is generated and provided to the user via voice. In this way, users can easily obtain useful information and apply it to their daily lives.

[0069] As described above, this system realizes a series of processes that starts with voice input and provides appropriate advice to the user by voice.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[0073] Step 2:

[0074] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[0075] Step 3:

[0076] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[0077] Step 4:

[0078] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[0079] Step 5:

[0080] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[0081] Step 6:

[0082] The server queries the database to retrieve relevant information based on the analysis results, for example, recommendations for fatigue recovery.

[0083] Step 7:

[0084] Based on the information acquired by the server, appropriate advice is generated for the user. For example, it might generate a sentence such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0085] Step 8:

[0086] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[0087] Step 9:

[0088] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[0089] Step 10:

[0090] The terminal plays back the received voice data and provides the user with the advice by voice. The played back voice conveys appropriate advice to the user.

[0091] Specific examples

[0092] The process after a user asks, "I've been feeling tired lately. What should I do?" is as follows:

[0093] Step 1: The user launches the app and says, "I've been feeling tired a lot lately. What should I do?"

[0094] Step 2: The device records this audio and saves it as audio data.

[0095] Step 3: The device sends the audio data to the server.

[0096] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[0097] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0098] Step 6: The server queries the database to obtain information about fatigue recovery.

[0099] Step 7: Based on the information obtained, the server generates advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0100] Step 8: The text data generated by the server is sent to the speech synthesis engine and converted into speech data.

[0101] Step 9: The server sends the audio data to the terminal.

[0102] Step 10: The terminal plays back the audio data and provides the advice to the user by voice.

[0103] This series of steps allows the user to easily obtain appropriate advice through voice input.

[0104] Example 1

[0105] 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."

[0106] In conventional systems, the process of providing appropriate advice to users based on voice input is complicated, making it difficult to accurately interpret the user's intentions and provide appropriate information. In particular, there were issues with the accuracy of speech recognition and natural language processing, making it impossible to quickly and accurately provide the information users were looking for. In addition, there was also the problem that when the advice generated was not in a natural dialogue format, it was difficult for users to understand the advice.

[0107] 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.

[0108] In this invention, the server includes means for converting voice data into text data, means for analyzing the user's intention based on the text data, means for retrieving related information from a database based on the analysis result, and means for using a generative AI model for generating generated advice information in a natural dialogue format. This makes it possible to accurately analyze the user's intention from voice input, quickly and accurately provide related information, and convey advice to the user in a natural dialogue format.

[0109] "User" refers to an individual who uses the system and provides voice input.

[0110] "Voice input information" refers to what a user speaks through a microphone using a smartphone app or digital device.

[0111] "Capturing means" refers to a device, such as a smartphone or microphone, that has the ability to capture audio input and store it as digital audio data.

[0112] "Voice data" refers to data in which information input by a user through voice is stored in digital form.

[0113] "Text data" refers to digital data that has been converted from audio data into text format.

[0114] "Speech recognition engine" refers to software or algorithms for converting voice data into text data.

[0115] "Means for analyzing user intent" refers to natural language processing technology that extracts important keywords and context from text data and analyzes their meaning.

[0116] A "natural language processing engine" refers to technology and software that analyzes text data and extracts user intent and keywords.

[0117] A "database" refers to a structured data storage for storing various information and providing relevant information in response to a query.

[0118] A "generative AI model" refers to an algorithm or software that generates natural-looking, conversational text based on input data.

[0119] "Means for obtaining related information" refers to a function for searching and obtaining appropriate information from a database based on the results of analyzing the user's intentions.

[0120] "Means for generating appropriate advice" refers to technologies and algorithms for automatically generating advice that is useful to users based on acquired information.

[0121] "Speech synthesis engine" refers to software or algorithms for converting text data into speech data.

[0122] "Means for providing to the user" refers to a device or system for playing back the generated audio data so that the user can hear it.

[0123] This invention relates to a system that provides appropriate advice based on information input by a user via voice. The system is designed to be easily accessible to users through a smartphone app. The system consists of the following main hardware and software components:

[0124] Basic system configuration

[0125] The system consists of the following main components:

[0126] 1. Audio input capture means (terminal)

[0127] The user starts a dedicated smartphone app and inputs voice. The device captures the voice through the microphone and saves it as digital voice data. This voice data is then stored in temporary storage on the smartphone.

[0128] 2. Speech recognition means (server)

[0129] The device compresses the captured voice data and sends it to the server, where a speech recognition engine such as the Google® Speech-to-Text API runs and converts the voice data into text data, which is then stored in a back-end database.

[0130] 3. Natural language processing means (server)

[0131] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for analysis. The engine extracts important keywords and user intent from the text. For example, the keywords "fatigue builds up" and "measures" are extracted. Analysis is also performed to take into account the user's emotions and context.

[0132] 4. Database access method (server)

[0133] Based on the result of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database, a health food database) to obtain relevant information. For example, it obtains "information on methods for reducing fatigue" from the database. This information is used for further processing.

[0134] 5. Dialogue Generation Means (Server)

[0135] The server uses a generative AI model (e.g., GPT-3 (registered trademark)) to generate advice in a natural conversational format based on information retrieved from the database. For example, specific advice might be generated, such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0136] 6. Audio output means (terminal)

[0137] The generated text data of the advice is sent to a speech synthesis engine (e.g., Google Text-to-Speech) on the server and converted into audio data. This audio data is then sent to the device, where it is played back through the device's speaker to provide the advice to the user.

[0138] Specific examples

[0139] For example, a user might ask their "inner self" a question such as, "Work has been so busy lately that I'm feeling stressed. What should I do?" This system captures the user's voice and converts it into text using speech recognition. Keywords such as "stress" and "ways to relieve stress" are extracted using natural language processing, and the server retrieves information about stress relief from a database. Based on the information retrieved, the server generates advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest," which the device then provides to the user via voice.

[0140] Prompt Sentence Examples

[0141] Example prompts for generative AI models:

[0142] "My work is so busy that I feel like I'm getting stressed. Can you tell me some effective ways to relieve stress?"

[0143] In this way, the system embodies a process for providing appropriate advice to the user based on voice input.

[0144] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0145] Step 1:

[0146] The user launches a dedicated smartphone app and inputs voice data. For example, they might say, "I've been feeling tired lately. What should I do?" This voice is captured by the device and saved as digital voice data. The user's voice is the input, and digital voice data is obtained as the output.

[0147] Step 2:

[0148] The terminal compresses the captured audio data and sends it to the server. The input is digital audio data, and the output is compressed audio data. The compressed audio data is transferred to the server via the Internet.

[0149] Step 3:

[0150] On the server, a speech recognition engine (e.g., Google Speech-to-Text API) receives the compressed voice data and converts it into text data. The input is the compressed voice data, and the output is the text data converted from the voice. For example, the voice saying "I've been feeling tired a lot lately. What should I do?" is converted into text data.

[0151] Step 4:

[0152] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for intent analysis. The input is text data, and the output is extracted keywords and intent. For example, the key keywords extracted are "fatigue builds up" and "measures."

[0153] Step 5:

[0154] Based on the results of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database or a health food database) to obtain related information. The input is the intent analysis result (keywords), and the output is related information. For example, data containing "methods for reducing fatigue" is obtained.

[0155] Step 6:

[0156] The server uses a generative AI model (e.g., GPT-3) to generate appropriate advice based on the acquired information. The input is relevant information acquired from the database, and the output is advice text in a natural conversational format. For example, the generated advice might be, "We recommend that you eat a diet rich in vitamin B, get at least 7-8 hours of quality sleep every day, and engage in light exercise."

[0157] Step 7:

[0158] The generated text data of the advice is converted into voice data by a speech synthesis engine (e.g., Google Text-to-Speech) on the server. The input is the text data of the advice, and the output is voice data. Once the voice data is generated, it is sent from the server to the device.

[0159] Step 8:

[0160] The terminal plays the received voice data through a speaker and provides advice to the user. The input is the voice data sent from the server, and the output is voice information that the user can hear. The user can receive appropriate advice through voice.

[0161] Through this series of steps, users can easily input their voice and receive appropriate advice via voice.

[0162] (Application example 1)

[0163] 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."

[0164] Conventional speech recognition systems have difficulty accurately analyzing user intent and providing relevant information, making it particularly difficult to suggest meal menus and restaurants that suit a user's preferences in the food delivery field. Furthermore, the inability to provide intuitive and prompt advice based on the voice information entered by the user creates a problem that impairs the user experience.

[0165] 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.

[0166] In this invention, the server includes a means for capturing voice-input information, a means for converting the information into text data, and a means for analyzing the user's intent, thereby smoothly generating suggestions suitable for food delivery and enabling the user to quickly and easily obtain appropriate meal menus and restaurant information.

[0167] A "means for capturing user voice input information" is a mechanism for obtaining user spoken voice data and storing it in digital form.

[0168] The "means for converting captured voice data into text data" is a mechanism for analyzing acquired voice data and converting it into corresponding text data.

[0169] "Means for analyzing user intent based on text data" refers to a mechanism that interprets the user's requests and intent from text data and extracts appropriate information based on that.

[0170] The "means for obtaining related information from a database based on the analysis results" is a mechanism for searching and obtaining information corresponding to a user request from a database.

[0171] The "means for generating appropriate advice for the user based on the acquired information" is a mechanism for automatically generating advice to be provided to the user based on the acquired information.

[0172] The "means for converting the generated advice from text data to voice data" is a mechanism for converting the generated advice in text format into voice format.

[0173] The "means for generating suggestions suitable for food delivery and providing feedback to the user" is a mechanism that generates suggestions for meal menus and restaurants suitable for food delivery based on requests input by the user via voice and communicates the results to the user.

[0174] This invention is a system that provides food delivery suggestions based on information input by voice from a user. The system includes the following main components:

[0175] Basic system configuration

[0176] 1. Audio input capture means (terminal)

[0177] 2. Speech recognition means (server)

[0178] 3. Natural language processing means (server)

[0179] 4. Database access method (server)

[0180] 5. Dialogue Generation Means (Server)

[0181] 6. Audio output means (terminal)

[0182] Program processing overview

[0183] Acquiring voice input

[0184] A user launches a smartphone app and speaks to ask a question or request a dish, such as "I want a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[0185] Converting audio data to text

[0186] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[0187] Intention analysis using natural language processing

[0188] The server then sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text. For example, key keywords like "healthy" and "lunch" are extracted.

[0189] Retrieving information from a database

[0190] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[0191] Dialogue generation

[0192] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[0193] Text-to-speech conversion

[0194] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0195] Audio output

[0196] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[0197] Specific examples

[0198] For example, if a user requests, "I want a light and healthy dinner," the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "light," "healthy," and "dinner," and retrieves corresponding menu and restaurant information from a database. A suggestion is generated and delivered to the user via voice: "How about a salad bowl and steamed fish set? You can order this at a nearby restaurant." In this way, users can easily obtain useful information and encourage their use of food delivery services.

[0199] Prompt Sentence Examples

[0200] markdown

[0201] Prompt statement

[0202] Your role is to provide an assistant that suggests appropriate meal options based on the meal requests entered by the user. Please analyze the following requests and suggest the corresponding menu options.

[0203] Question: I want a healthy lunch today.

[0204] Suggestion: How about a salad and grilled chicken combo for a healthy lunch?

[0205] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0206] Step 1:

[0207] A user launches a smartphone app and speaks to ask a question or describe a desired meal, for example, "I want to eat a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[0208] Input: User's voice data

[0209] Output: Digital audio data

[0210] Step 2:

[0211] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[0212] Input: Digital audio data

[0213] Output: Text data

[0214] Step 3:

[0215] The server sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text data. For example, key keywords like "healthy" and "lunch" are extracted.

[0216] Input: Text data

[0217] Output: Extracted keywords and intent

[0218] Step 4:

[0219] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[0220] Input: Extracted keywords and intent

[0221] Output: Information retrieved from the database

[0222] Step 5:

[0223] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[0224] Input: Information retrieved from the database

[0225] Output: Generated suggested text

[0226] Step 6:

[0227] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0228] Input: Generated suggested text

[0229] Output: Audio data

[0230] Step 7:

[0231] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[0232] Input: Audio data

[0233] Output: Providing spoken suggestions

[0234] 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.

[0235] This invention is a system that provides appropriate advice based on information input by voice by a user. By further combining this system with an emotion engine that recognizes the user's emotions, it is possible to provide more personalized advice based on emotions.

[0236] Basic system configuration

[0237] The system consists of the following main components:

[0238] 1. Audio input capture means (terminal)

[0239] 2. Speech recognition means (server)

[0240] 3. Natural language processing means (server)

[0241] 4. Emotion Recognition Method (Server)

[0242] 5. Database access method (server)

[0243] 6. Dialogue Generation Means (Server)

[0244] 7. Audio output means (terminal)

[0245] Program processing overview

[0246] The specific processing flow of the system will be explained in natural language below.

[0247] Acquiring voice input

[0248] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This speech is captured by the device and saved as digital audio data.

[0249] Converting audio data to text

[0250] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[0251] Intention analysis using natural language processing

[0252] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[0253] Emotion Recognition Processing

[0254] The server analyzes the user's emotions using emotion recognition means, using voice data and, if possible, facial expression data. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone and strength of the voice, choice of words, etc.

[0255] Retrieving information from a database

[0256] The server retrieves relevant information from the database based on the results of the intent analysis and emotion recognition. For example, it issues a query to retrieve information on fatigue reduction and stress relief, and retrieves the results.

[0257] Dialogue generation

[0258] The server generates appropriate advice for the user based on the acquired information and the user's emotions. For example, the server might generate advice such as, "I recommend that you eat a diet rich in vitamin B and get at least seven to eight hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[0259] Text-to-speech conversion

[0260] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0261] Audio output

[0262] Finally, the device plays back the audio data and provides the user with advice by voice, allowing the user to receive appropriate advice through voice and use it in their daily lives.

[0263] Specific examples

[0264] For example, if a user asks, "I'm so stressed out at work these days, what should I do?", the following steps are taken:

[0265] 1. The user speaks a question and the device records this speech.

[0266] 2. The device sends the recorded audio data to the server.

[0267] 3. The server uses a speech recognition engine to convert the voice data into text data.

[0268] 4. The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0269] 5. The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[0270] 6. The server queries the database to obtain information about stress relief.

[0271] 7. Based on the information obtained, the server generates the following advice: "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[0272] 8. The text data generated by the server is sent to the speech synthesis engine and converted into voice data.

[0273] 9. The server sends the audio data to the device.

[0274] 10. The terminal plays the audio data and provides advice to the user by voice.

[0275] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[0276] The processing flow will be explained below.

[0277] Step 1:

[0278] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[0279] Step 2:

[0280] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[0281] Step 3:

[0282] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[0283] Step 4:

[0284] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[0285] Step 5:

[0286] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[0287] Step 6:

[0288] The server uses voice data and facial expression data (if available) to analyze the user's emotions using an emotion engine. Emotions such as "fatigue," "stress," and "depression" are recognized from the tone and strength of the voice, as well as the choice of words.

[0289] Step 7:

[0290] The server queries the database and retrieves relevant information based on the analysis results, for example, recommendations for fatigue recovery and stress relief.

[0291] Step 8:

[0292] Based on the information acquired by the server, appropriate advice is generated for the user. Taking into account emotions, the advice generated is, "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[0293] Step 9:

[0294] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[0295] Step 10:

[0296] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[0297] Step 11:

[0298] The device plays back the received voice data and provides the user with advice in the form of voice. The played back voice conveys appropriate and emotionally sensitive advice to the user.

[0299] Specific examples

[0300] The process after a user asks, "I'm so stressed out at work these days, what should I do?":

[0301] Step 1: The user launches the app and says, "Work has been so busy lately that I'm feeling stressed. What should I do?"

[0302] Step 2: The device records this audio and saves it as audio data.

[0303] Step 3: The device sends the audio data to the server.

