System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving audio; means for converting the audio into textual data; means for inputting the textual data into a generative AI model to generate an appropriate answer; means for transmitting the generated answer as textual data; speech synthesis means for converting the textual data into audio; and means for playing the audio.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, many people feel lonely living alone and lack someone to talk to on a daily basis. While dialogue systems utilizing the latest artificial intelligence technology exist, they are difficult for many people to operate, limiting their adoption. There is a particular need for user interfaces that are easy for children and the elderly to use. To solve these issues, a conversational robot that is easy to use and approachable is needed. [Means for solving the problem]

[0005] The present invention provides a system including means for receiving voice data, means for converting the voice data into text data, means for inputting the text data into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis for converting the text data into voice data, and means for playing back the voice data. This system allows users to have natural conversations with an interactive robot, helping to alleviate the loneliness of living alone and the lack of someone to talk to on a daily basis. Furthermore, its simple operation makes it easy for children and the elderly to use. Furthermore, the generative AI model is regularly updated, allowing the system to always provide the latest information and functions.

[0006] "Voice data" refers to the digital audio signal captured by a microphone when a user speaks.

[0007] "Text data" refers to character string data converted from voice data by a voice recognition engine.

[0008] A "generative AI model" is an artificial intelligence algorithm that performs natural language processing and generates appropriate answers based on input text data.

[0009] A "voice recognition engine" is a collection of technologies that converts voice data into text data.

[0010] "Speech synthesis means" refers to technology that converts text data into voice data.

[0011] "Server" means a central computing device that receives voice data and processes it using a speech recognition engine and generative AI models.

[0012] A "terminal" is a device that a user directly operates, transmits voice data to a server, and receives generated text data.

[0013] "Playback means" refers to technology that uses a device such as a speaker to play back audio data as physical sound. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[0036] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This text data is then input into a generative AI model to generate appropriate responses to the user's questions and inquiries. These generated responses are then sent back to the device as text data. The server regularly updates the generative AI model to always provide the latest information and functions.

[0037] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends it to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. Finally, it plays this voice data back to the user through the speaker. This allows the user to receive answers to their questions and inquiries via voice.

[0038] Users can operate the system with simple speech. For example, if they ask, "What's the weather like today?", the device captures the speech and sends it to the server. The server converts this speech into text data and uses a generative AI model to generate an answer based on the weather information. Based on this, the server sends text data such as "The weather is sunny today" back to the device. The device then converts this text data into voice data and tells the user, "The weather is sunny today" over the speaker.

[0039] The system is intended to be installed in the living room for use by the whole family. Its simple and intuitive operation makes it suitable for a wide range of users, from children to the elderly. Regular updates ensure the system always provides the latest information and functions, ensuring long-term use.

[0040] By using this system, users can alleviate the loneliness of living alone and easily obtain information in their daily lives. They can also enjoy new forms of entertainment through dialogue. This system will provide users with new value and experiences, and will greatly enrich their lives in the future.

[0041] The processing flow will be explained below.

[0042] Step 1:

[0043] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[0044] Step 2:

[0045] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[0046] Step 3:

[0047] The device sends the audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[0048] Step 4:

[0049] The server receives the voice data sent from the device and temporarily stores it in the server's storage.

[0050] Step 5:

[0051] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[0052] Step 6:

[0053] The server inputs the text data generated by the speech recognition engine into the generative AI model, which then generates an appropriate answer based on the text data.

[0054] Step 7:

[0055] The server then reformats the text data of the answer returned by the generative AI model and prepares it for transmission to the terminal.

[0056] Step 8:

[0057] The server sends the formatted text data to the device, again via an HTTP response or a WebSocket message.

[0058] Step 9:

[0059] The terminal receives the text data from the server and temporarily stores it.

[0060] Step 10:

[0061] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[0062] Step 11:

[0063] The device plays the generated voice data over a speaker, allowing the user to hear the answer to their question aloud.

[0064] Step 12:

[0065] The user can obtain information by listening to the voice response played back from the terminal. This system allows users to easily obtain information and enjoy dialogue.

[0066] Example 1

[0067] 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."

[0068] Existing speech recognition systems and AI dialogue systems face several challenges in accurately recognizing speech data and generating natural dialogue. In particular, there are issues with misrecognition during the process of converting speech data into text data and the quality of the answers generated by the AI ​​model. Furthermore, the lack of regular model updates makes it difficult to provide appropriate answers based on the latest information. Furthermore, there is a need to efficiently transmit user speech data to a server and provide appropriate answers in real time.

[0069] 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.

[0070] In this invention, the server includes a speech recognition means, a means for inputting text data into a generative AI model to generate an answer, and a means for transmitting the generated answer as text data. This enables accurate conversion of speech data into text and generation of natural dialogue using the generative AI model. Furthermore, the generative AI model generates appropriate answers based on prompts from external input, improving the quality of dialogue with the user. Furthermore, the server periodically updates the generative AI model, allowing it to always provide answers based on the latest information.

[0071] A "means for receiving voice data" is a device or method that captures voice input from a user as a digital signal and receives it in a form that can be processed within the system.

[0072] "Speech recognition means" refers to a device or method that analyzes received voice data and converts it into text format data.

[0073] "Means for inputting text data into a generative AI model to generate an appropriate answer" refers to a device or method for inputting text data into an artificial intelligence model to generate an answer to a user's question or inquiry.

[0074] "Means for transmitting generated answers as text data" refers to a device or method for transmitting answers generated by a generative AI model to another device as text data.

[0075] "Speech synthesis means" refers to a device or method that converts text data into speech data and provides it to the user in a form that can be understood audibly.

[0076] "Means for reproducing audio data" refers to a device or method for reproducing audio data through an output device such as a speaker, and providing information to the user by voice.

[0077] "Means for a terminal device to communicate with a server, transmit voice data to the server, and receive text data from the server" refers to a device or method for communicating to transmit a user's voice data to the server and receive text data from the server.

[0078] "Means for regularly updating a generative AI model" means a device or method for regularly updating a generative AI model based on the latest information and technology, in order to always provide optimal performance and the latest information.

[0079] A "prompt" is an instruction or question input to a generative AI model, and is the input data that enables the model to generate an appropriate answer.

[0080] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[0081] The system of the present invention performs data processing and data calculation using the following hardware and software.

[0082] Hardware and Software Configuration

[0083] 1. Terminal equipment:

[0084] Microphone: A device for capturing audio data from a user.

[0085] Speaker: A device for playing back generated audio data to the user.