[0304] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[0305] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0306] Step 6: The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[0307] Step 7: The server queries the database to obtain information about stress relief.

[0308] Step 8: Based on the information obtained, the server generates the following advice: "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[0309] Step 9: The text data generated by the server is sent to a speech synthesis engine and converted into speech data.

[0310] Step 10: The server sends the voice data to the terminal.

[0311] Step 11: The terminal plays back the audio data and provides the advice to the user by voice.

[0312] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[0313] Example 2

[0314] 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."

[0315] Conventional voice input systems can only provide general responses to information entered by the user through voice, making it difficult to provide personalized advice based on the user's emotions and intentions. This often leaves users frustrated because they are unable to receive advice that is best suited to their situation. Furthermore, there are also issues with the accuracy of converting voice data into text data and the accuracy of the results of analyzing text data.

[0316] 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.

[0317] In this invention, the server includes a means for recognizing a user's emotion, a means for acquiring related information from a database based on the analysis result and the recognized emotion, and a means for generating appropriate advice for the user using a generative AI model based on the acquired information and the user's emotion, thereby making it possible to provide personalized advice that takes into account the user's emotion and intention.

[0318] "User" refers to a person who uses the system to input voice and receive advice.

[0319] "Voice input capture means" refers to a device or method that captures and stores a user's voice as digital audio data.

[0320] "Means for converting voice data into text data" refers to a device or method for converting captured voice data into text information, such as a voice recognition engine.

[0321] "Means for analyzing user intent based on text data" refers to a device or method for understanding a user's intent or main point from text data.

[0322] "Means for recognizing user emotions" refers to devices and methods that analyze a user's emotional state from information such as voice and facial expressions.

[0323] "Means for retrieving relevant information from a database based on the analysis result and the recognized emotion" refers to a device or method for searching and retrieving relevant information from a database based on the analyzed intention and the recognized emotion.

[0324] "Means for generating appropriate advice for a user using a generative AI model based on acquired information and the user's emotions" refers to a device or method for creating appropriate advice using a generative AI model, taking into account acquired information and the user's emotions.

[0325] The "means for converting text data into voice data" refers to a device or method for converting the text data of the generated advice into voice data.

[0326] The "means for providing the user with converted voice data" refers to a device or method for conveying the generated advice to the user as voice data.

[0327] The present invention provides a system for providing personalized advice based on information input by a user through speech, which includes, as its main components, a speech input capture unit, a speech recognition unit, a natural language processing unit, an emotion recognition unit, a database access unit, a dialogue generation unit, and a speech output unit.

[0328] Acquiring voice input

[0329] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" The device (smartphone) captures and saves this voice as digital audio data.

[0330] Converting audio data to text

[0331] The device sends the captured voice data over the Internet to a server, which then converts the voice data into text using a network-accessible speech recognition engine (e.g., Google Cloud Speech-to-Text).

[0332] Intention analysis using natural language processing

[0333] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) for analysis. By identifying important keywords and the user's intent from the analysis results, the keywords "fatigue builds up" and "measures" are extracted.

[0334] Emotion Recognition Processing

[0335] The server uses an emotion recognition engine on the voice data to analyze the user's emotions. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone of the voice and the words chosen.

[0336] Retrieving information from a database

[0337] Based on the results of intent analysis and emotion recognition, the server issues a query to retrieve relevant information from the database, for example, information on "fatigue reduction" and "stress relief."

[0338] Dialogue generation

[0339] The server considers the acquired information and the user's emotions and generates appropriate advice for the user using a generative AI model (e.g., OpenAI (registered trademark) GPT-3). For example, it generates text data such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[0340] Prompt Sentence Examples

[0341] In response to a user's question, "Work has been so busy lately that I'm feeling stressed. What should I do?", generate advice when the emotion recognition result includes "stress." Include advice on specific dietary choices, relaxation methods, and exercise.

[0342] Text-to-speech conversion

[0343] The server sends the generated advice text data to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data.

[0344] Audio output

[0345] The converted voice data is sent to the terminal, which plays it back to provide the user with advice in the form of voice. This allows the user to receive appropriate advice through voice and use it in their daily lives.

[0346] This system allows users to receive personalized advice based on their emotions and intentions, resulting in a more satisfying user experience.

[0347] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0348] Step 1:

[0349] The user launches the smartphone app and speaks into the microphone to ask a question or ask for advice. The spoken voice contains specific details such as, "I've been feeling tired a lot lately. What should I do?" The device captures and saves this voice as digital voice data. The input is the user's voice data, and the output is digital voice data.

[0350] Step 2:

[0351] The device sends the captured voice data to a server via the Internet. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data. In this step, the input is voice data and the output is text data.

[0352] Step 3:

[0353] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the intent of the text. The analysis identifies important keywords and the user's intent. For example, the keywords "fatigue builds up" and "measures" are extracted. The input is the text data, and the output is the analysis results of the keywords and intent.

[0354] Step 4:

[0355] The server sends the voice data to an emotion recognition engine to analyze the user's emotions. Emotion recognition uses the tone of the voice and the choice of words, so emotions such as "fatigue," "stress," and "depression" can be recognized. The input is the voice data, and the output is the recognized emotion data.

[0356] Step 5:

[0357] Based on the results of intent analysis and emotion recognition, the server queries a database to retrieve relevant information. The database contains information on fatigue reduction and stress relief. The input is the analysis results and emotion data, and the output is relevant information.

[0358] Step 6:

[0359] The server takes into account the acquired information and the user's emotions and inputs prompt sentences into a generative AI model (e.g., OpenAI GPT-3) to generate appropriate advice for the user. For example, text data such as "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation techniques into your routine" is generated. The input is the acquired information and emotion data, and the output is the text data of the advice.

[0360] Step 7:

[0361] The server sends the generated text data of advice to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data. The input is the text data of advice, and the output is voice data.

[0362] Step 8:

[0363] The converted voice data is sent to the terminal, which then plays it back. This allows the user to receive advice by voice and put it into practice in their daily lives. The input is voice data, and the output is the played-back voice.

[0364] (Application example 2)

[0365] 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."

[0366] Conventional food delivery services have struggled to provide personalized meal recommendations based on the user's emotions and specific circumstances. This has resulted in a lack of convenience for users to select meals that best suit their mood and health. Furthermore, methods for improving convenience through voice input and output have also been inadequate.

[0367] The specific processing by the specific 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 analyzing the user's emotions using emotion recognition means, means for generating personalized meal suggestions based on the user's emotions, and means for providing the personalized meal suggestions to the user by audio output. This enables personalized meal suggestions to be made according to the user's emotions and individual circumstances.

[0368] A "user" is an individual who wishes to use the system to receive meal suggestions.

[0369] "Audio input" refers to audio data that a user inputs into the system using a voice capture device such as a microphone.

[0370] A "speech recognition engine" is a technical device or software that converts voice input into text data.

[0371] A "natural language processing engine" is a technical device or software that analyzes user intent and keywords from text data.

[0372] An "emotion recognition means" is a technical device or software that analyzes a user's voice or text data to identify their emotions.

[0373] A "database" is a technical device or system for storing information that stores acquired information and provides required data based on a query.

[0374] The "dialogue generating means" is a technical device or software that generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions.

[0375] A "speech synthesis engine" is a technical device or software that converts text data into speech data.

[0376] A "meal suggestion" is a set of advice to provide the user with suitable meal options based on the user's emotions and situation.

[0377] This invention is a system that analyzes emotions based on information input by a user's voice and provides personalized meal suggestions. The system is composed of the following main components:

[0378] 1. Voice input capture method: The user uses a smartphone or other voice input device to input questions or requests by voice.

[0379] 2. Speech recognition means: Converts the voice data captured by the voice input capture means into text data. This process is performed using a voice recognition engine accessible via a network (e.g., Google Cloud Speech-to-Text).

[0380] 3. Natural language processing: Analyze user intent and keywords from text data. This process is performed using a natural language processing engine (e.g., SpaCy).

[0381] 4. Emotion Recognition: Analyzes the user's voice and text data to identify their emotions. This process is performed using generative AI models for emotion recognition (e.g., Transformers in HuggingFace).

[0382] 5. Database access method: Based on the analysis results, relevant information is retrieved from a database, which contains information about specific dishes and ingredients.

[0383] 6. Dialogue generation means: Based on the acquired information and the user's emotions, appropriate advice and meal suggestions are generated for the user.

[0384] 7. Voice output means: The generated advice is converted from text data to voice data. This process is performed using a voice synthesis engine (e.g., pyttsx3).

[0385] System operation explanation

[0386] Hardware and Software Requirements

[0387] Hardware: Smartphone or personal computer with microphone

[0388] Software: Python programs, SpeechRecognition, Pyttsx3, SpaCy, Transformers

[0389] Data processing and calculation

[0390] 1. Acquiring voice input: The microphone of a smartphone or personal computer is used to acquire the user's voice, which is then saved as digital voice data by the voice input capture means.

[0391] 2. Speech-to-text conversion: Using a speech recognition engine, the captured speech is converted into text, which is then used for further processing.

[0392] 3. Intent analysis using natural language processing: Analyze text data using a natural language processing engine to extract key keywords and their relevance.

[0393] 4. Emotion recognition processing: Emotion recognition means are used to analyze the user's emotions from voice and text data.

[0394] 5. Information retrieval from the database: Based on the analyzed intentions and emotions, information is retrieved from the database to make appropriate meal suggestions.

[0395] 6. Dialogue generation: Generate personalized meal suggestions based on the acquired information and user emotions.

[0396] 7. Voice output: The generated meal suggestions are converted into voice data using a voice synthesis engine and provided to the user audibly.

[0397] Specific examples

[0398] If a user speaks "What should I eat today?", the system will act as follows:

[0399] The speech recognition engine converts the speech to text, generating the text "What should I eat today?"

[0400] A natural language processing engine analyzes this text and extracts the request "what to eat" as a keyword.

[0401] The emotion recognition means identifies the user's emotion as "neutral" from the voice data.

[0402] A database access means retrieves meal suggestions based on "neutral" emotions from the database.

[0403] The dialogue generator generates appropriate dietary suggestions, creating advice such as, "Try a healthy diet. For example, grilled chicken and vegetables would be good."

[0404] A speech synthesis engine converts this advice into voice data and provides it to the user aloud.

[0405] Prompt Sentence Examples

[0406] What should I eat today?

[0407] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0408] Step 1:

[0409] A user uses the microphone of a smartphone or personal computer to input a question or request by voice. For example, the user might say, "What should I eat today?" The voice input capture means acquires the voice data and stores it as digital voice data. The input is the user's voice data, and the output is digital voice data stored on the device.

[0410] Step 2:

[0411] The device sends the captured voice data to the server, where a voice recognition engine runs and converts the voice data into text data. In this process, the voice data is the input and the text data is the output. Specifically, a voice recognition service such as Google Cloud Speech-to-Text is used.

[0412] Step 3:

[0413] The server sends the text data to a natural language processing engine for analysis. The natural language processing engine extracts key keywords and their relevance from the text. The input is the converted text data, and the output is the extracted keywords and their relevance. Specifically, the SpaCy library is used, and in this example, the keyword "what to eat" is extracted.

[0414] Step 4:

[0415] The server uses an emotion recognition mechanism to analyze emotions from the user's voice and text data. The input is the voice and text data, and the output is the identified emotion (e.g., "neutral"). Specific operations use the Transformers pipeline of HuggingFace.

[0416] Step 5:

[0417] Based on the analysis results, the server retrieves related information from a database. The input is the analyzed keywords and emotions, and the output is data on related meal suggestions. The database stores meal suggestions and ingredient information, and queries are issued to retrieve information. Specific operations involve SQL queries and API calls.

[0418] Step 6:

[0419] The server generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions. The input is information and emotion data acquired from the database, and the output is the generated advice text data. Specific operations include the use of a template engine and natural language generation technology.

[0420] Step 7:

[0421] The server sends the generated advice text data to a speech synthesis engine and converts it into speech data. The input is the advice text data, and the output is speech data. Specifically, the Pyttsx3 library is used.

[0422] Step 8:

[0423] The terminal receives the voice data and provides advice to the user by voice. The input is the voice data sent from the server, and the output is the voice output to the user. Specifically, the voice is played through the speaker.

[0424] Through the above processing steps, users can get appropriate and personalized meal suggestions through voice input and emotion recognition.

[0425] 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.

[0426] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0427] 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.

[0428] [Second embodiment]

[0429] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0430] 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.

[0431] 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).

[0432] 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.

[0433] 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.

[0434] 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).

[0435] 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. 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.

[0436] 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.

[0437] 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.

[0438] 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.

[0439] In the smart glasses 214, 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.

[0440] 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."

[0441] This invention is a system that provides appropriate advice based on information input by voice by the user. The system is designed to be easy to use by users through a smartphone app.

[0442] Basic system configuration

[0443] The system consists of the following main components:

[0444] 1. Audio input capture means (terminal)

[0445] 2. Speech recognition means (server)

[0446] 3. Natural language processing means (server)

[0447] 4. Database access method (server)

[0448] 5. Dialogue Generation Means (Server)

[0449] 6. Audio output means (terminal)

[0450] Program processing overview

[0451] The specific processing flow of the system will be explained in natural language below.

[0452] Acquiring voice input

[0453] The user starts the smartphone app and inputs a question or request by voice. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This voice is captured by the device and saved as digital audio data.

[0454] Converting audio data to text

[0455] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[0456] Intention analysis using natural language processing

[0457] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[0458] Retrieving information from a database

[0459] The server retrieves relevant information from a database based on the result of the intent analysis, for example, by issuing a query to retrieve information about fatigue reduction, and then retrieves the result.

[0460] Dialogue generation

[0461] The server generates appropriate advice for the user based on the acquired information. For example, it might say, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0462] Text-to-speech conversion

[0463] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0464] Audio output

[0465] Finally, the device plays back the audio data and provides advice to the user, who can receive appropriate advice through audio and use it in their daily lives.

[0466] Specific examples

[0467] For example, if a user asks "Inner Self F" a question like, "I've been so busy at work lately that I'm feeling stressed. What should I do?", the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "stress" and "ways to relieve stress," and retrieves useful ways to relieve stress from a database. Advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest" is generated and provided to the user via voice. In this way, users can easily obtain useful information and apply it to their daily lives.

[0468] As described above, this system realizes a series of processes that starts with voice input and provides appropriate advice to the user by voice.

[0469] The processing flow will be explained below.

[0470] Step 1:

[0471] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[0472] Step 2:

[0473] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[0474] Step 3:

[0475] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[0476] Step 4:

[0477] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[0478] Step 5:

[0479] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[0480] Step 6:

[0481] The server queries the database to retrieve relevant information based on the analysis results, for example, recommendations for fatigue recovery.

[0482] Step 7:

[0483] Based on the information acquired by the server, appropriate advice is generated for the user. For example, it might generate a sentence such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0484] Step 8:

[0485] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[0486] Step 9:

[0487] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[0488] Step 10:

[0489] The terminal plays back the received voice data and provides the user with the advice by voice. The played back voice conveys appropriate advice to the user.

[0490] Specific examples

[0491] The process after a user asks, "I've been feeling tired lately. What should I do?" is as follows:

[0492] Step 1: The user launches the app and says, "I've been feeling tired a lot lately. What should I do?"

[0493] Step 2: The device records this audio and saves it as audio data.

[0494] Step 3: The device sends the audio data to the server.

[0495] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[0496] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0497] Step 6: The server queries the database to obtain information about fatigue recovery.

[0498] Step 7: Based on the information obtained, the server generates advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0499] Step 8: The text data generated by the server is sent to the speech synthesis engine and converted into speech data.

[0500] Step 9: The server sends the audio data to the terminal.

[0501] Step 10: The terminal plays back the audio data and provides the advice to the user by voice.

[0502] This series of steps allows the user to easily obtain appropriate advice through voice input.

[0503] Example 1

[0504] 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."

[0505] In conventional systems, the process of providing appropriate advice to users based on voice input is complicated, making it difficult to accurately interpret the user's intentions and provide appropriate information. In particular, there were issues with the accuracy of speech recognition and natural language processing, making it impossible to quickly and accurately provide the information users were looking for. In addition, there was also the problem that when the advice generated was not in a natural dialogue format, it was difficult for users to understand the advice.

[0506] 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.

[0507] In this invention, the server includes means for converting voice data into text data, means for analyzing the user's intention based on the text data, means for retrieving related information from a database based on the analysis result, and means for using a generative AI model for generating generated advice information in a natural dialogue format. This makes it possible to accurately analyze the user's intention from voice input, quickly and accurately provide related information, and convey advice to the user in a natural dialogue format.

[0508] "User" refers to an individual who uses the system and provides voice input.

[0509] "Voice input information" refers to what a user speaks through a microphone using a smartphone app or digital device.

[0510] "Capturing means" refers to a device, such as a smartphone or microphone, that has the ability to capture audio input and store it as digital audio data.

[0511] "Voice data" refers to data in which information input by a user through voice is stored in digital form.

[0512] "Text data" refers to digital data that has been converted from audio data into text format.

[0513] "Speech recognition engine" refers to software or algorithms for converting voice data into text data.

[0514] "Means for analyzing user intent" refers to natural language processing technology that extracts important keywords and context from text data and analyzes their meaning.

[0515] A "natural language processing engine" refers to technology and software that analyzes text data and extracts user intent and keywords.

[0516] A "database" refers to a structured data storage for storing various information and providing relevant information in response to a query.

[0517] A "generative AI model" refers to an algorithm or software that generates natural-looking, conversational text based on input data.

[0518] "Means for obtaining related information" refers to a function for searching and obtaining appropriate information from a database based on the results of analyzing the user's intentions.

[0519] "Means for generating appropriate advice" refers to technologies and algorithms for automatically generating advice that is useful to users based on acquired information.

[0520] "Speech synthesis engine" refers to software or algorithms for converting text data into speech data.

[0521] "Means for providing to the user" refers to a device or system for playing back the generated audio data so that the user can hear it.

[0522] This invention relates to a system that provides appropriate advice based on information input by a user via voice. The system is designed to be easily accessible to users through a smartphone app. The system consists of the following main hardware and software components:

[0523] Basic system configuration

[0524] The system consists of the following main components:

[0525] 1. Audio input capture means (terminal)

[0526] The user starts a dedicated smartphone app and inputs voice. The device captures the voice through the microphone and saves it as digital voice data. This voice data is then stored in temporary storage on the smartphone.

[0527] 2. Speech recognition means (server)

[0528] The device compresses the captured voice data and sends it to the server, where a speech recognition engine such as the Google Speech-to-Text API runs and converts the voice data into text data, which is then stored in a back-end database.

[0529] 3. Natural language processing means (server)

[0530] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for analysis. The engine extracts important keywords and user intent from the text. For example, the keywords "fatigue builds up" and "measures" are extracted. Analysis is also performed to take into account the user's emotions and context.

[0531] 4. Database access method (server)

[0532] Based on the result of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database, a health food database) to obtain relevant information. For example, it obtains "information on methods for reducing fatigue" from the database. This information is used for further processing.

[0533] 5. Dialogue Generation Means (Server)

[0534] The server uses a generative AI model (e.g., GPT-3) to generate advice in a natural conversational format based on information retrieved from the database. For example, it might generate specific advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0535] 6. Audio output means (terminal)

[0536] The generated text data of the advice is sent to a speech synthesis engine (e.g., Google Text-to-Speech) on the server and converted into audio data. This audio data is then sent to the device, where it is played back through the device's speaker to provide the advice to the user.

[0537] Specific examples

[0538] For example, a user might ask their "inner self" a question such as, "Work has been so busy lately that I'm feeling stressed. What should I do?" This system captures the user's voice and converts it into text using speech recognition. Keywords such as "stress" and "ways to relieve stress" are extracted using natural language processing, and the server retrieves information about stress relief from a database. Based on the information retrieved, the server generates advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest," which the device then provides to the user via voice.

[0539] Prompt Sentence Examples

[0540] Example prompts for generative AI models:

[0541] "My work is so busy that I feel like I'm getting stressed. Can you tell me some effective ways to relieve stress?"

[0542] In this way, the system embodies a process for providing appropriate advice to the user based on voice input.

[0543] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0544] Step 1:

[0545] The user launches a dedicated smartphone app and inputs voice data. For example, they might say, "I've been feeling tired lately. What should I do?" This voice is captured by the device and saved as digital voice data. The user's voice is the input, and digital voice data is obtained as the output.

[0546] Step 2:

[0547] The terminal compresses the captured audio data and sends it to the server. The input is digital audio data, and the output is compressed audio data. The compressed audio data is transferred to the server via the Internet.

[0548] Step 3:

[0549] On the server, a speech recognition engine (e.g., Google Speech-to-Text API) receives the compressed voice data and converts it into text data. The input is the compressed voice data, and the output is the text data converted from the voice. For example, the voice saying "I've been feeling tired a lot lately. What should I do?" is converted into text data.

[0550] Step 4:

[0551] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for intent analysis. The input is text data, and the output is extracted keywords and intent. For example, the key keywords extracted are "fatigue builds up" and "measures."

[0552] Step 5:

[0553] Based on the results of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database or a health food database) to obtain related information. The input is the intent analysis result (keywords), and the output is related information. For example, data containing "methods for reducing fatigue" is obtained.

[0554] Step 6:

[0555] The server uses a generative AI model (e.g., GPT-3) to generate appropriate advice based on the acquired information. The input is relevant information acquired from the database, and the output is advice text in a natural conversational format. For example, the generated advice might be, "We recommend that you eat a diet rich in vitamin B, get at least 7-8 hours of quality sleep every day, and engage in light exercise."

[0556] Step 7:

[0557] The generated text data of the advice is converted into voice data by a speech synthesis engine (e.g., Google Text-to-Speech) on the server. The input is the text data of the advice, and the output is voice data. Once the voice data is generated, it is sent from the server to the device.

[0558] Step 8:

[0559] The terminal plays the received voice data through a speaker and provides advice to the user. The input is the voice data sent from the server, and the output is voice information that the user can hear. The user can receive appropriate advice through voice.

[0560] Through this series of steps, users can easily input their voice and receive appropriate advice via voice.

[0561] (Application example 1)

[0562] 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."

[0563] Conventional speech recognition systems have difficulty accurately analyzing user intent and providing relevant information, making it particularly difficult to suggest meal menus and restaurants that suit a user's preferences in the food delivery field. Furthermore, the inability to provide intuitive and prompt advice based on the voice information entered by the user creates a problem that impairs the user experience.

[0564] 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.

[0565] In this invention, the server includes a means for capturing voice-input information, a means for converting the information into text data, and a means for analyzing the user's intent, thereby smoothly generating suggestions suitable for food delivery and enabling the user to quickly and easily obtain appropriate meal menus and restaurant information.

[0566] A "means for capturing user voice input information" is a mechanism for obtaining user spoken voice data and storing it in digital form.

[0567] The "means for converting captured voice data into text data" is a mechanism for analyzing acquired voice data and converting it into corresponding text data.

[0568] "Means for analyzing user intent based on text data" refers to a mechanism that interprets the user's requests and intent from text data and extracts appropriate information based on that.

[0569] The "means for obtaining related information from a database based on the analysis results" is a mechanism for searching and obtaining information corresponding to a user request from a database.

[0570] The "means for generating appropriate advice for the user based on the acquired information" is a mechanism for automatically generating advice to be provided to the user based on the acquired information.

[0571] The "means for converting the generated advice from text data to voice data" is a mechanism for converting the generated advice in text format into voice format.

[0572] The "means for generating suggestions suitable for food delivery and providing feedback to the user" is a mechanism that generates suggestions for meal menus and restaurants suitable for food delivery based on requests input by the user via voice and communicates the results to the user.

[0573] This invention is a system that provides food delivery suggestions based on information input by voice from a user. The system includes the following main components:

[0574] Basic system configuration

[0575] 1. Audio input capture means (terminal)

[0576] 2. Speech recognition means (server)

[0577] 3. Natural language processing means (server)

[0578] 4. Database access method (server)

[0579] 5. Dialogue Generation Means (Server)

[0580] 6. Audio output means (terminal)

[0581] Program processing overview

[0582] Acquiring voice input

[0583] A user launches a smartphone app and speaks to ask a question or request a dish, such as "I want a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[0584] Converting audio data to text

[0585] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[0586] Intention analysis using natural language processing

[0587] The server then sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text. For example, key keywords like "healthy" and "lunch" are extracted.

[0588] Retrieving information from a database

[0589] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[0590] Dialogue generation

[0591] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[0592] Text-to-speech conversion

[0593] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0594] Audio output

[0595] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[0596] Specific examples

[0597] For example, if a user requests, "I want a light and healthy dinner," the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "light," "healthy," and "dinner," and retrieves corresponding menu and restaurant information from a database. A suggestion is generated and delivered to the user via voice: "How about a salad bowl and steamed fish set? You can order this at a nearby restaurant." In this way, users can easily obtain useful information and encourage their use of food delivery services.

[0598] Prompt Sentence Examples

[0599] markdown

[0600] Prompt statement

[0601] Your role is to provide an assistant that suggests appropriate meal options based on the meal requests entered by the user. Please analyze the following requests and suggest the corresponding menu options.

[0602] Question: I want a healthy lunch today.

[0603] Suggestion: How about a salad and grilled chicken combo for a healthy lunch?

[0604] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0605] Step 1:

[0606] A user launches a smartphone app and speaks to ask a question or describe a desired meal, for example, "I want to eat a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[0607] Input: User's voice data

[0608] Output: Digital audio data

[0609] Step 2:

[0610] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[0611] Input: Digital audio data

[0612] Output: Text data

[0613] Step 3:

[0614] The server sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text data. For example, key keywords like "healthy" and "lunch" are extracted.

[0615] Input: Text data

[0616] Output: Extracted keywords and intent

[0617] Step 4:

[0618] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[0619] Input: Extracted keywords and intent

[0620] Output: Information retrieved from the database

[0621] Step 5:

[0622] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[0623] Input: Information retrieved from the database

[0624] Output: Generated suggested text

[0625] Step 6:

[0626] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0627] Input: Generated suggested text

[0628] Output: Audio data

[0629] Step 7:

[0630] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[0631] Input: Audio data

[0632] Output: Providing spoken suggestions

[0633] 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.

[0634] This invention is a system that provides appropriate advice based on information input by voice by a user. By further combining this system with an emotion engine that recognizes the user's emotions, it is possible to provide more personalized advice based on emotions.

[0635] Basic system configuration

[0636] The system consists of the following main components:

[0637] 1. Audio input capture means (terminal)

[0638] 2. Speech recognition means (server)

[0639] 3. Natural language processing means (server)

[0640] 4. Emotion Recognition Method (Server)

[0641] 5. Database access method (server)

[0642] 6. Dialogue Generation Means (Server)

[0643] 7. Audio output means (terminal)

[0644] Program processing overview

[0645] The specific processing flow of the system will be explained in natural language below.

[0646] Acquiring voice input

[0647] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This speech is captured by the device and saved as digital audio data.

[0648] Converting audio data to text

[0649] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[0650] Intention analysis using natural language processing

[0651] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[0652] Emotion Recognition Processing

[0653] The server analyzes the user's emotions using emotion recognition means, using voice data and, if possible, facial expression data. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone and strength of the voice, choice of words, etc.

[0654] Retrieving information from a database

[0655] The server retrieves relevant information from the database based on the results of the intent analysis and emotion recognition. For example, it issues a query to retrieve information on fatigue reduction and stress relief, and retrieves the results.

[0656] Dialogue generation

[0657] The server generates appropriate advice for the user based on the acquired information and the user's emotions. For example, the server might generate advice such as, "I recommend that you eat a diet rich in vitamin B and get at least seven to eight hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[0658] Text-to-speech conversion

[0659] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0660] Audio output

[0661] Finally, the device plays back the audio data and provides the user with advice by voice, allowing the user to receive appropriate advice through voice and use it in their daily lives.

[0662] Specific examples

[0663] For example, if a user asks, "I'm so stressed out at work these days, what should I do?", the following steps are taken:

[0664] 1. The user speaks a question and the device records this speech.

[0665] 2. The device sends the recorded audio data to the server.

[0666] 3. The server uses a speech recognition engine to convert the voice data into text data.

[0667] 4. The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0668] 5. The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[0669] 6. The server queries the database to obtain information about stress relief.

[0670] 7. Based on the information obtained, the server generates the following advice: "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[0671] 8. The text data generated by the server is sent to the speech synthesis engine and converted into voice data.

[0672] 9. The server sends the audio data to the device.

[0673] 10. The terminal plays the audio data and provides advice to the user by voice.

[0674] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[0675] The processing flow will be explained below.

[0676] Step 1:

[0677] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[0678] Step 2:

[0679] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[0680] Step 3:

[0681] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[0682] Step 4:

[0683] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[0684] Step 5:

[0685] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[0686] Step 6:

[0687] The server uses voice data and facial expression data (if available) to analyze the user's emotions using an emotion engine. Emotions such as "fatigue," "stress," and "depression" are recognized from the tone and strength of the voice, as well as the choice of words.

[0688] Step 7:

[0689] The server queries the database and retrieves relevant information based on the analysis results, for example, recommendations for fatigue recovery and stress relief.

[0690] Step 8:

[0691] Based on the information acquired by the server, appropriate advice is generated for the user. Taking into account emotions, the advice generated is, "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[0692] Step 9:

[0693] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[0694] Step 10:

[0695] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[0696] Step 11:

[0697] The device plays back the received voice data and provides the user with advice in the form of voice. The played back voice conveys appropriate and emotionally sensitive advice to the user.

[0698] Specific examples

[0699] The process after a user asks, "I'm so stressed out at work these days, what should I do?":

[0700] Step 1: The user launches the app and says, "Work has been so busy lately that I'm feeling stressed. What should I do?"

[0701] Step 2: The device records this audio and saves it as audio data.

[0702] Step 3: The device sends the audio data to the server.

[0703] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[0704] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0705] Step 6: The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[0706] Step 7: The server queries the database to obtain information about stress relief.

[0707] Step 8: Based on the information obtained, the server generates the following advice: "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[0708] Step 9: The text data generated by the server is sent to a speech synthesis engine and converted into speech data.

[0709] Step 10: The server sends the voice data to the terminal.

[0710] Step 11: The terminal plays back the audio data and provides the advice to the user by voice.

[0711] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[0712] Example 2

[0713] 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."

[0714] Conventional voice input systems can only provide general responses to information entered by the user through voice, making it difficult to provide personalized advice based on the user's emotions and intentions. This often leaves users frustrated because they are unable to receive advice that is best suited to their situation. Furthermore, there are also issues with the accuracy of converting voice data into text data and the accuracy of the results of analyzing text data.

[0715] 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.

[0716] In this invention, the server includes a means for recognizing a user's emotion, a means for acquiring related information from a database based on the analysis result and the recognized emotion, and a means for generating appropriate advice for the user using a generative AI model based on the acquired information and the user's emotion, thereby making it possible to provide personalized advice that takes into account the user's emotion and intention.

[0717] "User" refers to a person who uses the system to input voice and receive advice.

[0718] "Voice input capture means" refers to a device or method that captures and stores a user's voice as digital audio data.

[0719] "Means for converting voice data into text data" refers to a device or method for converting captured voice data into text information, such as a voice recognition engine.

[0720] "Means for analyzing user intent based on text data" refers to a device or method for understanding a user's intent or main point from text data.

[0721] "Means for recognizing user emotions" refers to devices and methods that analyze a user's emotional state from information such as voice and facial expressions.