[0086] Network interface: A device for communication between a terminal device and a server.

[0087] Speech recognition software: Software that converts voice data into text data. For example, Google Cloud Speech-to-Text API is used.

[0088] 2. Server Device:

[0089] Generative AI model: An artificial intelligence model that inputs text data and generates appropriate answers to user questions or inquiries. For example, OpenAI GPT-4 is used.

[0090] Database: A database for managing update information for generative AI models and speech recognition engines.

[0091] Network interface: A device for communication between a server device and a terminal.

[0092] Specific processing flow

[0093] 1. Capture audio data:

[0094] When a user speaks to the system, "What's the weather like today?", the device's microphone converts the voice signal into digital data, which is then stored in buffer memory.

[0095] 2. Sending audio data:

[0096] The device sends the captured audio data to the server using an HTTP POST request. This audio data is encoded in, for example, WAV format.

[0097] 3. Audio to text conversion:

[0098] The server inputs the received voice data into a voice recognition engine (such as Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, the voice data is converted into text data such as "What's the weather like today?"

[0099] 4. Answer generation using generative AI models:

[0100] The server inputs the generated text data into a generative AI model (OpenAI GPT-4) to generate an appropriate answer. For example, in response to the input "What's the weather like today?", the answer generated is "Today's weather is sunny."

[0101] 5. Submit the generated answer:

[0102] The server sends the generated text data to the terminal via an HTTP POST request.

[0103] 6. Text-to-speech conversion:

[0104] The device converts the received text data into voice data using a speech synthesis engine (such as Amazon Polly), which is then converted into a format that the user can understand.

[0105] 7. Playing audio data:

[0106] The device's speaker will play a voice message saying "Today's weather is sunny" and return the answer to the user.

[0107] Example prompt sentence:

[0108] A user asks about today's weather. Please respond in the following format:

[0109] User asks: "What's the weather like today?"

[0110] answer:

[0111] As described above, this invention allows users to easily obtain information through voice and enjoy natural dialogue. By regularly updating the generative AI model, it is possible to always provide the latest information and improve the quality of the dialogue.

[0112] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0113] Step 1:

[0114] The user speaks to the system, "What's the weather like today?" The device's microphone captures this voice in real time. The voice signal is converted into digital data and stored in buffer memory. The input here is the user's voice, and the output is digital voice data.

[0115] Step 2:

[0116] The device sends the captured audio data to the server in fixed buffer size increments. The audio data (e.g., WAV format) is transferred to the server using an HTTP POST request. The input is the digital audio data, and the output is a transmission request to the server.

[0117] Step 3:

[0118] The server inputs the received voice data into a voice recognition engine (e.g., Google Cloud Speech-to-Text API) and converts it into text data. Specifically, the voice data is analyzed and recognized as text, such as "What's the weather like today?" The input here is digital voice data, and the output is text data.

[0119] Step 4:

[0120] The server inputs the text data obtained from the speech recognition engine into a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate answer to the user's question. The prompt text used is "The user asked about today's weather. Please answer in the following format: User question: 'What's the weather like today?' Answer:". Based on this prompt, the generative AI model generates an answer such as "The weather is sunny today." The input is text data and the prompt text, and the output is the answer text data.

[0121] Step 5:

[0122] The server sends the generated answer text data to the terminal via an HTTP POST request. The input is the generated answer text data, and the output is a request to send to the terminal.

[0123] Step 6:

[0124] The device inputs the received text data into a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. Specifically, it converts the text "Today's weather is sunny" into voice data. The input is text data, and the output is voice data.

[0125] Step 7:

[0126] The terminal speaker plays the converted voice data, and the answer "Today's weather is sunny" is spoken to the user. The input is voice data, and the output is voice playback.

[0127] Through the above steps, the user can have a natural voice interaction with the system.

[0128] (Application example 1)

[0129] 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."

[0130] Conventional food delivery systems have the drawback of being difficult to use, requiring users to input their order details and questions. Furthermore, there is a demand for systems that can respond quickly and appropriately to special requests and questions. Therefore, there is a need for the development of a system that allows users to order food delivery via intuitive voice commands, providing a comfortable user experience.

[0131] 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.

[0132] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for inputting the text data into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis for converting the text data into voice data, means for playing the voice data, means for receiving order details from a user via voice and confirming questions or special requests through natural dialogue, and means for placing a food delivery order based on the voice instructions. This allows users to place food delivery orders and ask questions through intuitive voice operations, improving the user experience.

[0133] "Means for receiving voice data" refers to equipment or software for capturing the voice spoken by the user as digital data.

[0134] The "means for converting voice data into text data" refers to a voice recognition technology that analyzes received voice data and expresses it as text information.

[0135] "Means of inputting into a generative AI model to generate an appropriate response" refers to the process by which the generative AI uses text data obtained through speech recognition to create an appropriate response.

[0136] "Means for transmitting the generated answer as text data" refers to means for transmitting the text-formatted response created by the generation AI to another device or server via communication.

[0137] "Speech synthesis means for converting text data into voice data" is a technology for artificially generating voice using text data.

[0138] "Means for reproducing voice data" refers to a device or software that allows the user to hear the voice data generated by voice synthesis through a speaker.

[0139] "A means of receiving order details via user voice and confirming questions or special requests through natural dialogue" refers to technology that interactively confirms the order details or questions communicated by the user and obtains additional information as needed.

[0140] "Means for ordering food delivery based on voice instructions" refers to technology that automates the ordering process by sending order information to a food delivery service based on the user's voice instructions.

[0141] This invention provides a system that allows users to intuitively order food delivery and ask questions through voice. To implement the invention, a system configuration using mainly a server and a terminal is required.

[0142] Terminal configuration and operation

[0143] The terminal is equipped with a microphone for capturing the user's voice data, a speaker for playing back the synthesized voice data, and a network interface for communicating with the server. Specifically, a mobile terminal such as a smartphone is assumed. The application installed on the terminal has the following functions:

[0144] 1. Audio capture function:

[0145] Use the device's microphone to receive user voice input.

[0146] 2. Voice recognition function:

[0147] The received voice data is converted into text data by a voice recognition engine.

[0148] Software used: Google Speech Recognition API

[0149] 3. Communication functions:

[0150] The converted text data is sent to a server and a response is received from the server.

[0151] Software used: Communication library using HTTP / HTTPS protocol

[0152] Server configuration and operation

[0153] The server analyzes the received text data and generates an appropriate answer using a generative AI model. The generated answer is sent to the device as text data. The server has the following functions implemented:

[0154] 1. Data receiving function:

[0155] Receive text data sent from the terminal.

[0156] 2. Generative AI Model:

[0157] The received text data is used as input to generate an appropriate answer.

[0158] Software used: OpenAI GPT-3

[0159] 3. Data transmission function:

[0160] The generated answer is sent to the terminal as text data.

[0161] Speech synthesis and output

[0162] The device converts the text data received from the server into voice data using a speech synthesis engine and plays a voice response to the user. The software used is Google Text-to-Speech (gTTS).

[0163] Specific examples

[0164] For example, if a user says, "I'd like to order pizza. What's your recommendation?", the process will proceed as follows:

[0165] 1. The device captures the audio data.

[0166] 2. The voice data is converted into text data by a voice recognition engine.

[0167] 3. The text data is sent to the server.

[0168] 4. The server generates an appropriate answer using the generative AI model. Example prompt:

[0169] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[0170] Please write an appropriate answer in Japanese.

[0171] 5. The server generates a response and sends it to the terminal as text data.

[0172] 6. The text data received by the device is converted into voice data using a voice synthesis engine.

[0173] 7. Play the answer back to the user through the speaker.

[0174] In this way, users can smoothly place food delivery orders and ask questions through voice interaction, improving convenience.

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

[0176] Step 1:

[0177] Users can order food delivery or ask questions by speaking into the microphone on their smartphone.

[0178] Input: User's voice data

[0179] Output: Audio data captured by the smartphone microphone

[0180] What happens: The user says, "I'd like to order a pizza. What do you recommend?"

[0181] Step 2:

[0182] The device sends the captured voice data to a voice recognition engine, which converts it into text data.

[0183] Input: Audio data

[0184] Output: Text data as a string

[0185] Specific operation: The voice recognition software (Google Speech Recognition API) on the smartphone analyzes the voice data and generates the corresponding text data: "I'd like to order a pizza. What do you recommend?"

[0186] Step 3:

[0187] The terminal transmits the text data to the server.

[0188] Input: Text data

[0189] Output: Text data sent to the server

[0190] Specific operation: The smartphone sends the generated text data to the server using the HTTP / HTTPS protocol.

[0191] Step 4:

[0192] The server inputs the received text data into a generative AI model to generate an appropriate answer.

[0193] Input: Text data

[0194] Output: Text data of generated answers

[0195] Specific operation: The generative AI model (OpenAI GPT-3) on the server generates an appropriate answer based on the following prompt.

[0196] Example prompt sentence:

[0197] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[0198] Please write an appropriate answer in Japanese.

[0199] Sample generated answer: "Recommended pizzas are the Margherita and seafood pizza."

[0200] Step 5:

[0201] The server sends the generated response to the terminal as text data.

[0202] Input: Text data of generated answers

[0203] Output: Text data sent to the terminal

[0204] Specific operation: The server uses the HTTP / HTTPS protocol to send the generated response to the terminal as text data.

[0205] Step 6:

[0206] The terminal inputs the received text data into a speech synthesis engine and converts it into voice data.

[0207] Input: Text data of generated answers

[0208] Output: Audio data

[0209] Specific operation: The speech synthesis software (Google Text-to-Speech, gTTS) on the smartphone converts the text data "Recommended pizza menu items are Margherita and seafood pizza" into speech data.

[0210] Step 7:

[0211] The device plays the converted audio data to the user through the speaker.

[0212] Input: Audio data

[0213] Output: The audio the user hears

[0214] Specific operation: The smartphone speaker plays audio data, telling the user, "Our recommended pizza menu items are Margherita and seafood pizza."

[0215] 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.

[0216] This invention is a system that receives voice data and uses a speech recognition engine and a generative AI model to have a natural dialogue with the user. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides appropriate responses according to the user's emotions.

[0217] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This converted text data is also input into an emotion engine to recognize the user's emotions. Once the emotion engine identifies the user's emotions (e.g., joy, sadness, anger, etc.), it provides that information to a generative AI model. The generative AI model generates an appropriate response based on the user's emotions based on the text data and emotion information. This generated response is formatted as text data and sent to the device. The server also regularly updates the generative AI model and emotion engine to provide the latest information and functions.

[0218] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends the acquired voice data to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. The device plays the converted voice data through a speaker and also provides feedback to the user on the emotions recognized by the emotion engine. This allows the user to confirm that their emotions are being understood by the system.

[0219] Users can operate the system with simple speech. For example, if a user asks, "What's the weather like today?", the device captures the voice and sends it to the server. The server converts the voice data into text and uses an emotion engine to analyze the user's emotions. Based on the analysis results, the generative AI model generates an appropriate answer and returns text data such as "The weather is sunny today." The device then converts this text data into speech and tells the user, along with the emotion result, "The weather is sunny today" in a bright tone.

[0220] This system is installed in the living room and is designed to blend into the user's daily life. Its simple operation makes it easy to use for a wide range of users, from children to the elderly. Regular updates ensure that the latest information and functions are always provided, ensuring long-term usability. Using this system can alleviate the loneliness of living alone and allow users to easily access information in their daily lives. Furthermore, emotion-based responses allow users to enjoy a richer interactive experience. In this way, the present invention provides users with new value and experiences, supporting a richer lifestyle.

[0221] The processing flow will be explained below.

[0222] Step 1:

[0223] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[0224] Step 2:

[0225] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[0226] Step 3:

[0227] The device sends the captured audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[0228] Step 4:

[0229] The server receives the voice data sent from the terminal and temporarily stores it in the server's storage.

[0230] Step 5:

[0231] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[0232] Step 6:

[0233] The server sends the converted text data to the emotion engine, which analyzes the text data and recognizes the user's emotions (e.g., joy, sadness, anger, etc.).

[0234] Step 7:

[0235] The server inputs the emotional information and text data obtained from the emotion engine into the generative AI model, which then generates an appropriate answer based on the user's emotions and text data.

[0236] Step 8:

[0237] The server formats the answer obtained from the generative AI model as text data and prepares it to be sent to the device.

[0238] Step 9:

[0239] The server sends the formatted text data and emotion information to the device, again via HTTP responses or WebSocket messages.

[0240] Step 10:

[0241] The device receives the text data and emotion information from the server and temporarily stores them.

[0242] Step 11:

[0243] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[0244] Step 12:

[0245] The device then plays the generated voice data over a speaker, allowing the user to hear the answer to their question. The tone and intonation of the voice may be adjusted based on the emotional information.