[0722] "Means for retrieving relevant information from a database based on the analysis result and the recognized emotion" refers to a device or method for searching and retrieving relevant information from a database based on the analyzed intention and the recognized emotion.

[0723] "Means for generating appropriate advice for a user using a generative AI model based on acquired information and the user's emotions" refers to a device or method for creating appropriate advice using a generative AI model, taking into account acquired information and the user's emotions.

[0724] The "means for converting text data into voice data" refers to a device or method for converting the text data of the generated advice into voice data.

[0725] The "means for providing the user with converted voice data" refers to a device or method for conveying the generated advice to the user as voice data.

[0726] The present invention provides a system for providing personalized advice based on information input by a user through speech, which includes, as its main components, a speech input capture unit, a speech recognition unit, a natural language processing unit, an emotion recognition unit, a database access unit, a dialogue generation unit, and a speech output unit.

[0727] Acquiring voice input

[0728] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" The device (smartphone) captures and saves this voice as digital audio data.

[0729] Converting audio data to text

[0730] The device sends the captured voice data over the Internet to a server, which then converts the voice data into text using a network-accessible speech recognition engine (e.g., Google Cloud Speech-to-Text).

[0731] Intention analysis using natural language processing

[0732] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) for analysis. By identifying important keywords and the user's intent from the analysis results, the keywords "fatigue builds up" and "measures" are extracted.

[0733] Emotion Recognition Processing

[0734] The server uses an emotion recognition engine on the voice data to analyze the user's emotions. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone of the voice and the words chosen.

[0735] Retrieving information from a database

[0736] Based on the results of intent analysis and emotion recognition, the server issues a query to retrieve relevant information from the database, for example, information on "fatigue reduction" and "stress relief."

[0737] Dialogue generation

[0738] The server considers the acquired information and the user's emotions and generates appropriate advice for the user using a generative AI model (e.g., OpenAI GPT-3). For example, it generates text data such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[0739] Prompt Sentence Examples

[0740] In response to a user's question, "Work has been so busy lately that I'm feeling stressed. What should I do?", generate advice when the emotion recognition result includes "stress." Include advice on specific dietary choices, relaxation methods, and exercise.

[0741] Text-to-speech conversion

[0742] The server sends the generated advice text data to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data.

[0743] Audio output

[0744] The converted voice data is sent to the terminal, which plays it back to provide the user with advice in the form of voice. This allows the user to receive appropriate advice through voice and use it in their daily lives.

[0745] This system allows users to receive personalized advice based on their emotions and intentions, resulting in a more satisfying user experience.

[0746] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0747] Step 1:

[0748] The user launches the smartphone app and speaks into the microphone to ask a question or ask for advice. The spoken voice contains specific details such as, "I've been feeling tired a lot lately. What should I do?" The device captures and saves this voice as digital voice data. The input is the user's voice data, and the output is digital voice data.

[0749] Step 2:

[0750] The device sends the captured voice data to a server via the Internet. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data. In this step, the input is voice data and the output is text data.

[0751] Step 3:

[0752] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the intent of the text. The analysis identifies important keywords and the user's intent. For example, the keywords "fatigue builds up" and "measures" are extracted. The input is the text data, and the output is the analysis results of the keywords and intent.

[0753] Step 4:

[0754] The server sends the voice data to an emotion recognition engine to analyze the user's emotions. Emotion recognition uses the tone of the voice and the choice of words, so emotions such as "fatigue," "stress," and "depression" can be recognized. The input is the voice data, and the output is the recognized emotion data.

[0755] Step 5:

[0756] Based on the results of intent analysis and emotion recognition, the server queries a database to retrieve relevant information. The database contains information on fatigue reduction and stress relief. The input is the analysis results and emotion data, and the output is relevant information.

[0757] Step 6:

[0758] The server takes into account the acquired information and the user's emotions and inputs prompt sentences into a generative AI model (e.g., OpenAI GPT-3) to generate appropriate advice for the user. For example, text data such as "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation techniques into your routine" is generated. The input is the acquired information and emotion data, and the output is the text data of the advice.

[0759] Step 7:

[0760] The server sends the generated text data of advice to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data. The input is the text data of advice, and the output is voice data.

[0761] Step 8:

[0762] The converted voice data is sent to the terminal, which then plays it back. This allows the user to receive advice by voice and put it into practice in their daily lives. The input is voice data, and the output is the played-back voice.

[0763] (Application example 2)

[0764] 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."

[0765] Conventional food delivery services have struggled to provide personalized meal recommendations based on the user's emotions and specific circumstances. This has resulted in a lack of convenience for users to select meals that best suit their mood and health. Furthermore, methods for improving convenience through voice input and output have also been inadequate.

[0766] The specific processing by the specific 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 analyzing the user's emotions using emotion recognition means, means for generating personalized meal suggestions based on the user's emotions, and means for providing the personalized meal suggestions to the user by audio output. This enables personalized meal suggestions to be made according to the user's emotions and individual circumstances.

[0767] A "user" is an individual who wishes to use the system to receive meal suggestions.

[0768] "Audio input" refers to audio data that a user inputs into the system using a voice capture device such as a microphone.

[0769] A "speech recognition engine" is a technical device or software that converts voice input into text data.

[0770] A "natural language processing engine" is a technical device or software that analyzes user intent and keywords from text data.

[0771] An "emotion recognition means" is a technical device or software that analyzes a user's voice or text data to identify their emotions.

[0772] A "database" is a technical device or system for storing information that stores acquired information and provides required data based on a query.

[0773] The "dialogue generating means" is a technical device or software that generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions.

[0774] A "speech synthesis engine" is a technical device or software that converts text data into speech data.

[0775] A "meal suggestion" is a set of advice to provide the user with suitable meal options based on the user's emotions and situation.

[0776] This invention is a system that analyzes emotions based on information input by a user's voice and provides personalized meal suggestions. The system is composed of the following main components:

[0777] 1. Voice input capture method: The user uses a smartphone or other voice input device to input questions or requests by voice.

[0778] 2. Speech recognition means: Converts the voice data captured by the voice input capture means into text data. This process is performed using a voice recognition engine accessible via a network (e.g., Google Cloud Speech-to-Text).

[0779] 3. Natural language processing: Analyze user intent and keywords from text data. This process is performed using a natural language processing engine (e.g., SpaCy).

[0780] 4. Emotion Recognition: Analyzes the user's voice and text data to identify their emotions. This process is performed using generative AI models for emotion recognition (e.g., Transformers in HuggingFace).

[0781] 5. Database access method: Based on the analysis results, relevant information is retrieved from a database, which contains information about specific dishes and ingredients.

[0782] 6. Dialogue generation means: Based on the acquired information and the user's emotions, appropriate advice and meal suggestions are generated for the user.

[0783] 7. Voice output means: The generated advice is converted from text data to voice data. This process is performed using a voice synthesis engine (e.g., pyttsx3).

[0784] System operation explanation

[0785] Hardware and Software Requirements

[0786] Hardware: Smartphone or personal computer with microphone

[0787] Software: Python programs, SpeechRecognition, Pyttsx3, SpaCy, Transformers

[0788] Data processing and calculation

[0789] 1. Acquiring voice input: The microphone of a smartphone or personal computer is used to acquire the user's voice, which is then saved as digital voice data by the voice input capture means.

[0790] 2. Speech-to-text conversion: Using a speech recognition engine, the captured speech is converted into text, which is then used for further processing.

[0791] 3. Intent analysis using natural language processing: Analyze text data using a natural language processing engine to extract key keywords and their relevance.

[0792] 4. Emotion recognition processing: Emotion recognition means are used to analyze the user's emotions from voice and text data.

[0793] 5. Information retrieval from the database: Based on the analyzed intentions and emotions, information is retrieved from the database to make appropriate meal suggestions.

[0794] 6. Dialogue generation: Generate personalized meal suggestions based on the acquired information and user emotions.

[0795] 7. Voice output: The generated meal suggestions are converted into voice data using a voice synthesis engine and provided to the user audibly.

[0796] Specific examples

[0797] If a user speaks "What should I eat today?", the system will act as follows:

[0798] The speech recognition engine converts the speech to text, generating the text "What should I eat today?"

[0799] A natural language processing engine analyzes this text and extracts the request "what to eat" as a keyword.

[0800] The emotion recognition means identifies the user's emotion as "neutral" from the voice data.

[0801] A database access means retrieves meal suggestions based on "neutral" emotions from the database.

[0802] The dialogue generator generates appropriate dietary suggestions, creating advice such as, "Try a healthy diet. For example, grilled chicken and vegetables would be good."

[0803] A speech synthesis engine converts this advice into voice data and provides it to the user aloud.

[0804] Prompt Sentence Examples

[0805] What should I eat today?

[0806] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0807] Step 1:

[0808] A user uses the microphone of a smartphone or personal computer to input a question or request by voice. For example, the user might say, "What should I eat today?" The voice input capture means acquires the voice data and stores it as digital voice data. The input is the user's voice data, and the output is digital voice data stored on the device.

[0809] Step 2:

[0810] The device sends the captured voice data to the server, where a voice recognition engine runs and converts the voice data into text data. In this process, the voice data is the input and the text data is the output. Specifically, a voice recognition service such as Google Cloud Speech-to-Text is used.

[0811] Step 3:

[0812] The server sends the text data to a natural language processing engine for analysis. The natural language processing engine extracts key keywords and their relevance from the text. The input is the converted text data, and the output is the extracted keywords and their relevance. Specifically, the SpaCy library is used, and in this example, the keyword "what to eat" is extracted.

[0813] Step 4:

[0814] The server uses an emotion recognition mechanism to analyze emotions from the user's voice and text data. The input is the voice and text data, and the output is the identified emotion (e.g., "neutral"). Specific operations use the Transformers pipeline of HuggingFace.

[0815] Step 5:

[0816] Based on the analysis results, the server retrieves related information from a database. The input is the analyzed keywords and emotions, and the output is data on related meal suggestions. The database stores meal suggestions and ingredient information, and queries are issued to retrieve information. Specific operations involve SQL queries and API calls.

[0817] Step 6:

[0818] The server generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions. The input is information and emotion data acquired from the database, and the output is the generated advice text data. Specific operations include the use of a template engine and natural language generation technology.

[0819] Step 7:

[0820] The server sends the generated advice text data to a speech synthesis engine and converts it into speech data. The input is the advice text data, and the output is speech data. Specifically, the Pyttsx3 library is used.

[0821] Step 8:

[0822] The terminal receives the voice data and provides advice to the user by voice. The input is the voice data sent from the server, and the output is the voice output to the user. Specifically, the voice is played through the speaker.

[0823] Through the above processing steps, users can get appropriate and personalized meal suggestions through voice input and emotion recognition.

[0824] 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.

[0825] 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.

[0826] 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.

[0827] [Third embodiment]

[0828] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0829] 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.

[0830] 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).

[0831] 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.

[0832] 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.

[0833] 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).

[0834] 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. 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.

[0835] 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.

[0836] 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.

[0837] 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.

[0838] 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.

[0839] 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."

[0840] This invention is a system that provides appropriate advice based on information input by voice by the user. The system is designed to be easy to use by users through a smartphone app.

[0841] Basic system configuration

[0842] The system consists of the following main components:

[0843] 1. Audio input capture means (terminal)

[0844] 2. Speech recognition means (server)

[0845] 3. Natural language processing means (server)

[0846] 4. Database access method (server)

[0847] 5. Dialogue Generation Means (Server)

[0848] 6. Audio output means (terminal)

[0849] Program processing overview

[0850] The specific processing flow of the system will be explained in natural language below.

[0851] Acquiring voice input

[0852] The user starts the smartphone app and inputs a question or request by voice. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This voice is captured by the device and saved as digital audio data.

[0853] Converting audio data to text

[0854] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[0855] Intention analysis using natural language processing

[0856] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[0857] Retrieving information from a database

[0858] The server retrieves relevant information from a database based on the result of the intent analysis, for example, by issuing a query to retrieve information about fatigue reduction, and then retrieves the result.

[0859] Dialogue generation

[0860] The server generates appropriate advice for the user based on the acquired information. For example, it might say, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0861] Text-to-speech conversion

[0862] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0863] Audio output

[0864] Finally, the device plays back the audio data and provides advice to the user, who can receive appropriate advice through audio and use it in their daily lives.

[0865] Specific examples

[0866] For example, if a user asks "Inner Self F" a question like, "I've been so busy at work lately that I'm feeling stressed. What should I do?", the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "stress" and "ways to relieve stress," and retrieves useful ways to relieve stress from a database. Advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest" is generated and provided to the user via voice. In this way, users can easily obtain useful information and apply it to their daily lives.

[0867] As described above, this system realizes a series of processes that starts with voice input and provides appropriate advice to the user by voice.

[0868] The processing flow will be explained below.

[0869] Step 1:

[0870] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[0871] Step 2:

[0872] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[0873] Step 3:

[0874] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[0875] Step 4:

[0876] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[0877] Step 5:

[0878] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[0879] Step 6:

[0880] The server queries the database to retrieve relevant information based on the analysis results, for example, recommendations for fatigue recovery.

[0881] Step 7:

[0882] Based on the information acquired by the server, appropriate advice is generated for the user. For example, it might generate a sentence such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0883] Step 8:

[0884] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[0885] Step 9:

[0886] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[0887] Step 10:

[0888] The terminal plays back the received voice data and provides the user with the advice by voice. The played back voice conveys appropriate advice to the user.

[0889] Specific examples

[0890] The process after a user asks, "I've been feeling tired lately. What should I do?" is as follows:

[0891] Step 1: The user launches the app and says, "I've been feeling tired a lot lately. What should I do?"

[0892] Step 2: The device records this audio and saves it as audio data.

[0893] Step 3: The device sends the audio data to the server.

[0894] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[0895] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[0896] Step 6: The server queries the database to obtain information about fatigue recovery.

[0897] Step 7: Based on the information obtained, the server generates advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0898] Step 8: The text data generated by the server is sent to the speech synthesis engine and converted into speech data.

[0899] Step 9: The server sends the audio data to the terminal.

[0900] Step 10: The terminal plays back the audio data and provides the advice to the user by voice.

[0901] This series of steps allows the user to easily obtain appropriate advice through voice input.

[0902] Example 1

[0903] 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."

[0904] In conventional systems, the process of providing appropriate advice to users based on voice input is complicated, making it difficult to accurately interpret the user's intentions and provide appropriate information. In particular, there were issues with the accuracy of speech recognition and natural language processing, making it impossible to quickly and accurately provide the information users were looking for. In addition, there was also the problem that when the advice generated was not in a natural dialogue format, it was difficult for users to understand the advice.

[0905] 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.

[0906] In this invention, the server includes means for converting voice data into text data, means for analyzing the user's intention based on the text data, means for retrieving related information from a database based on the analysis result, and means for using a generative AI model for generating generated advice information in a natural dialogue format. This makes it possible to accurately analyze the user's intention from voice input, quickly and accurately provide related information, and convey advice to the user in a natural dialogue format.

[0907] "User" refers to an individual who uses the system and provides voice input.

[0908] "Voice input information" refers to what a user speaks through a microphone using a smartphone app or digital device.

[0909] "Capturing means" refers to a device, such as a smartphone or microphone, that has the ability to capture audio input and store it as digital audio data.

[0910] "Voice data" refers to data in which information input by a user through voice is stored in digital form.

[0911] "Text data" refers to digital data that has been converted from audio data into text format.

[0912] "Speech recognition engine" refers to software or algorithms for converting voice data into text data.

[0913] "Means for analyzing user intent" refers to natural language processing technology that extracts important keywords and context from text data and analyzes their meaning.

[0914] A "natural language processing engine" refers to technology and software that analyzes text data and extracts user intent and keywords.

[0915] A "database" refers to a structured data storage for storing various information and providing relevant information in response to a query.