[0246] Step 13:

[0247] The user listens to the voice response played back from the device. The emotional feedback confirms that their feelings are understood. This system allows the user to feel emotional satisfaction through the interaction.

[0248] Example 2

[0249] 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."

[0250] Conventional voice recognition systems do not take user emotions into account, making dialogue mechanical and difficult to achieve natural communication. Furthermore, the lack of regular updates to generative AI models and emotion recognition engines has led to problems with the system's accuracy and response quality declining. Furthermore, users are unable to confirm whether their emotions are being understood, which leads to a decline in dialogue satisfaction.

[0251] 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.

[0252] In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion recognition engine to analyze the user's emotions, means for inputting the text data and emotion information into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis to convert the text data into voice data, means for playing the voice data, and means for feeding back the emotion recognition results to the user. This enables natural dialogue according to the user's emotions, and regular updates can maintain the accuracy and quality of the system's responses, improving user satisfaction.

[0253] "Voice Data" means a digital representation of voice information uttered by a User.

[0254] "Text data" is digital information that is audio data converted into text format.

[0255] An "emotion recognition engine" is a software module that analyzes text data to identify a user's emotions.

[0256] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers based on input text data and emotional information.

[0257] A "voice recognition engine" is a software module for converting voice data into text data.

[0258] A "speech synthesis means" is a software module and hardware device for converting text data into speech data.

[0259] A "terminal device" is a hardware device that captures voice data from a user and transmits it to a server.

[0260] The "server" is a central processing unit that processes voice data and performs emotion recognition and response generation.

[0261] A "prompt" is an instruction or question to be input into a generative AI model.

[0262] This invention is a system that receives voice data and uses a generative AI model and an emotion recognition engine to engage in natural dialogue with the user. It also recognizes the user's emotions and provides appropriate responses based on those emotions.

[0263] Specifically, the following hardware and software are used.

[0264] Hardware:

[0265] 1. Terminal equipment:

[0266] Microphone: Captures the user's voice.

[0267] Speaker: Provides a spoken response to the user.

[0268] Communication module: Performs data communication with the server.

[0269] 2. Server:

[0270] Central processing unit: processes voice data, recognizes emotions, and generates responses.

[0271] software:

[0272] 1. Speech recognition engine (e.g. Google Speech-to-Text API)

[0273] Convert the audio data into text data.

[0274] 2. Emotion recognition engine (e.g., Microsoft Azure Text Analytics)

[0275] Analyze text data to identify user sentiment.

[0276] 3. Generative AI models (e.g., OpenAI GPT-4)

[0277] Generate appropriate answers based on text data and emotional information.

[0278] 4. Speech synthesis engine (e.g. IBM Watson Text-to-Speech)

[0279] Convert text data into audio data.

[0280] Detailed instructions on how to do this:

[0281] The server uses a voice recognition engine to convert the voice data sent from the terminal device into text data. The converted text data is analyzed by an emotion recognition engine to identify the user's emotion. This text data and emotion information are then input into a generative AI model to generate an appropriate answer. This answer is formatted as text data and sent from the server to the terminal device. The terminal device then converts the received text data into voice data using a voice synthesis engine and responds to the user audibly using a speaker.

[0282] Examples:

[0283] When a user asks, "What's the weather today?", the terminal device captures the voice and sends it to the server. The server uses a speech recognition engine to convert this voice data into text data, generating the text "What's the weather today?". It then uses an emotion recognition engine to identify the user's emotion (e.g., "interested") from this text and provides that emotion information to the generative AI model. The generative AI model then generates an appropriate answer based on the following prompt sentence, for example:

[0284] Example prompt sentence:

[0285] "A user asks, 'What's the weather like today?' The user's emotion is 'interested.' Generate an appropriate answer based on the user's emotion."

[0286] The generative AI model generates a response such as "Today's weather is sunny," which the server sends to the terminal device as text data. The terminal device then uses a speech synthesis engine to convert the text data into voice data, responding to the user in a bright tone: "Today's weather is sunny."

[0287] This system allows users to enjoy a natural, emotion-sensitive conversational experience, helping them to obtain information on a daily basis and alleviating the loneliness of living alone. Furthermore, feedback on emotion recognition results allows users to confirm that their emotions are being understood by the system.

[0288] Furthermore, the server regularly updates the generative AI model and emotion recognition engine, allowing it to continually provide the latest information and advanced functionality, thereby increasing the long-term utility value of the system and improving user satisfaction.

[0289] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0290] Step 1:

[0291] The user provides voice input.

[0292] Specifically, the user speaks a question such as "What's the weather like today?" This speech is input into the system. The input data is the user's speech, and the output is voice data.

[0293] Step 2:

[0294] The device captures the audio data.

[0295] The device's microphone captures the user's voice and records it as digital voice data. During this process, voice data is input and the voice data is output as is. For example, the voice saying "What's the weather like today?" is captured in digital form.

[0296] Step 3:

[0297] The terminal transmits the voice data to the server.

[0298] The communication module of the device sends the captured audio data to the server. The input here is the audio data of the device, and the output is the transmission of audio data to the server. Specifically, the device sends the audio data to the server using an HTTP request.

[0299] Step 4:

[0300] The server converts the voice data into text data.

[0301] The server converts the received voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). The input of this step is voice data, and the output is text data. For example, the voice data "What's the weather like today?" is converted into the text data "What's the weather like today?"

[0302] Step 5:

[0303] The server analyzes emotions using an emotion recognition engine.

[0304] The server inputs the text data into an emotion recognition engine (e.g., Microsoft Azure Text Analytics) to analyze the user's emotions. The input is text data, and the output is emotion information. Specifically, the emotion "interest" is identified from the text data "What's the weather like today?"

[0305] Step 6:

[0306] The server inputs a prompt into the generative AI model and generates an appropriate answer.

[0307] The server inputs a prompt based on text data and emotional information into a generative AI model (e.g., OpenAI GPT-4) and generates an appropriate answer. The input for this step is text data and emotional information, and the output is the generated answer text data. As a specific example, the prompt "The user asked, 'What's the weather like today?' The user's emotion at that time is 'interested.' Please generate an appropriate answer based on the user's emotion." is input, and the AI ​​model generates the text "The weather is sunny today."

[0308] Step 7:

[0309] The server transmits the generated text data to the terminal.

[0310] The server sends the generated answer to the terminal as text data. The input is the generated text data, and the output is the transmission of the text data to the terminal. Specifically, the text data "Today's weather is sunny" is sent to the terminal.