[0916] A "generative AI model" refers to an algorithm or software that generates natural-looking, conversational text based on input data.

[0917] "Means for obtaining related information" refers to a function for searching and obtaining appropriate information from a database based on the results of analyzing the user's intentions.

[0918] "Means for generating appropriate advice" refers to technologies and algorithms for automatically generating advice that is useful to users based on acquired information.

[0919] "Speech synthesis engine" refers to software or algorithms for converting text data into speech data.

[0920] "Means for providing to the user" refers to a device or system for playing back the generated audio data so that the user can hear it.

[0921] This invention relates to a system that provides appropriate advice based on information input by a user via voice. The system is designed to be easily accessible to users through a smartphone app. The system consists of the following main hardware and software components:

[0922] Basic system configuration

[0923] The system consists of the following main components:

[0924] 1. Audio input capture means (terminal)

[0925] The user starts a dedicated smartphone app and inputs voice. The device captures the voice through the microphone and saves it as digital voice data. This voice data is then stored in temporary storage on the smartphone.

[0926] 2. Speech recognition means (server)

[0927] The device compresses the captured voice data and sends it to the server, where a speech recognition engine such as the Google Speech-to-Text API runs and converts the voice data into text data, which is then stored in a back-end database.

[0928] 3. Natural language processing means (server)

[0929] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for analysis. The engine extracts important keywords and user intent from the text. For example, the keywords "fatigue builds up" and "measures" are extracted. Analysis is also performed to take into account the user's emotions and context.

[0930] 4. Database access method (server)

[0931] Based on the result of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database, a health food database) to obtain relevant information. For example, it obtains "information on methods for reducing fatigue" from the database. This information is used for further processing.

[0932] 5. Dialogue Generation Means (Server)

[0933] The server uses a generative AI model (e.g., GPT-3) to generate advice in a natural conversational format based on information retrieved from the database. For example, it might generate specific advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[0934] 6. Audio output means (terminal)

[0935] The generated text data of the advice is sent to a speech synthesis engine (e.g., Google Text-to-Speech) on the server and converted into audio data. This audio data is then sent to the device, where it is played back through the device's speaker to provide the advice to the user.

[0936] Specific examples

[0937] For example, a user might ask their "inner self" a question such as, "Work has been so busy lately that I'm feeling stressed. What should I do?" This system captures the user's voice and converts it into text using speech recognition. Keywords such as "stress" and "ways to relieve stress" are extracted using natural language processing, and the server retrieves information about stress relief from a database. Based on the information retrieved, the server generates advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest," which the device then provides to the user via voice.

[0938] Prompt Sentence Examples

[0939] Example prompts for generative AI models:

[0940] "My work is so busy that I feel like I'm getting stressed. Can you tell me some effective ways to relieve stress?"

[0941] In this way, the system embodies a process for providing appropriate advice to the user based on voice input.

[0942] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0943] Step 1:

[0944] The user launches a dedicated smartphone app and inputs voice data. For example, they might say, "I've been feeling tired lately. What should I do?" This voice is captured by the device and saved as digital voice data. The user's voice is the input, and digital voice data is obtained as the output.

[0945] Step 2:

[0946] The terminal compresses the captured audio data and sends it to the server. The input is digital audio data, and the output is compressed audio data. The compressed audio data is transferred to the server via the Internet.

[0947] Step 3:

[0948] On the server, a speech recognition engine (e.g., Google Speech-to-Text API) receives the compressed voice data and converts it into text data. The input is the compressed voice data, and the output is the text data converted from the voice. For example, the voice saying "I've been feeling tired a lot lately. What should I do?" is converted into text data.

[0949] Step 4:

[0950] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for intent analysis. The input is text data, and the output is extracted keywords and intent. For example, the key keywords extracted are "fatigue builds up" and "measures."

[0951] Step 5:

[0952] Based on the results of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database or a health food database) to obtain related information. The input is the intent analysis result (keywords), and the output is related information. For example, data containing "methods for reducing fatigue" is obtained.

[0953] Step 6:

[0954] The server uses a generative AI model (e.g., GPT-3) to generate appropriate advice based on the acquired information. The input is relevant information acquired from the database, and the output is advice text in a natural conversational format. For example, the generated advice might be, "We recommend that you eat a diet rich in vitamin B, get at least 7-8 hours of quality sleep every day, and engage in light exercise."

[0955] Step 7:

[0956] The generated text data of the advice is converted into voice data by a speech synthesis engine (e.g., Google Text-to-Speech) on the server. The input is the text data of the advice, and the output is voice data. Once the voice data is generated, it is sent from the server to the device.

[0957] Step 8:

[0958] The terminal plays the received voice data through a speaker and provides advice to the user. The input is the voice data sent from the server, and the output is voice information that the user can hear. The user can receive appropriate advice through voice.

[0959] Through this series of steps, users can easily input their voice and receive appropriate advice via voice.

[0960] (Application example 1)

[0961] 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."

[0962] Conventional speech recognition systems have difficulty accurately analyzing user intent and providing relevant information, making it particularly difficult to suggest meal menus and restaurants that suit a user's preferences in the food delivery field. Furthermore, the inability to provide intuitive and prompt advice based on the voice information entered by the user creates a problem that impairs the user experience.

[0963] 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.

[0964] In this invention, the server includes a means for capturing voice-input information, a means for converting the information into text data, and a means for analyzing the user's intent, thereby smoothly generating suggestions suitable for food delivery and enabling the user to quickly and easily obtain appropriate meal menus and restaurant information.

[0965] A "means for capturing user voice input information" is a mechanism for obtaining user spoken voice data and storing it in digital form.

[0966] The "means for converting captured voice data into text data" is a mechanism for analyzing acquired voice data and converting it into corresponding text data.

[0967] "Means for analyzing user intent based on text data" refers to a mechanism that interprets the user's requests and intent from text data and extracts appropriate information based on that.

[0968] The "means for obtaining related information from a database based on the analysis results" is a mechanism for searching and obtaining information corresponding to a user request from a database.

[0969] The "means for generating appropriate advice for the user based on the acquired information" is a mechanism for automatically generating advice to be provided to the user based on the acquired information.

[0970] The "means for converting the generated advice from text data to voice data" is a mechanism for converting the generated advice in text format into voice format.

[0971] The "means for generating suggestions suitable for food delivery and providing feedback to the user" is a mechanism that generates suggestions for meal menus and restaurants suitable for food delivery based on requests input by the user via voice and communicates the results to the user.

[0972] This invention is a system that provides food delivery suggestions based on information input by voice from a user. The system includes the following main components:

[0973] Basic system configuration

[0974] 1. Audio input capture means (terminal)

[0975] 2. Speech recognition means (server)

[0976] 3. Natural language processing means (server)

[0977] 4. Database access method (server)

[0978] 5. Dialogue Generation Means (Server)

[0979] 6. Audio output means (terminal)

[0980] Program processing overview

[0981] Acquiring voice input

[0982] A user launches a smartphone app and speaks to ask a question or request a dish, such as "I want a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[0983] Converting audio data to text

[0984] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[0985] Intention analysis using natural language processing

[0986] The server then sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text. For example, key keywords like "healthy" and "lunch" are extracted.

[0987] Retrieving information from a database

[0988] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[0989] Dialogue generation

[0990] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[0991] Text-to-speech conversion

[0992] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[0993] Audio output

[0994] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[0995] Specific examples

[0996] For example, if a user requests, "I want a light and healthy dinner," the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "light," "healthy," and "dinner," and retrieves corresponding menu and restaurant information from a database. A suggestion is generated and delivered to the user via voice: "How about a salad bowl and steamed fish set? You can order this at a nearby restaurant." In this way, users can easily obtain useful information and encourage their use of food delivery services.

[0997] Prompt Sentence Examples

[0998] markdown

[0999] Prompt statement

[1000] Your role is to provide an assistant that suggests appropriate meal options based on the meal requests entered by the user. Please analyze the following requests and suggest the corresponding menu options.

[1001] Question: I want a healthy lunch today.

[1002] Suggestion: How about a salad and grilled chicken combo for a healthy lunch?

[1003] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1004] Step 1:

[1005] A user launches a smartphone app and speaks to ask a question or describe a desired meal, for example, "I want to eat a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[1006] Input: User's voice data

[1007] Output: Digital audio data

[1008] Step 2:

[1009] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[1010] Input: Digital audio data

[1011] Output: Text data

[1012] Step 3:

[1013] The server sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text data. For example, key keywords like "healthy" and "lunch" are extracted.

[1014] Input: Text data

[1015] Output: Extracted keywords and intent

[1016] Step 4:

[1017] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[1018] Input: Extracted keywords and intent

[1019] Output: Information retrieved from the database

[1020] Step 5:

[1021] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[1022] Input: Information retrieved from the database

[1023] Output: Generated suggested text

[1024] Step 6:

[1025] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[1026] Input: Generated suggested text

[1027] Output: Audio data

[1028] Step 7:

[1029] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[1030] Input: Audio data

[1031] Output: Providing spoken suggestions

[1032] 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.

[1033] This invention is a system that provides appropriate advice based on information input by voice by a user. By further combining this system with an emotion engine that recognizes the user's emotions, it is possible to provide more personalized advice based on emotions.

[1034] Basic system configuration

[1035] The system consists of the following main components:

[1036] 1. Audio input capture means (terminal)

[1037] 2. Speech recognition means (server)

[1038] 3. Natural language processing means (server)

[1039] 4. Emotion Recognition Method (Server)

[1040] 5. Database access method (server)

[1041] 6. Dialogue Generation Means (Server)

[1042] 7. Audio output means (terminal)

[1043] Program processing overview

[1044] The specific processing flow of the system will be explained in natural language below.

[1045] Acquiring voice input

[1046] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This speech is captured by the device and saved as digital audio data.

[1047] Converting audio data to text

[1048] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[1049] Intention analysis using natural language processing

[1050] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[1051] Emotion Recognition Processing

[1052] The server analyzes the user's emotions using emotion recognition means, using voice data and, if possible, facial expression data. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone and strength of the voice, choice of words, etc.

[1053] Retrieving information from a database

[1054] The server retrieves relevant information from the database based on the results of the intent analysis and emotion recognition. For example, it issues a query to retrieve information on fatigue reduction and stress relief, and retrieves the results.

[1055] Dialogue generation

[1056] The server generates appropriate advice for the user based on the acquired information and the user's emotions. For example, the server might generate advice such as, "I recommend that you eat a diet rich in vitamin B and get at least seven to eight hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[1057] Text-to-speech conversion

[1058] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[1059] Audio output

[1060] Finally, the device plays back the audio data and provides the user with advice by voice, allowing the user to receive appropriate advice through voice and use it in their daily lives.

[1061] Specific examples

[1062] For example, if a user asks, "I'm so stressed out at work these days, what should I do?", the following steps are taken:

[1063] 1. The user speaks a question and the device records this speech.

[1064] 2. The device sends the recorded audio data to the server.

[1065] 3. The server uses a speech recognition engine to convert the voice data into text data.

[1066] 4. The server uses a natural language processing engine to analyze the text data and extract key keywords.

[1067] 5. The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[1068] 6. The server queries the database to obtain information about stress relief.

[1069] 7. Based on the information obtained, the server generates the following advice: "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[1070] 8. The text data generated by the server is sent to the speech synthesis engine and converted into voice data.

[1071] 9. The server sends the audio data to the device.

[1072] 10. The terminal plays the audio data and provides advice to the user by voice.

[1073] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[1074] The processing flow will be explained below.

[1075] Step 1:

[1076] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[1077] Step 2:

[1078] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[1079] Step 3:

[1080] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[1081] Step 4:

[1082] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[1083] Step 5:

[1084] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[1085] Step 6:

[1086] The server uses voice data and facial expression data (if available) to analyze the user's emotions using an emotion engine. Emotions such as "fatigue," "stress," and "depression" are recognized from the tone and strength of the voice, as well as the choice of words.

[1087] Step 7:

[1088] The server queries the database and retrieves relevant information based on the analysis results, for example, recommendations for fatigue recovery and stress relief.

[1089] Step 8:

[1090] Based on the information acquired by the server, appropriate advice is generated for the user. Taking into account emotions, the advice generated is, "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[1091] Step 9:

[1092] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[1093] Step 10:

[1094] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[1095] Step 11:

[1096] The device plays back the received voice data and provides the user with advice in the form of voice. The played back voice conveys appropriate and emotionally sensitive advice to the user.

[1097] Specific examples

[1098] The process after a user asks, "I'm so stressed out at work these days, what should I do?":

[1099] Step 1: The user launches the app and says, "Work has been so busy lately that I'm feeling stressed. What should I do?"

[1100] Step 2: The device records this audio and saves it as audio data.

[1101] Step 3: The device sends the audio data to the server.

[1102] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[1103] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[1104] Step 6: The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[1105] Step 7: The server queries the database to obtain information about stress relief.

[1106] Step 8: Based on the information obtained, the server generates the following advice: "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[1107] Step 9: The text data generated by the server is sent to a speech synthesis engine and converted into speech data.

[1108] Step 10: The server sends the voice data to the terminal.

[1109] Step 11: The terminal plays back the audio data and provides the advice to the user by voice.

[1110] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[1111] Example 2

[1112] 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."

[1113] Conventional voice input systems can only provide general responses to information entered by the user through voice, making it difficult to provide personalized advice based on the user's emotions and intentions. This often leaves users frustrated because they are unable to receive advice that is best suited to their situation. Furthermore, there are also issues with the accuracy of converting voice data into text data and the accuracy of the results of analyzing text data.

[1114] 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.

[1115] In this invention, the server includes a means for recognizing a user's emotion, a means for acquiring related information from a database based on the analysis result and the recognized emotion, and a means for generating appropriate advice for the user using a generative AI model based on the acquired information and the user's emotion, thereby making it possible to provide personalized advice that takes into account the user's emotion and intention.

[1116] "User" refers to a person who uses the system to input voice and receive advice.

[1117] "Voice input capture means" refers to a device or method that captures and stores a user's voice as digital audio data.

[1118] "Means for converting voice data into text data" refers to a device or method for converting captured voice data into text information, such as a voice recognition engine.

[1119] "Means for analyzing user intent based on text data" refers to a device or method for understanding a user's intent or main point from text data.

[1120] "Means for recognizing user emotions" refers to devices and methods that analyze a user's emotional state from information such as voice and facial expressions.

[1121] "Means for retrieving relevant information from a database based on the analysis result and the recognized emotion" refers to a device or method for searching and retrieving relevant information from a database based on the analyzed intention and the recognized emotion.

[1122] "Means for generating appropriate advice for a user using a generative AI model based on acquired information and the user's emotions" refers to a device or method for creating appropriate advice using a generative AI model, taking into account acquired information and the user's emotions.

[1123] The "means for converting text data into voice data" refers to a device or method for converting the text data of the generated advice into voice data.

[1124] The "means for providing the user with converted voice data" refers to a device or method for conveying the generated advice to the user as voice data.

[1125] The present invention provides a system for providing personalized advice based on information input by a user through speech, which includes, as its main components, a speech input capture unit, a speech recognition unit, a natural language processing unit, an emotion recognition unit, a database access unit, a dialogue generation unit, and a speech output unit.

[1126] Acquiring voice input

[1127] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" The device (smartphone) captures and saves this voice as digital audio data.

[1128] Converting audio data to text

[1129] The device sends the captured voice data over the Internet to a server, which then converts the voice data into text using a network-accessible speech recognition engine (e.g., Google Cloud Speech-to-Text).

[1130] Intention analysis using natural language processing

[1131] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) for analysis. By identifying important keywords and the user's intent from the analysis results, the keywords "fatigue builds up" and "measures" are extracted.

[1132] Emotion Recognition Processing

[1133] The server uses an emotion recognition engine on the voice data to analyze the user's emotions. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone of the voice and the words chosen.