[0311] Step 8:

[0312] The terminal converts the text data into audio data.

[0313] The device uses a speech synthesis engine (e.g., IBM Watson Text-to-Speech) to convert text data into voice data. The input is text data and the output is voice data. For example, the text data "The weather is sunny today" is converted into voice data.

[0314] Step 9:

[0315] The device responds to the user with voice and also provides feedback on the emotional results.

[0316] The device plays the converted voice data through a speaker and also feeds back the emotional result to the user. The input is the voice data and the emotional result, and the output is a voice response to the user and emotional feedback. As a concrete example, the device tells the user, "Today's weather is sunny," in a bright tone.

[0317] Step 10:

[0318] The server periodically updates the generative AI model and emotion recognition engine.

[0319] The server regularly updates the generative AI model and emotion recognition engine to maintain the system's accuracy and response quality. The input is a new dataset and algorithm, and the output is an updated generative AI model and emotion recognition engine. Specifically, the server trains the generative AI model using a new dataset once a month and updates the emotion recognition engine algorithm to the latest version.

[0320] (Application example 2)

[0321] 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."

[0322] Conventional voice dialogue systems simply convert voice data into text and generate responses using a generative AI model, limiting their ability to provide personalized responses that take the user's emotions into account. In brick-and-mortar stores in particular, appropriate responses based on customer emotions are often required, and in this regard, existing technologies are inadequate. Furthermore, delays in information delivery during the voice recognition and response generation process, as well as inaccurate emotion recognition, are factors that reduce the quality of customer service. It is necessary to solve these problems.

[0323] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion engine to recognize the user's emotion, means for inputting the text data and emotion information into a generative AI model to generate a response in an appropriate tone, means for transmitting the generated response as text data, voice synthesis means for converting the text data into voice data, and means for playing back the voice data. This makes it possible to quickly provide an appropriate response according to the user's emotion and improve customer satisfaction.

[0324] "Audio data" refers to data in which audio is recorded in digital format.

[0325] "Text data" is data expressed in the form of a string of characters.

[0326] A "generative AI model" is an artificial intelligence model that analyzes user input data and generates appropriate answers based on it.

[0327] An "emotion engine" is a system that estimates a user's emotions from text and voice data.

[0328] "Appropriate tone" refers to the appropriate speaking style and tone of voice that corresponds to the user's emotions.

[0329] A "voice recognition engine" is a system that analyzes voice data and converts it into text data.

[0330] "Speech synthesis means" is a technology that converts text data into voice data.

[0331] "User" refers to an individual or group that uses the system.

[0332] A "server" is a central computing device that performs various processes for the system.

[0333] A "terminal device" is a device operated by a user that captures and plays audio.

[0334] System Configuration

[0335] The embodiment of the invention employs the following system configuration: The system is mainly composed of three entities: a server, a terminal device, and a user, and each function operates in cooperation with the others.

[0336] Hardware Configuration

[0337] Server: A high-performance computing device that utilizes cloud or on-premise servers.

[0338] Terminal Device: A user interface device such as a smartphone, smart glasses, or tablet.

[0339] Microphone: A device for inputting the user's voice.

[0340] Speaker: A device for playing back generated audio data.

[0341] Software Configuration

[0342] Voice capture module: Installed on the terminal device, it captures the user's voice data.

[0343] Speech recognition engine: Used to convert voice data into text data (e.g., Google Cloud Speech-to-Text).

[0344] Emotion engine: Recognizes user emotions from text data (e.g., IBM Watson Tone Analyzer).

[0345] Generative AI models: Generate appropriate answers based on user input and emotional information (e.g., OpenAI GPT-4).

[0346] Speech synthesis engine: Converts the generated text data into speech data (e.g., Amazon Polly).

[0347] System Operation Overview

[0348] Step 1: Capture audio

[0349] The voice capture module of the terminal device uses a microphone to capture the user's voice data, which is then transmitted to the server in real time.

[0350] Step 2: Voice Recognition

[0351] The server converts the received voice data into text data using a voice recognition engine. For example, if a user says, "I want new shoes," the voice data is converted into text data saying, "I want new shoes."

[0352] Step 3: Emotion Recognition

[0353] The converted text data is then fed into an emotion engine, which analyzes the user's emotions. For example, if the user's comments are recognized as "excited," that emotional information is provided to the generative AI model.

[0354] Step 4: Answer Generation

[0355] The generative AI model generates an appropriate response based on the text data and emotional information. For example, based on the text data "I want new shoes" and the emotional information "I'm excited," the generative AI model generates the following response: "Thank you for visiting! New shoes, then let me show you our latest collection!"

[0356] Step 5: Text-to-Speech

[0357] A speech synthesis engine is used to convert the generated response text data into speech data, which generates a cheerful voice saying, "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!"

[0358] Step 6: Play back the response

[0359] The terminal device plays back the voice data generated by the voice synthesis engine, and the response is sent to the user through the speaker.

[0360] Prompt Sentence Examples

[0361] The customer says, "I want new shoes." The customer is now excited. An appropriate response would be:

[0362] This allows users to receive appropriate responses based on their emotions, improving the customer service experience in physical stores.

[0363] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0364] Step 1:

[0365] The terminal uses the voice capture module to capture the user's voice data through the microphone. The input is the user's voice, and the output is the voice data. This voice data is sent to the server in real time.

[0366] Step 2:

[0367] The voice data received by the server is converted into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text). The input is voice data and the output is text data. For example, voice data saying "I want new shoes" is converted into text data saying "I want new shoes."

[0368] Step 3:

[0369] The converted text data is input into the server's emotion engine (e.g., IBM Watson Tone Analyzer), which analyzes the user's emotions. The input is text data, and the output is emotional information. For example, the emotional information "excited" can be obtained from the text data "I want new shoes."

[0370] Step 4:

[0371] The server inputs text data and emotional information into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates an appropriate answer based on this information. The input is text data and emotional information, and the output is response text data. For example, from the information "I want new shoes" and "Excited," the response text data "Thank you for visiting! New shoes, let me show you our latest collection!" is generated.

[0372] Step 5:

[0373] The generated response text data is input to a speech synthesis engine (e.g., Amazon Polly) and converted into voice data. The input is the response text data, and the output is the response voice data. The speech synthesis engine converts the text data into voice data with a bright tone.

[0374] Step 6:

[0375] The device plays back the response voice data generated by the speech synthesis engine. The input is the response voice data, and the output is a voice response that the user can hear. For example, a voice saying "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!" is played through the device's speaker.

[0376] The above is the specific processing flow of the program of the system that realizes the application example.

[0377] 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.

[0378] 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.

[0379] 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.

[0380] [Second embodiment]

[0381] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0382] 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.

[0383] 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).

[0384] 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.

[0385] 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.

[0386] 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).

[0387] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0388] 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.

[0389] 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.

[0390] 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.

[0391] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0392] 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."

[0393] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[0394] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This text data is then input into a generative AI model to generate appropriate responses to the user's questions and inquiries. These generated responses are then sent back to the device as text data. The server regularly updates the generative AI model to always provide the latest information and functions.

[0395] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends it to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. Finally, it plays this voice data back to the user through the speaker. This allows the user to receive answers to their questions and inquiries via voice.

[0396] Users can operate the system with simple speech. For example, if they ask, "What's the weather like today?", the device captures the speech and sends it to the server. The server converts this speech into text data and uses a generative AI model to generate an answer based on the weather information. Based on this, the server sends text data such as "The weather is sunny today" back to the device. The device then converts this text data into voice data and tells the user, "The weather is sunny today" over the speaker.

[0397] The system is intended to be installed in the living room for use by the whole family. Its simple and intuitive operation makes it suitable for a wide range of users, from children to the elderly. Regular updates ensure the system always provides the latest information and functions, ensuring long-term use.

[0398] By using this system, users can alleviate the loneliness of living alone and easily obtain information in their daily lives. They can also enjoy new forms of entertainment through dialogue. This system will provide users with new value and experiences, and will greatly enrich their lives in the future.

[0399] The processing flow will be explained below.

[0400] Step 1:

[0401] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[0402] Step 2:

[0403] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[0404] Step 3:

[0405] The device sends the audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[0406] Step 4:

[0407] The server receives the voice data sent from the device and temporarily stores it in the server's storage.

[0408] Step 5:

[0409] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[0410] Step 6:

[0411] The server inputs the text data generated by the speech recognition engine into the generative AI model, which then generates an appropriate answer based on the text data.

[0412] Step 7:

[0413] The server then reformats the text data of the answer returned by the generative AI model and prepares it for transmission to the terminal.

[0414] Step 8:

[0415] The server sends the formatted text data to the device, again via an HTTP response or a WebSocket message.

[0416] Step 9:

[0417] The terminal receives the text data from the server and temporarily stores it.

[0418] Step 10:

[0419] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[0420] Step 11:

[0421] The device plays the generated voice data over a speaker, allowing the user to hear the answer to their question aloud.

[0422] Step 12:

[0423] The user can obtain information by listening to the voice response played back from the terminal. This system allows users to easily obtain information and enjoy dialogue.

[0424] Example 1

[0425] 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."

[0426] Existing speech recognition systems and AI dialogue systems face several challenges in accurately recognizing speech data and generating natural dialogue. In particular, there are issues with misrecognition during the process of converting speech data into text data and the quality of the answers generated by the AI ​​model. Furthermore, the lack of regular model updates makes it difficult to provide appropriate answers based on the latest information. Furthermore, there is a need to efficiently transmit user speech data to a server and provide appropriate answers in real time.

[0427] 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.

[0428] In this invention, the server includes a speech recognition means, a means for inputting text data into a generative AI model to generate an answer, and a means for transmitting the generated answer as text data. This enables accurate conversion of speech data into text and generation of natural dialogue using the generative AI model. Furthermore, the generative AI model generates appropriate answers based on prompts from external input, improving the quality of dialogue with the user. Furthermore, the server periodically updates the generative AI model, allowing it to always provide answers based on the latest information.

[0429] A "means for receiving voice data" is a device or method that captures voice input from a user as a digital signal and receives it in a form that can be processed within the system.

[0430] "Speech recognition means" refers to a device or method that analyzes received voice data and converts it into text format data.

[0431] "Means for inputting text data into a generative AI model to generate an appropriate answer" refers to a device or method for inputting text data into an artificial intelligence model to generate an answer to a user's question or inquiry.

[0432] "Means for transmitting generated answers as text data" refers to a device or method for transmitting answers generated by a generative AI model to another device as text data.

[0433] "Speech synthesis means" refers to a device or method that converts text data into speech data and provides it to the user in a form that can be understood audibly.

[0434] "Means for reproducing audio data" refers to a device or method for reproducing audio data through an output device such as a speaker, and providing information to the user by voice.

[0435] "Means for a terminal device to communicate with a server, transmit voice data to the server, and receive text data from the server" refers to a device or method for communicating to transmit a user's voice data to the server and receive text data from the server.

[0436] "Means for regularly updating a generative AI model" means a device or method for regularly updating a generative AI model based on the latest information and technology, in order to always provide optimal performance and the latest information.

[0437] A "prompt" is an instruction or question input to a generative AI model, and is the input data that enables the model to generate an appropriate answer.

[0438] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[0439] The system of the present invention performs data processing and data calculation using the following hardware and software.

[0440] Hardware and Software Configuration

[0441] 1. Terminal equipment:

[0442] Microphone: A device for capturing audio data from a user.

[0443] Speaker: A device for playing back generated audio data to the user.

[0444] Network interface: A device for communication between a terminal device and a server.

[0445] Speech recognition software: Software that converts voice data into text data. For example, Google Cloud Speech-to-Text API is used.

[0446] 2. Server Device:

[0447] Generative AI model: An artificial intelligence model that inputs text data and generates appropriate answers to user questions or inquiries. For example, OpenAI GPT-4 is used.

[0448] Database: A database for managing update information for generative AI models and speech recognition engines.

[0449] Network interface: A device for communication between a server device and a terminal.

[0450] Specific processing flow

[0451] 1. Capture audio data:

[0452] When a user speaks to the system, "What's the weather like today?", the device's microphone converts the voice signal into digital data, which is then stored in buffer memory.

[0453] 2. Sending audio data:

[0454] The device sends the captured audio data to the server using an HTTP POST request. This audio data is encoded in, for example, WAV format.

[0455] 3. Audio to text conversion:

[0456] The server inputs the received voice data into a voice recognition engine (such as Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, the voice data is converted into text data such as "What's the weather like today?"

[0457] 4. Answer generation using generative AI models:

[0458] The server inputs the generated text data into a generative AI model (OpenAI GPT-4) to generate an appropriate answer. For example, in response to the input "What's the weather like today?", the answer generated is "Today's weather is sunny."