[1134] Retrieving information from a database

[1135] Based on the results of intent analysis and emotion recognition, the server issues a query to retrieve relevant information from the database, for example, information on "fatigue reduction" and "stress relief."

[1136] Dialogue generation

[1137] The server considers the acquired information and the user's emotions and generates appropriate advice for the user using a generative AI model (e.g., OpenAI GPT-3). For example, it generates text data such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[1138] Prompt Sentence Examples

[1139] In response to a user's question, "Work has been so busy lately that I'm feeling stressed. What should I do?", generate advice when the emotion recognition result includes "stress." Include advice on specific dietary choices, relaxation methods, and exercise.

[1140] Text-to-speech conversion

[1141] The server sends the generated advice text data to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data.

[1142] Audio output

[1143] The converted voice data is sent to the terminal, which plays it back to provide the user with advice in the form of voice. This allows the user to receive appropriate advice through voice and use it in their daily lives.

[1144] This system allows users to receive personalized advice based on their emotions and intentions, resulting in a more satisfying user experience.

[1145] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1146] Step 1:

[1147] The user launches the smartphone app and speaks into the microphone to ask a question or ask for advice. The spoken voice contains specific details such as, "I've been feeling tired a lot lately. What should I do?" The device captures and saves this voice as digital voice data. The input is the user's voice data, and the output is digital voice data.

[1148] Step 2:

[1149] The device sends the captured voice data to a server via the Internet. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data. In this step, the input is voice data and the output is text data.

[1150] Step 3:

[1151] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the intent of the text. The analysis identifies important keywords and the user's intent. For example, the keywords "fatigue builds up" and "measures" are extracted. The input is the text data, and the output is the analysis results of the keywords and intent.

[1152] Step 4:

[1153] The server sends the voice data to an emotion recognition engine to analyze the user's emotions. Emotion recognition uses the tone of the voice and the choice of words, so emotions such as "fatigue," "stress," and "depression" can be recognized. The input is the voice data, and the output is the recognized emotion data.

[1154] Step 5:

[1155] Based on the results of intent analysis and emotion recognition, the server queries a database to retrieve relevant information. The database contains information on fatigue reduction and stress relief. The input is the analysis results and emotion data, and the output is relevant information.

[1156] Step 6:

[1157] The server takes into account the acquired information and the user's emotions and inputs prompt sentences into a generative AI model (e.g., OpenAI GPT-3) to generate appropriate advice for the user. For example, text data such as "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation techniques into your routine" is generated. The input is the acquired information and emotion data, and the output is the text data of the advice.

[1158] Step 7:

[1159] The server sends the generated text data of advice to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data. The input is the text data of advice, and the output is voice data.

[1160] Step 8:

[1161] The converted voice data is sent to the terminal, which then plays it back. This allows the user to receive advice by voice and put it into practice in their daily lives. The input is voice data, and the output is the played-back voice.

[1162] (Application example 2)

[1163] 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."

[1164] Conventional food delivery services have struggled to provide personalized meal recommendations based on the user's emotions and specific circumstances. This has resulted in a lack of convenience for users to select meals that best suit their mood and health. Furthermore, methods for improving convenience through voice input and output have also been inadequate.

[1165] The specific processing by the specific 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 analyzing the user's emotions using emotion recognition means, means for generating personalized meal suggestions based on the user's emotions, and means for providing the personalized meal suggestions to the user by audio output. This enables personalized meal suggestions to be made according to the user's emotions and individual circumstances.

[1166] A "user" is an individual who wishes to use the system to receive meal suggestions.

[1167] "Audio input" refers to audio data that a user inputs into the system using a voice capture device such as a microphone.

[1168] A "speech recognition engine" is a technical device or software that converts voice input into text data.

[1169] A "natural language processing engine" is a technical device or software that analyzes user intent and keywords from text data.

[1170] An "emotion recognition means" is a technical device or software that analyzes a user's voice or text data to identify their emotions.

[1171] A "database" is a technical device or system for storing information that stores acquired information and provides required data based on a query.

[1172] The "dialogue generating means" is a technical device or software that generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions.

[1173] A "speech synthesis engine" is a technical device or software that converts text data into speech data.

[1174] A "meal suggestion" is a set of advice to provide the user with suitable meal options based on the user's emotions and situation.

[1175] This invention is a system that analyzes emotions based on information input by a user's voice and provides personalized meal suggestions. The system is composed of the following main components:

[1176] 1. Voice input capture method: The user uses a smartphone or other voice input device to input questions or requests by voice.

[1177] 2. Speech recognition means: Converts the voice data captured by the voice input capture means into text data. This process is performed using a voice recognition engine accessible via a network (e.g., Google Cloud Speech-to-Text).

[1178] 3. Natural language processing: Analyze user intent and keywords from text data. This process is performed using a natural language processing engine (e.g., SpaCy).

[1179] 4. Emotion Recognition: Analyzes the user's voice and text data to identify their emotions. This process is performed using generative AI models for emotion recognition (e.g., Transformers in HuggingFace).

[1180] 5. Database access method: Based on the analysis results, relevant information is retrieved from a database, which contains information about specific dishes and ingredients.

[1181] 6. Dialogue generation means: Based on the acquired information and the user's emotions, appropriate advice and meal suggestions are generated for the user.

[1182] 7. Voice output means: The generated advice is converted from text data to voice data. This process is performed using a voice synthesis engine (e.g., pyttsx3).

[1183] System operation explanation

[1184] Hardware and Software Requirements

[1185] Hardware: Smartphone or personal computer with microphone

[1186] Software: Python programs, SpeechRecognition, Pyttsx3, SpaCy, Transformers

[1187] Data processing and calculation

[1188] 1. Acquiring voice input: The microphone of a smartphone or personal computer is used to acquire the user's voice, which is then saved as digital voice data by the voice input capture means.

[1189] 2. Speech-to-text conversion: Using a speech recognition engine, the captured speech is converted into text, which is then used for further processing.

[1190] 3. Intent analysis using natural language processing: Analyze text data using a natural language processing engine to extract key keywords and their relevance.

[1191] 4. Emotion recognition processing: Emotion recognition means are used to analyze the user's emotions from voice and text data.

[1192] 5. Information retrieval from the database: Based on the analyzed intentions and emotions, information is retrieved from the database to make appropriate meal suggestions.

[1193] 6. Dialogue generation: Generate personalized meal suggestions based on the acquired information and user emotions.

[1194] 7. Voice output: The generated meal suggestions are converted into voice data using a voice synthesis engine and provided to the user audibly.

[1195] Specific examples

[1196] If a user speaks "What should I eat today?", the system will act as follows:

[1197] The speech recognition engine converts the speech to text, generating the text "What should I eat today?"

[1198] A natural language processing engine analyzes this text and extracts the request "what to eat" as a keyword.

[1199] The emotion recognition means identifies the user's emotion as "neutral" from the voice data.

[1200] A database access means retrieves meal suggestions based on "neutral" emotions from the database.

[1201] The dialogue generator generates appropriate dietary suggestions, creating advice such as, "Try a healthy diet. For example, grilled chicken and vegetables would be good."

[1202] A speech synthesis engine converts this advice into voice data and provides it to the user aloud.

[1203] Prompt Sentence Examples

[1204] What should I eat today?

[1205] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1206] Step 1:

[1207] A user uses the microphone of a smartphone or personal computer to input a question or request by voice. For example, the user might say, "What should I eat today?" The voice input capture means acquires the voice data and stores it as digital voice data. The input is the user's voice data, and the output is digital voice data stored on the device.

[1208] Step 2:

[1209] The device sends the captured voice data to the server, where a voice recognition engine runs and converts the voice data into text data. In this process, the voice data is the input and the text data is the output. Specifically, a voice recognition service such as Google Cloud Speech-to-Text is used.

[1210] Step 3:

[1211] The server sends the text data to a natural language processing engine for analysis. The natural language processing engine extracts key keywords and their relevance from the text. The input is the converted text data, and the output is the extracted keywords and their relevance. Specifically, the SpaCy library is used, and in this example, the keyword "what to eat" is extracted.

[1212] Step 4:

[1213] The server uses an emotion recognition mechanism to analyze emotions from the user's voice and text data. The input is the voice and text data, and the output is the identified emotion (e.g., "neutral"). Specific operations use the Transformers pipeline of HuggingFace.

[1214] Step 5:

[1215] Based on the analysis results, the server retrieves related information from a database. The input is the analyzed keywords and emotions, and the output is data on related meal suggestions. The database stores meal suggestions and ingredient information, and queries are issued to retrieve information. Specific operations involve SQL queries and API calls.

[1216] Step 6:

[1217] The server generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions. The input is information and emotion data acquired from the database, and the output is the generated advice text data. Specific operations include the use of a template engine and natural language generation technology.

[1218] Step 7:

[1219] The server sends the generated advice text data to a speech synthesis engine and converts it into speech data. The input is the advice text data, and the output is speech data. Specifically, the Pyttsx3 library is used.

[1220] Step 8:

[1221] The terminal receives the voice data and provides advice to the user by voice. The input is the voice data sent from the server, and the output is the voice output to the user. Specifically, the voice is played through the speaker.

[1222] Through the above processing steps, users can get appropriate and personalized meal suggestions through voice input and emotion recognition.

[1223] 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.

[1224] 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.

[1225] 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.

[1226] [Fourth embodiment]

[1227] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1228] 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.

[1229] 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).

[1230] 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.

[1231] 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.

[1232] 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).

[1233] 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. 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.

[1234] 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.

[1235] 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.

[1236] 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.

[1237] 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.

[1238] 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.

[1239] 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."

[1240] This invention is a system that provides appropriate advice based on information input by voice by the user. The system is designed to be easy to use by users through a smartphone app.

[1241] Basic system configuration

[1242] The system consists of the following main components:

[1243] 1. Audio input capture means (terminal)

[1244] 2. Speech recognition means (server)

[1245] 3. Natural language processing means (server)

[1246] 4. Database access method (server)

[1247] 5. Dialogue Generation Means (Server)

[1248] 6. Audio output means (terminal)

[1249] Program processing overview

[1250] The specific processing flow of the system will be explained in natural language below.

[1251] Acquiring voice input

[1252] The user starts the smartphone app and inputs a question or request by voice. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This voice is captured by the device and saved as digital audio data.

[1253] Converting audio data to text

[1254] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[1255] Intention analysis using natural language processing

[1256] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[1257] Retrieving information from a database

[1258] The server retrieves relevant information from a database based on the result of the intent analysis, for example, by issuing a query to retrieve information about fatigue reduction, and then retrieves the result.

[1259] Dialogue generation

[1260] The server generates appropriate advice for the user based on the acquired information. For example, it might say, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[1261] Text-to-speech conversion

[1262] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[1263] Audio output

[1264] Finally, the device plays back the audio data and provides advice to the user, who can receive appropriate advice through audio and use it in their daily lives.

[1265] Specific examples

[1266] For example, if a user asks "Inner Self F" a question like, "I've been so busy at work lately that I'm feeling stressed. What should I do?", the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "stress" and "ways to relieve stress," and retrieves useful ways to relieve stress from a database. Advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest" is generated and provided to the user via voice. In this way, users can easily obtain useful information and apply it to their daily lives.

[1267] As described above, this system realizes a series of processes that starts with voice input and provides appropriate advice to the user by voice.

[1268] The processing flow will be explained below.

[1269] Step 1:

[1270] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[1271] Step 2:

[1272] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[1273] Step 3:

[1274] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[1275] Step 4:

[1276] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[1277] Step 5:

[1278] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[1279] Step 6:

[1280] The server queries the database to retrieve relevant information based on the analysis results, for example, recommendations for fatigue recovery.

[1281] Step 7:

[1282] Based on the information acquired by the server, appropriate advice is generated for the user. For example, it might generate a sentence such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[1283] Step 8:

[1284] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[1285] Step 9:

[1286] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[1287] Step 10:

[1288] The terminal plays back the received voice data and provides the user with the advice by voice. The played back voice conveys appropriate advice to the user.

[1289] Specific examples

[1290] The process after a user asks, "I've been feeling tired lately. What should I do?" is as follows:

[1291] Step 1: The user launches the app and says, "I've been feeling tired a lot lately. What should I do?"

[1292] Step 2: The device records this audio and saves it as audio data.

[1293] Step 3: The device sends the audio data to the server.

[1294] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[1295] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[1296] Step 6: The server queries the database to obtain information about fatigue recovery.

[1297] Step 7: Based on the information obtained, the server generates advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[1298] Step 8: The text data generated by the server is sent to the speech synthesis engine and converted into speech data.

[1299] Step 9: The server sends the audio data to the terminal.

[1300] Step 10: The terminal plays back the audio data and provides the advice to the user by voice.

[1301] This series of steps allows the user to easily obtain appropriate advice through voice input.

[1302] Example 1

[1303] 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."

[1304] In conventional systems, the process of providing appropriate advice to users based on voice input is complicated, making it difficult to accurately interpret the user's intentions and provide appropriate information. In particular, there were issues with the accuracy of speech recognition and natural language processing, making it impossible to quickly and accurately provide the information users were looking for. In addition, there was also the problem that when the advice generated was not in a natural dialogue format, it was difficult for users to understand the advice.

[1305] 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.

[1306] In this invention, the server includes means for converting voice data into text data, means for analyzing the user's intention based on the text data, means for retrieving related information from a database based on the analysis result, and means for using a generative AI model for generating generated advice information in a natural dialogue format. This makes it possible to accurately analyze the user's intention from voice input, quickly and accurately provide related information, and convey advice to the user in a natural dialogue format.

[1307] "User" refers to an individual who uses the system and provides voice input.

[1308] "Voice input information" refers to what a user speaks through a microphone using a smartphone app or digital device.

[1309] "Capturing means" refers to a device, such as a smartphone or microphone, that has the ability to capture audio input and store it as digital audio data.

[1310] "Voice data" refers to data in which information input by a user through voice is stored in digital form.

[1311] "Text data" refers to digital data that has been converted from audio data into text format.

[1312] "Speech recognition engine" refers to software or algorithms for converting voice data into text data.

[1313] "Means for analyzing user intent" refers to natural language processing technology that extracts important keywords and context from text data and analyzes their meaning.

[1314] A "natural language processing engine" refers to technology and software that analyzes text data and extracts user intent and keywords.

[1315] A "database" refers to a structured data storage for storing various information and providing relevant information in response to a query.

[1316] A "generative AI model" refers to an algorithm or software that generates natural-looking, conversational text based on input data.

[1317] "Means for obtaining related information" refers to a function for searching and obtaining appropriate information from a database based on the results of analyzing the user's intentions.

[1318] "Means for generating appropriate advice" refers to technologies and algorithms for automatically generating advice that is useful to users based on acquired information.

[1319] "Speech synthesis engine" refers to software or algorithms for converting text data into speech data.

[1320] "Means for providing to the user" refers to a device or system for playing back the generated audio data so that the user can hear it.

[1321] This invention relates to a system that provides appropriate advice based on information input by a user via voice. The system is designed to be easily accessible to users through a smartphone app. The system consists of the following main hardware and software components:

[1322] Basic system configuration

[1323] The system consists of the following main components:

[1324] 1. Audio input capture means (terminal)

[1325] The user starts a dedicated smartphone app and inputs voice. The device captures the voice through the microphone and saves it as digital voice data. This voice data is then stored in temporary storage on the smartphone.

[1326] 2. Speech recognition means (server)

[1327] The device compresses the captured voice data and sends it to the server, where a speech recognition engine such as the Google Speech-to-Text API runs and converts the voice data into text data, which is then stored in a back-end database.

[1328] 3. Natural language processing means (server)

[1329] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for analysis. The engine extracts important keywords and user intent from the text. For example, the keywords "fatigue builds up" and "measures" are extracted. Analysis is also performed to take into account the user's emotions and context.

[1330] 4. Database access method (server)

[1331] Based on the result of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database, a health food database) to obtain relevant information. For example, it obtains "information on methods for reducing fatigue" from the database. This information is used for further processing.