[0459] 5. Submit the generated answer:

[0460] The server sends the generated text data to the terminal via an HTTP POST request.

[0461] 6. Text-to-speech conversion:

[0462] The device converts the received text data into voice data using a speech synthesis engine (such as Amazon Polly), which is then converted into a format that the user can understand.

[0463] 7. Playing audio data:

[0464] The device's speaker will play a voice message saying "Today's weather is sunny" and return the answer to the user.

[0465] Example prompt sentence:

[0466] A user asks about today's weather. Please respond in the following format:

[0467] User asks: "What's the weather like today?"

[0468] answer:

[0469] As described above, this invention allows users to easily obtain information through voice and enjoy natural dialogue. By regularly updating the generative AI model, it is possible to always provide the latest information and improve the quality of the dialogue.

[0470] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0471] Step 1:

[0472] The user speaks to the system, "What's the weather like today?" The device's microphone captures this voice in real time. The voice signal is converted into digital data and stored in buffer memory. The input here is the user's voice, and the output is digital voice data.

[0473] Step 2:

[0474] The device sends the captured audio data to the server in fixed buffer size increments. The audio data (e.g., WAV format) is transferred to the server using an HTTP POST request. The input is the digital audio data, and the output is a transmission request to the server.

[0475] Step 3:

[0476] The server inputs the received voice data into a voice recognition engine (e.g., Google Cloud Speech-to-Text API) and converts it into text data. Specifically, the voice data is analyzed and recognized as text, such as "What's the weather like today?" The input here is digital voice data, and the output is text data.

[0477] Step 4:

[0478] The server inputs the text data obtained from the speech recognition engine into a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate answer to the user's question. The prompt text used is "The user asked about today's weather. Please answer in the following format: User question: 'What's the weather like today?' Answer:". Based on this prompt, the generative AI model generates an answer such as "The weather is sunny today." The input is text data and the prompt text, and the output is the answer text data.

[0479] Step 5:

[0480] The server sends the generated answer text data to the terminal via an HTTP POST request. The input is the generated answer text data, and the output is a request to send to the terminal.

[0481] Step 6:

[0482] The device inputs the received text data into a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. Specifically, it converts the text "Today's weather is sunny" into voice data. The input is text data, and the output is voice data.

[0483] Step 7:

[0484] The terminal speaker plays the converted voice data, and the answer "Today's weather is sunny" is spoken to the user. The input is voice data, and the output is voice playback.

[0485] Through the above steps, the user can have a natural voice interaction with the system.

[0486] (Application example 1)

[0487] 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."

[0488] Conventional food delivery systems have the drawback of being difficult to use, requiring users to input their order details and questions. Furthermore, there is a demand for systems that can respond quickly and appropriately to special requests and questions. Therefore, there is a need for the development of a system that allows users to order food delivery via intuitive voice commands, providing a comfortable user experience.

[0489] 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.

[0490] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for inputting the text data into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis for converting the text data into voice data, means for playing the voice data, means for receiving order details from a user via voice and confirming questions or special requests through natural dialogue, and means for placing a food delivery order based on the voice instructions. This allows users to place food delivery orders and ask questions through intuitive voice operations, improving the user experience.

[0491] "Means for receiving voice data" refers to equipment or software for capturing the voice spoken by the user as digital data.

[0492] The "means for converting voice data into text data" refers to a voice recognition technology that analyzes received voice data and expresses it as text information.

[0493] "Means of inputting into a generative AI model to generate an appropriate response" refers to the process by which the generative AI uses text data obtained through speech recognition to create an appropriate response.

[0494] "Means for transmitting the generated answer as text data" refers to means for transmitting the text-formatted response created by the generation AI to another device or server via communication.

[0495] "Speech synthesis means for converting text data into voice data" is a technology for artificially generating voice using text data.

[0496] "Means for reproducing voice data" refers to a device or software that allows the user to hear the voice data generated by voice synthesis through a speaker.

[0497] "A means of receiving order details via user voice and confirming questions or special requests through natural dialogue" refers to technology that interactively confirms the order details or questions communicated by the user and obtains additional information as needed.

[0498] "Means for ordering food delivery based on voice instructions" refers to technology that automates the ordering process by sending order information to a food delivery service based on the user's voice instructions.

[0499] This invention provides a system that allows users to intuitively order food delivery and ask questions through voice. To implement the invention, a system configuration using mainly a server and a terminal is required.

[0500] Terminal configuration and operation

[0501] The terminal is equipped with a microphone for capturing the user's voice data, a speaker for playing back the synthesized voice data, and a network interface for communicating with the server. Specifically, a mobile terminal such as a smartphone is assumed. The application installed on the terminal has the following functions:

[0502] 1. Audio capture function:

[0503] Use the device's microphone to receive user voice input.

[0504] 2. Voice recognition function:

[0505] The received voice data is converted into text data by a voice recognition engine.

[0506] Software used: Google Speech Recognition API

[0507] 3. Communication functions:

[0508] The converted text data is sent to a server and a response is received from the server.

[0509] Software used: Communication library using HTTP / HTTPS protocol

[0510] Server configuration and operation

[0511] The server analyzes the received text data and generates an appropriate answer using a generative AI model. The generated answer is sent to the device as text data. The server has the following functions implemented:

[0512] 1. Data receiving function:

[0513] Receive text data sent from the terminal.

[0514] 2. Generative AI Model:

[0515] The received text data is used as input to generate an appropriate answer.

[0516] Software used: OpenAI GPT-3

[0517] 3. Data transmission function:

[0518] The generated answer is sent to the terminal as text data.

[0519] Speech synthesis and output

[0520] The device converts the text data received from the server into voice data using a speech synthesis engine and plays a voice response to the user. The software used is Google Text-to-Speech (gTTS).

[0521] Specific examples

[0522] For example, if a user says, "I'd like to order pizza. What's your recommendation?", the process will proceed as follows:

[0523] 1. The device captures the audio data.

[0524] 2. The voice data is converted into text data by a voice recognition engine.

[0525] 3. The text data is sent to the server.

[0526] 4. The server generates an appropriate answer using the generative AI model. Example prompt:

[0527] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[0528] Please write an appropriate answer in Japanese.

[0529] 5. The server generates a response and sends it to the terminal as text data.

[0530] 6. The text data received by the device is converted into voice data using a voice synthesis engine.

[0531] 7. Play the answer back to the user through the speaker.

[0532] In this way, users can smoothly place food delivery orders and ask questions through voice interaction, improving convenience.