[1332] 5. Dialogue Generation Means (Server)

[1333] The server uses a generative AI model (e.g., GPT-3) to generate advice in a natural conversational format based on information retrieved from the database. For example, it might generate specific advice such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective."

[1334] 6. Audio output means (terminal)

[1335] The generated text data of the advice is sent to a speech synthesis engine (e.g., Google Text-to-Speech) on the server and converted into audio data. This audio data is then sent to the device, where it is played back through the device's speaker to provide the advice to the user.

[1336] Specific examples

[1337] For example, a user might ask their "inner self" a question such as, "Work has been so busy lately that I'm feeling stressed. What should I do?" This system captures the user's voice and converts it into text using speech recognition. Keywords such as "stress" and "ways to relieve stress" are extracted using natural language processing, and the server retrieves information about stress relief from a database. Based on the information retrieved, the server generates advice such as "get some exercise," "enjoy your hobbies," and "get plenty of rest," which the device then provides to the user via voice.

[1338] Prompt Sentence Examples

[1339] Example prompts for generative AI models:

[1340] "My work is so busy that I feel like I'm getting stressed. Can you tell me some effective ways to relieve stress?"

[1341] In this way, the system embodies a process for providing appropriate advice to the user based on voice input.

[1342] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1343] Step 1:

[1344] The user launches a dedicated smartphone app and inputs voice data. For example, they might say, "I've been feeling tired lately. What should I do?" This voice is captured by the device and saved as digital voice data. The user's voice is the input, and digital voice data is obtained as the output.

[1345] Step 2:

[1346] The terminal compresses the captured audio data and sends it to the server. The input is digital audio data, and the output is compressed audio data. The compressed audio data is transferred to the server via the Internet.

[1347] Step 3:

[1348] On the server, a speech recognition engine (e.g., Google Speech-to-Text API) receives the compressed voice data and converts it into text data. The input is the compressed voice data, and the output is the text data converted from the voice. For example, the voice saying "I've been feeling tired a lot lately. What should I do?" is converted into text data.

[1349] Step 4:

[1350] The server then sends the converted text data to a natural language processing engine (e.g., SpaCy or BERT) for intent analysis. The input is text data, and the output is extracted keywords and intent. For example, the key keywords extracted are "fatigue builds up" and "measures."

[1351] Step 5:

[1352] Based on the results of the intent analysis, the server issues a query to an internal or external database (e.g., a medical database or a health food database) to obtain related information. The input is the intent analysis result (keywords), and the output is related information. For example, data containing "methods for reducing fatigue" is obtained.

[1353] Step 6:

[1354] The server uses a generative AI model (e.g., GPT-3) to generate appropriate advice based on the acquired information. The input is relevant information acquired from the database, and the output is advice text in a natural conversational format. For example, the generated advice might be, "We recommend that you eat a diet rich in vitamin B, get at least 7-8 hours of quality sleep every day, and engage in light exercise."

[1355] Step 7:

[1356] The generated text data of the advice is converted into voice data by a speech synthesis engine (e.g., Google Text-to-Speech) on the server. The input is the text data of the advice, and the output is voice data. Once the voice data is generated, it is sent from the server to the device.

[1357] Step 8:

[1358] The terminal plays the received voice data through a speaker and provides advice to the user. The input is the voice data sent from the server, and the output is voice information that the user can hear. The user can receive appropriate advice through voice.

[1359] Through this series of steps, users can easily input their voice and receive appropriate advice via voice.

[1360] (Application example 1)

[1361] 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."

[1362] Conventional speech recognition systems have difficulty accurately analyzing user intent and providing relevant information, making it particularly difficult to suggest meal menus and restaurants that suit a user's preferences in the food delivery field. Furthermore, the inability to provide intuitive and prompt advice based on the voice information entered by the user creates a problem that impairs the user experience.

[1363] 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.

[1364] In this invention, the server includes a means for capturing voice-input information, a means for converting the information into text data, and a means for analyzing the user's intent, thereby smoothly generating suggestions suitable for food delivery and enabling the user to quickly and easily obtain appropriate meal menus and restaurant information.

[1365] A "means for capturing user voice input information" is a mechanism for obtaining user spoken voice data and storing it in digital form.

[1366] The "means for converting captured voice data into text data" is a mechanism for analyzing acquired voice data and converting it into corresponding text data.

[1367] "Means for analyzing user intent based on text data" refers to a mechanism that interprets the user's requests and intent from text data and extracts appropriate information based on that.

[1368] The "means for obtaining related information from a database based on the analysis results" is a mechanism for searching and obtaining information corresponding to a user request from a database.

[1369] The "means for generating appropriate advice for the user based on the acquired information" is a mechanism for automatically generating advice to be provided to the user based on the acquired information.

[1370] The "means for converting the generated advice from text data to voice data" is a mechanism for converting the generated advice in text format into voice format.

[1371] The "means for generating suggestions suitable for food delivery and providing feedback to the user" is a mechanism that generates suggestions for meal menus and restaurants suitable for food delivery based on requests input by the user via voice and communicates the results to the user.

[1372] This invention is a system that provides food delivery suggestions based on information input by voice from a user. The system includes the following main components:

[1373] Basic system configuration

[1374] 1. Audio input capture means (terminal)

[1375] 2. Speech recognition means (server)

[1376] 3. Natural language processing means (server)

[1377] 4. Database access method (server)

[1378] 5. Dialogue Generation Means (Server)

[1379] 6. Audio output means (terminal)

[1380] Program processing overview

[1381] Acquiring voice input

[1382] A user launches a smartphone app and speaks to ask a question or request a dish, such as "I want a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[1383] Converting audio data to text

[1384] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[1385] Intention analysis using natural language processing

[1386] The server then sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text. For example, key keywords like "healthy" and "lunch" are extracted.

[1387] Retrieving information from a database

[1388] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[1389] Dialogue generation

[1390] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[1391] Text-to-speech conversion

[1392] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[1393] Audio output

[1394] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[1395] Specific examples

[1396] For example, if a user requests, "I want a light and healthy dinner," the system captures the user's voice and converts it into text using speech recognition. Natural language processing extracts the keywords "light," "healthy," and "dinner," and retrieves corresponding menu and restaurant information from a database. A suggestion is generated and delivered to the user via voice: "How about a salad bowl and steamed fish set? You can order this at a nearby restaurant." In this way, users can easily obtain useful information and encourage their use of food delivery services.

[1397] Prompt Sentence Examples

[1398] markdown

[1399] Prompt statement

[1400] Your role is to provide an assistant that suggests appropriate meal options based on the meal requests entered by the user. Please analyze the following requests and suggest the corresponding menu options.

[1401] Question: I want a healthy lunch today.

[1402] Suggestion: How about a salad and grilled chicken combo for a healthy lunch?

[1403] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1404] Step 1:

[1405] A user launches a smartphone app and speaks to ask a question or describe a desired meal, for example, "I want to eat a healthy lunch today." This speech is captured by the device and stored as digital audio data.

[1406] Input: User's voice data

[1407] Output: Digital audio data

[1408] Step 2:

[1409] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I want to eat a healthy lunch today."

[1410] Input: Digital audio data

[1411] Output: Text data

[1412] Step 3:

[1413] The server sends the text data to a natural language processing engine for analysis, which extracts key keywords and intent from the text data. For example, key keywords like "healthy" and "lunch" are extracted.

[1414] Input: Text data

[1415] Output: Extracted keywords and intent

[1416] Step 4:

[1417] The server retrieves relevant information from a database based on the result of intent analysis. For example, it issues a query to retrieve information about healthy lunch menus and restaurants that serve them, and retrieves the results.

[1418] Input: Extracted keywords and intent

[1419] Output: Information retrieved from the database

[1420] Step 5:

[1421] The server generates appropriate suggestions for the user based on the acquired information, such as "How about a salad and grilled chicken combo? You can order it at a nearby restaurant."

[1422] Input: Information retrieved from the database

[1423] Output: Generated suggested text

[1424] Step 6:

[1425] The generated text data of the proposal is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[1426] Input: Generated suggested text

[1427] Output: Audio data

[1428] Step 7:

[1429] Finally, the device plays the audio data and provides suggestions to the user, allowing the user to easily receive appropriate food delivery information through voice.

[1430] Input: Audio data

[1431] Output: Providing spoken suggestions

[1432] 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.

[1433] This invention is a system that provides appropriate advice based on information input by voice by a user. By further combining this system with an emotion engine that recognizes the user's emotions, it is possible to provide more personalized advice based on emotions.

[1434] Basic system configuration

[1435] The system consists of the following main components:

[1436] 1. Audio input capture means (terminal)

[1437] 2. Speech recognition means (server)

[1438] 3. Natural language processing means (server)

[1439] 4. Emotion Recognition Method (Server)

[1440] 5. Database access method (server)

[1441] 6. Dialogue Generation Means (Server)

[1442] 7. Audio output means (terminal)

[1443] Program processing overview

[1444] The specific processing flow of the system will be explained in natural language below.

[1445] Acquiring voice input

[1446] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" This speech is captured by the device and saved as digital audio data.

[1447] Converting audio data to text

[1448] The device sends the captured voice data to the server, where a speech recognition engine runs and converts the voice data into text data. The converted text data corresponds to the sentence, "I've been feeling tired a lot lately. What should I do?"

[1449] Intention analysis using natural language processing

[1450] The server then sends the text data to a natural language processing engine for analysis. The engine extracts key keywords and intent from the text. For example, the key keywords "fatigue builds up" and "measures" are extracted.

[1451] Emotion Recognition Processing

[1452] The server analyzes the user's emotions using emotion recognition means, using voice data and, if possible, facial expression data. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone and strength of the voice, choice of words, etc.

[1453] Retrieving information from a database

[1454] The server retrieves relevant information from the database based on the results of the intent analysis and emotion recognition. For example, it issues a query to retrieve information on fatigue reduction and stress relief, and retrieves the results.

[1455] Dialogue generation

[1456] The server generates appropriate advice for the user based on the acquired information and the user's emotions. For example, the server might generate advice such as, "I recommend that you eat a diet rich in vitamin B and get at least seven to eight hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[1457] Text-to-speech conversion

[1458] The generated text data of the advice is sent from the server to a speech synthesis engine, where it is converted into voice data, which is then sent back to the device.

[1459] Audio output

[1460] Finally, the device plays back the audio data and provides the user with advice by voice, allowing the user to receive appropriate advice through voice and use it in their daily lives.

[1461] Specific examples

[1462] For example, if a user asks, "I'm so stressed out at work these days, what should I do?", the following steps are taken:

[1463] 1. The user speaks a question and the device records this speech.

[1464] 2. The device sends the recorded audio data to the server.

[1465] 3. The server uses a speech recognition engine to convert the voice data into text data.

[1466] 4. The server uses a natural language processing engine to analyze the text data and extract key keywords.

[1467] 5. The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[1468] 6. The server queries the database to obtain information about stress relief.

[1469] 7. Based on the information obtained, the server generates the following advice: "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[1470] 8. The text data generated by the server is sent to the speech synthesis engine and converted into voice data.

[1471] 9. The server sends the audio data to the device.

[1472] 10. The terminal plays the audio data and provides advice to the user by voice.

[1473] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[1474] The processing flow will be explained below.

[1475] Step 1:

[1476] The user starts the smartphone app and voice-inputs a question or request for advice, for example, "I've been feeling tired a lot lately. What should I do?"

[1477] Step 2:

[1478] The device records the user's voice and saves it as digital audio data. The recorded audio data is temporarily stored on the device.

[1479] Step 3:

[1480] The device sends the recorded voice data to the server, which then transfers the voice data to the server via the network.

[1481] Step 4:

[1482] The server converts the transmitted voice data into text data using a speech recognition engine, which analyzes the voice signal and generates a corresponding string of characters.

[1483] Step 5:

[1484] The server uses a natural language processing engine to analyze the text data and extract key keywords and intent. The keywords "fatigue builds up" and "measures" are extracted.

[1485] Step 6:

[1486] The server uses voice data and facial expression data (if available) to analyze the user's emotions using an emotion engine. Emotions such as "fatigue," "stress," and "depression" are recognized from the tone and strength of the voice, as well as the choice of words.

[1487] Step 7:

[1488] The server queries the database and retrieves relevant information based on the analysis results, for example, recommendations for fatigue recovery and stress relief.

[1489] Step 8:

[1490] Based on the information acquired by the server, appropriate advice is generated for the user. Taking into account emotions, the advice generated is, "I recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[1491] Step 9:

[1492] The text data generated by the server is sent to a speech synthesis engine, which converts it into voice data. The speech synthesis engine generates a voice signal based on the input text.

[1493] Step 10:

[1494] The server transmits the generated voice data to the terminal, which then transfers the voice data to the terminal via the network.

[1495] Step 11:

[1496] The device plays back the received voice data and provides the user with advice in the form of voice. The played back voice conveys appropriate and emotionally sensitive advice to the user.

[1497] Specific examples

[1498] The process after a user asks, "I'm so stressed out at work these days, what should I do?":

[1499] Step 1: The user launches the app and says, "Work has been so busy lately that I'm feeling stressed. What should I do?"

[1500] Step 2: The device records this audio and saves it as audio data.

[1501] Step 3: The device sends the audio data to the server.

[1502] Step 4: The server uses a speech recognition engine to convert the voice data into text data.

[1503] Step 5: The server uses a natural language processing engine to analyze the text data and extract key keywords.

[1504] Step 6: The server recognizes the user's emotions from the voice data and identifies the emotion "stress."

[1505] Step 7: The server queries the database to obtain information about stress relief.

[1506] Step 8: Based on the information obtained, the server generates the following advice: "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. Since you seem to be feeling stressed, it would be a good idea to incorporate relaxation into your routine."

[1507] Step 9: The text data generated by the server is sent to a speech synthesis engine and converted into speech data.

[1508] Step 10: The server sends the voice data to the terminal.

[1509] Step 11: The terminal plays back the audio data and provides the advice to the user by voice.

[1510] This series of steps allows users to receive relevant and personalized advice through voice input and emotion recognition.

[1511] Example 2

[1512] 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."

[1513] Conventional voice input systems can only provide general responses to information entered by the user through voice, making it difficult to provide personalized advice based on the user's emotions and intentions. This often leaves users frustrated because they are unable to receive advice that is best suited to their situation. Furthermore, there are also issues with the accuracy of converting voice data into text data and the accuracy of the results of analyzing text data.

[1514] 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.

[1515] In this invention, the server includes a means for recognizing a user's emotion, a means for acquiring related information from a database based on the analysis result and the recognized emotion, and a means for generating appropriate advice for the user using a generative AI model based on the acquired information and the user's emotion, thereby making it possible to provide personalized advice that takes into account the user's emotion and intention.

[1516] "User" refers to a person who uses the system to input voice and receive advice.

[1517] "Voice input capture means" refers to a device or method that captures and stores a user's voice as digital audio data.

[1518] "Means for converting voice data into text data" refers to a device or method for converting captured voice data into text information, such as a voice recognition engine.

[1519] "Means for analyzing user intent based on text data" refers to a device or method for understanding a user's intent or main point from text data.

[1520] "Means for recognizing user emotions" refers to devices and methods that analyze a user's emotional state from information such as voice and facial expressions.

[1521] "Means for retrieving relevant information from a database based on the analysis result and the recognized emotion" refers to a device or method for searching and retrieving relevant information from a database based on the analyzed intention and the recognized emotion.

[1522] "Means for generating appropriate advice for a user using a generative AI model based on acquired information and the user's emotions" refers to a device or method for creating appropriate advice using a generative AI model, taking into account acquired information and the user's emotions.

[1523] The "means for converting text data into voice data" refers to a device or method for converting the text data of the generated advice into voice data.

[1524] The "means for providing the user with converted voice data" refers to a device or method for conveying the generated advice to the user as voice data.

[1525] The present invention provides a system for providing personalized advice based on information input by a user through speech, which includes, as its main components, a speech input capture unit, a speech recognition unit, a natural language processing unit, an emotion recognition unit, a database access unit, a dialogue generation unit, and a speech output unit.

[1526] Acquiring voice input

[1527] The user launches the smartphone app and speaks their question or inquiry. For example, they might ask, "I've been feeling tired a lot lately. What should I do?" The device (smartphone) captures and saves this voice as digital audio data.

[1528] Converting audio data to text

[1529] The device sends the captured voice data over the Internet to a server, which then converts the voice data into text using a network-accessible speech recognition engine (e.g., Google Cloud Speech-to-Text).

[1530] Intention analysis using natural language processing

[1531] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) for analysis. By identifying important keywords and the user's intent from the analysis results, the keywords "fatigue builds up" and "measures" are extracted.

[1532] Emotion Recognition Processing

[1533] The server uses an emotion recognition engine on the voice data to analyze the user's emotions. For example, emotions such as "fatigue," "stress," and "depression" can be recognized from the tone of the voice and the words chosen.

[1534] Retrieving information from a database

[1535] Based on the results of intent analysis and emotion recognition, the server issues a query to retrieve relevant information from the database, for example, information on "fatigue reduction" and "stress relief."

[1536] Dialogue generation

[1537] The server considers the acquired information and the user's emotions and generates appropriate advice for the user using a generative AI model (e.g., OpenAI GPT-3). For example, it generates text data such as, "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation into your routine."

[1538] Prompt Sentence Examples

[1539] In response to a user's question, "Work has been so busy lately that I'm feeling stressed. What should I do?", generate advice when the emotion recognition result includes "stress." Include advice on specific dietary choices, relaxation methods, and exercise.

[1540] Text-to-speech conversion

[1541] The server sends the generated advice text data to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data.

[1542] Audio output

[1543] The converted voice data is sent to the terminal, which plays it back to provide the user with advice in the form of voice. This allows the user to receive appropriate advice through voice and use it in their daily lives.

[1544] This system allows users to receive personalized advice based on their emotions and intentions, resulting in a more satisfying user experience.

[1545] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1546] Step 1:

[1547] The user launches the smartphone app and speaks into the microphone to ask a question or ask for advice. The spoken voice contains specific details such as, "I've been feeling tired a lot lately. What should I do?" The device captures and saves this voice as digital voice data. The input is the user's voice data, and the output is digital voice data.

[1548] Step 2:

[1549] The device sends the captured voice data to a server via the Internet. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data. In this step, the input is voice data and the output is text data.

[1550] Step 3:

[1551] The server sends the converted text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the intent of the text. The analysis identifies important keywords and the user's intent. For example, the keywords "fatigue builds up" and "measures" are extracted. The input is the text data, and the output is the analysis results of the keywords and intent.

[1552] Step 4:

[1553] The server sends the voice data to an emotion recognition engine to analyze the user's emotions. Emotion recognition uses the tone of the voice and the choice of words, so emotions such as "fatigue," "stress," and "depression" can be recognized. The input is the voice data, and the output is the recognized emotion data.

[1554] Step 5:

[1555] Based on the results of intent analysis and emotion recognition, the server queries a database to retrieve relevant information. The database contains information on fatigue reduction and stress relief. The input is the analysis results and emotion data, and the output is relevant information.

[1556] Step 6:

[1557] The server takes into account the acquired information and the user's emotions and inputs prompt sentences into a generative AI model (e.g., OpenAI GPT-3) to generate appropriate advice for the user. For example, text data such as "We recommend that you eat a diet rich in vitamin B and get at least 7 to 8 hours of quality sleep every day. Light exercise is also effective. You seem to be feeling stressed, so it would be a good idea to incorporate relaxation techniques into your routine" is generated. The input is the acquired information and emotion data, and the output is the text data of the advice.

[1558] Step 7:

[1559] The server sends the generated text data of advice to a speech synthesis engine (e.g., Google Cloud Text-to-Speech) and converts it into voice data. The input is the text data of advice, and the output is voice data.

[1560] Step 8:

[1561] The converted voice data is sent to the terminal, which then plays it back. This allows the user to receive advice by voice and put it into practice in their daily lives. The input is voice data, and the output is the played-back voice.

[1562] (Application example 2)

[1563] 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."

[1564] Conventional food delivery services have struggled to provide personalized meal recommendations based on the user's emotions and specific circumstances. This has resulted in a lack of convenience for users to select meals that best suit their mood and health. Furthermore, methods for improving convenience through voice input and output have also been inadequate.

[1565] The specific processing by the specific 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 analyzing the user's emotions using emotion recognition means, means for generating personalized meal suggestions based on the user's emotions, and means for providing the personalized meal suggestions to the user by audio output. This enables personalized meal suggestions to be made according to the user's emotions and individual circumstances.

[1566] A "user" is an individual who wishes to use the system to receive meal suggestions.

[1567] "Audio input" refers to audio data that a user inputs into the system using a voice capture device such as a microphone.

[1568] A "speech recognition engine" is a technical device or software that converts voice input into text data.

[1569] A "natural language processing engine" is a technical device or software that analyzes user intent and keywords from text data.

[1570] An "emotion recognition means" is a technical device or software that analyzes a user's voice or text data to identify their emotions.

[1571] A "database" is a technical device or system for storing information that stores acquired information and provides required data based on a query.

[1572] The "dialogue generating means" is a technical device or software that generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions.

[1573] A "speech synthesis engine" is a technical device or software that converts text data into speech data.

[1574] A "meal suggestion" is a set of advice to provide the user with suitable meal options based on the user's emotions and situation.

[1575] This invention is a system that analyzes emotions based on information input by a user's voice and provides personalized meal suggestions. The system is composed of the following main components:

[1576] 1. Voice input capture method: The user uses a smartphone or other voice input device to input questions or requests by voice.

[1577] 2. Speech recognition means: Converts the voice data captured by the voice input capture means into text data. This process is performed using a voice recognition engine accessible via a network (e.g., Google Cloud Speech-to-Text).

[1578] 3. Natural language processing: Analyze user intent and keywords from text data. This process is performed using a natural language processing engine (e.g., SpaCy).

[1579] 4. Emotion Recognition: Analyzes the user's voice and text data to identify their emotions. This process is performed using generative AI models for emotion recognition (e.g., Transformers in HuggingFace).

[1580] 5. Database access method: Based on the analysis results, relevant information is retrieved from a database, which contains information about specific dishes and ingredients.

[1581] 6. Dialogue generation means: Based on the acquired information and the user's emotions, appropriate advice and meal suggestions are generated for the user.

[1582] 7. Voice output means: The generated advice is converted from text data to voice data. This process is performed using a voice synthesis engine (e.g., pyttsx3).

[1583] System operation explanation

[1584] Hardware and Software Requirements

[1585] Hardware: Smartphone or personal computer with microphone

[1586] Software: Python programs, SpeechRecognition, Pyttsx3, SpaCy, Transformers

[1587] Data processing and calculation

[1588] 1. Acquiring voice input: The microphone of a smartphone or personal computer is used to acquire the user's voice, which is then saved as digital voice data by the voice input capture means.

[1589] 2. Speech-to-text conversion: Using a speech recognition engine, the captured speech is converted into text, which is then used for further processing.

[1590] 3. Intent analysis using natural language processing: Analyze text data using a natural language processing engine to extract key keywords and their relevance.

[1591] 4. Emotion recognition processing: Emotion recognition means are used to analyze the user's emotions from voice and text data.

[1592] 5. Information retrieval from the database: Based on the analyzed intentions and emotions, information is retrieved from the database to make appropriate meal suggestions.

[1593] 6. Dialogue generation: Generate personalized meal suggestions based on the acquired information and user emotions.

[1594] 7. Voice output: The generated meal suggestions are converted into voice data using a voice synthesis engine and provided to the user audibly.

[1595] Specific examples

[1596] If a user speaks "What should I eat today?", the system will act as follows:

[1597] The speech recognition engine converts the speech to text, generating the text "What should I eat today?"

[1598] A natural language processing engine analyzes this text and extracts the request "what to eat" as a keyword.

[1599] The emotion recognition means identifies the user's emotion as "neutral" from the voice data.

[1600] A database access means retrieves meal suggestions based on "neutral" emotions from the database.

[1601] The dialogue generator generates appropriate dietary suggestions, creating advice such as, "Try a healthy diet. For example, grilled chicken and vegetables would be good."

[1602] A speech synthesis engine converts this advice into voice data and provides it to the user aloud.

[1603] Prompt Sentence Examples

[1604] What should I eat today?

[1605] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1606] Step 1:

[1607] A user uses the microphone of a smartphone or personal computer to input a question or request by voice. For example, the user might say, "What should I eat today?" The voice input capture means acquires the voice data and stores it as digital voice data. The input is the user's voice data, and the output is digital voice data stored on the device.

[1608] Step 2:

[1609] The device sends the captured voice data to the server, where a voice recognition engine runs and converts the voice data into text data. In this process, the voice data is the input and the text data is the output. Specifically, a voice recognition service such as Google Cloud Speech-to-Text is used.

[1610] Step 3:

[1611] The server sends the text data to a natural language processing engine for analysis. The natural language processing engine extracts key keywords and their relevance from the text. The input is the converted text data, and the output is the extracted keywords and their relevance. Specifically, the SpaCy library is used, and in this example, the keyword "what to eat" is extracted.

[1612] Step 4:

[1613] The server uses an emotion recognition mechanism to analyze emotions from the user's voice and text data. The input is the voice and text data, and the output is the identified emotion (e.g., "neutral"). Specific operations use the Transformers pipeline of HuggingFace.

[1614] Step 5:

[1615] Based on the analysis results, the server retrieves related information from a database. The input is the analyzed keywords and emotions, and the output is data on related meal suggestions. The database stores meal suggestions and ingredient information, and queries are issued to retrieve information. Specific operations involve SQL queries and API calls.

[1616] Step 6:

[1617] The server generates appropriate advice and meal suggestions for the user based on the acquired information and the user's emotions. The input is information and emotion data acquired from the database, and the output is the generated advice text data. Specific operations include the use of a template engine and natural language generation technology.

[1618] Step 7:

[1619] The server sends the generated advice text data to a speech synthesis engine and converts it into speech data. The input is the advice text data, and the output is speech data. Specifically, the Pyttsx3 library is used.

[1620] Step 8:

[1621] The terminal receives the voice data and provides advice to the user by voice. The input is the voice data sent from the server, and the output is the voice output to the user. Specifically, the voice is played through the speaker.

[1622] Through the above processing steps, users can get appropriate and personalized meal suggestions through voice input and emotion recognition.

[1623] 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.

[1624] 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.

[1625] 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.

[1626] 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.

[1627] 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.

[1628] 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.

[1629] 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).

[1630] 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.

[1631] 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."

[1632] 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.

[1633] 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).

[1634] 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.

[1635] 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.

[1636] 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.

[1637] 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.

[1638] 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.

[1639] 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.

[1640] 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.

[1641] 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.

[1642] 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.

[1643] 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.

[1644] The following is further disclosed regarding the above embodiment.

[1645] (Claim 1)

[1646] means for capturing information input by voice from a user;

[1647] means for converting the captured audio data into text data;

[1648] means for analyzing user intent based on text data;

[1649] A means for retrieving related information from a database based on the analysis results;

[1650] A means for generating appropriate advice for a user based on the acquired information;

[1651] means for converting the generated advice from text data into voice data;

[1652] means for providing the converted audio data to a user;

[1653] A system including:

[1654] (Claim 2)

[1655] 2. The system according to claim 1, wherein the means for converting voice data into text data utilizes a voice recognition engine accessible via a network.

[1656] (Claim 3)

[1657] 10. The system of claim 1, wherein the means for analyzing user intent uses a natural language processing engine to extract keywords and associations.

[1658] "Example 1"

[1659] (Claim 1)

[1660] means for capturing information input by voice from a user;

[1661] means for converting the captured audio data into text data;

[1662] means for analyzing user intent based on text data;

[1663] A means for retrieving related information from a database based on the analysis results;

[1664] A means for generating appropriate advice for a user based on the acquired information;

[1665] means for converting the generated advice from text data into voice data;

[1666] means for providing the converted audio data to a user;

[1667] a means for using a generative AI model to generate generated advice information in a natural, interactive format;

[1668] A system including:

[1669] (Claim 2)

[1670] 2. The system according to claim 1, wherein the means for converting voice data into text data utilizes a voice recognition engine accessible via a network.

[1671] (Claim 3)

[1672] 10. The system of claim 1, wherein the means for analyzing user intent uses a natural language processing engine to extract keywords and associations.

[1673] "Application Example 1"

[1674] (Claim 1)

[1675] means for capturing information input by voice from a user;

[1676] means for converting the captured audio data into text data;

[1677] means for analyzing user intent based on text data;

[1678] A means for retrieving related information from a database based on the analysis results;

[1679] A means for generating appropriate advice for a user based on the acquired information;

[1680] means for converting the generated advice from text data into voice data;

[1681] means for generating food delivery suitable suggestions and providing feedback to the user;

[1682] A system including:

[1683] (Claim 2)

[1684] 2. The system according to claim 1, wherein the means for converting voice data into text data utilizes a voice recognition engine accessible via a network.

[1685] (Claim 3)

[1686] 2. The system of claim 1, wherein the means for analyzing user intent uses a natural language processing engine to extract keywords and associations, and provides a meal menu based on the extracted information.

[1687] "Example 2: Combining Emotion Engines"

[1688] (Claim 1)

[1689] means for capturing information input by voice from a user;

[1690] means for converting the captured audio data into text data;

[1691] means for analyzing user intent based on text data;

[1692] means for recognizing a user's emotion;

[1693] means for retrieving relevant information from a database based on the analysis result and the recognized emotion;

[1694] A means for generating appropriate advice for a user using a generative AI model based on the acquired information and the user's emotions;

[1695] means for converting the generated advice from text data into voice data;

[1696] means for providing the converted audio data to a user;

[1697] A system including:

[1698] (Claim 2)

[1699] 2. The system according to claim 1, wherein the means for converting voice data into text data utilizes a voice recognition engine accessible via a network.

[1700] (Claim 3)

[1701] 10. The system of claim 1, wherein the means for analyzing user intent uses a natural language processing engine to extract keywords and associations.

[1702] "Application example 2 when combining emotion engines"

[1703] (Claim 1)

[1704] means for capturing information input by voice from a user;

[1705] means for converting the captured audio data into text data;

[1706] means for analyzing user intent based on text data;

[1707] A means for retrieving related information from a database based on the analysis results;

[1708] A means for generating appropriate advice for a user based on the acquired information;

[1709] means for converting the generated advice from text data into voice data;

[1710] means for providing the converted audio data to a user;

[1711] means for analyzing a user's emotion by an emotion recognition means;

[1712] means for generating personalized meal suggestions based on user emotions;

[1713] means for providing personalized meal suggestions to the user via audio output;

[1714] A system including:

[1715] (Claim 2)

[1716] 2. The system according to claim 1, wherein the means for converting voice data into text data utilizes a voice recognition engine accessible via a network.

[1717] (Claim 3)

[1718] 10. The system of claim 1, wherein the means for analyzing user intent uses a natural language processing engine to extract keywords and associations. [Explanation of symbols]

[1719] 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 capturing information input by voice from a user; means for converting the captured audio data into text data; means for analyzing user intent based on text data; A means for retrieving related information from a database based on the analysis results; A means for generating appropriate advice for a user based on the acquired information; means for converting the generated advice from text data into voice data; means for providing the converted audio data to a user; A system including:

2. 2. The system according to claim 1, wherein the means for converting voice data into text data utilizes a voice recognition engine accessible via a network.

3. 10. The system of claim 1, wherein the means for analyzing user intent uses a natural language processing engine to extract keywords and associations.

Images