[0533] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0534] Step 1:

[0535] Users can order food delivery or ask questions by speaking into the microphone on their smartphone.

[0536] Input: User's voice data

[0537] Output: Audio data captured by the smartphone microphone

[0538] What happens: The user says, "I'd like to order a pizza. What do you recommend?"

[0539] Step 2:

[0540] The device sends the captured voice data to a voice recognition engine, which converts it into text data.

[0541] Input: Audio data

[0542] Output: Text data as a string

[0543] Specific operation: The voice recognition software (Google Speech Recognition API) on the smartphone analyzes the voice data and generates the corresponding text data: "I'd like to order a pizza. What do you recommend?"

[0544] Step 3:

[0545] The terminal transmits the text data to the server.

[0546] Input: Text data

[0547] Output: Text data sent to the server

[0548] Specific operation: The smartphone sends the generated text data to the server using the HTTP / HTTPS protocol.

[0549] Step 4:

[0550] The server inputs the received text data into a generative AI model to generate an appropriate answer.

[0551] Input: Text data

[0552] Output: Text data of generated answers

[0553] Specific operation: The generative AI model (OpenAI GPT-3) on the server generates an appropriate answer based on the following prompt.

[0554] Example prompt sentence:

[0555] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[0556] Please write an appropriate answer in Japanese.

[0557] Sample generated answer: "Recommended pizzas are the Margherita and seafood pizza."

[0558] Step 5:

[0559] The server sends the generated response to the terminal as text data.

[0560] Input: Text data of generated answers

[0561] Output: Text data sent to the terminal

[0562] Specific operation: The server uses the HTTP / HTTPS protocol to send the generated response to the terminal as text data.

[0563] Step 6:

[0564] The terminal inputs the received text data into a speech synthesis engine and converts it into voice data.

[0565] Input: Text data of generated answers

[0566] Output: Audio data

[0567] Specific operation: The speech synthesis software (Google Text-to-Speech, gTTS) on the smartphone converts the text data "Recommended pizza menu items are Margherita and seafood pizza" into speech data.

[0568] Step 7:

[0569] The device plays the converted audio data to the user through the speaker.

[0570] Input: Audio data

[0571] Output: The audio the user hears

[0572] Specific operation: The smartphone speaker plays audio data, telling the user, "Our recommended pizza menu items are Margherita and seafood pizza."

[0573] 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.

[0574] This invention is a system that receives voice data and uses a speech recognition engine and a generative AI model to have a natural dialogue with the user. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides appropriate responses according to the user's emotions.

[0575] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This converted text data is also input into an emotion engine to recognize the user's emotions. Once the emotion engine identifies the user's emotions (e.g., joy, sadness, anger, etc.), it provides that information to a generative AI model. The generative AI model generates an appropriate response based on the user's emotions based on the text data and emotion information. This generated response is formatted as text data and sent to the device. The server also regularly updates the generative AI model and emotion engine to provide the latest information and functions.

[0576] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends the acquired voice data to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. The device plays the converted voice data through a speaker and also provides feedback to the user on the emotions recognized by the emotion engine. This allows the user to confirm that their emotions are being understood by the system.

[0577] Users can operate the system with simple speech. For example, if a user asks, "What's the weather like today?", the device captures the voice and sends it to the server. The server converts the voice data into text and uses an emotion engine to analyze the user's emotions. Based on the analysis results, the generative AI model generates an appropriate answer and returns text data such as "The weather is sunny today." The device then converts this text data into speech and tells the user, along with the emotion result, "The weather is sunny today" in a bright tone.

[0578] This system is installed in the living room and is designed to blend into the user's daily life. Its simple operation makes it easy to use for a wide range of users, from children to the elderly. Regular updates ensure that the latest information and functions are always provided, ensuring long-term usability. Using this system can alleviate the loneliness of living alone and allow users to easily access information in their daily lives. Furthermore, emotion-based responses allow users to enjoy a richer interactive experience. In this way, the present invention provides users with new value and experiences, supporting a richer lifestyle.

[0579] The processing flow will be explained below.

[0580] Step 1:

[0581] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[0582] Step 2:

[0583] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[0584] Step 3:

[0585] The device sends the captured audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[0586] Step 4:

[0587] The server receives the voice data sent from the terminal and temporarily stores it in the server's storage.

[0588] Step 5:

[0589] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[0590] Step 6:

[0591] The server sends the converted text data to the emotion engine, which analyzes the text data and recognizes the user's emotions (e.g., joy, sadness, anger, etc.).

[0592] Step 7:

[0593] The server inputs the emotional information and text data obtained from the emotion engine into the generative AI model, which then generates an appropriate answer based on the user's emotions and text data.

[0594] Step 8:

[0595] The server formats the answer obtained from the generative AI model as text data and prepares it to be sent to the device.

[0596] Step 9:

[0597] The server sends the formatted text data and emotion information to the device, again via HTTP responses or WebSocket messages.

[0598] Step 10:

[0599] The device receives the text data and emotion information from the server and temporarily stores them.

[0600] Step 11:

[0601] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[0602] Step 12:

[0603] The device then plays the generated voice data over a speaker, allowing the user to hear the answer to their question. The tone and intonation of the voice may be adjusted based on the emotional information.

[0604] Step 13:

[0605] The user listens to the voice response played back from the device. The emotional feedback confirms that their feelings are understood. This system allows the user to feel emotional satisfaction through the interaction.

[0606] Example 2

[0607] 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."

[0608] Conventional voice recognition systems do not take user emotions into account, making dialogue mechanical and difficult to achieve natural communication. Furthermore, the lack of regular updates to generative AI models and emotion recognition engines has led to problems with the system's accuracy and response quality declining. Furthermore, users are unable to confirm whether their emotions are being understood, which leads to a decline in dialogue satisfaction.

[0609] 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.

[0610] In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion recognition engine to analyze the user's emotions, means for inputting the text data and emotion information into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis to convert the text data into voice data, means for playing the voice data, and means for feeding back the emotion recognition results to the user. This enables natural dialogue according to the user's emotions, and regular updates can maintain the accuracy and quality of the system's responses, improving user satisfaction.

[0611] "Voice Data" means a digital representation of voice information uttered by a User.

[0612] "Text data" is digital information that is audio data converted into text format.

[0613] An "emotion recognition engine" is a software module that analyzes text data to identify a user's emotions.

[0614] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers based on input text data and emotional information.

[0615] A "voice recognition engine" is a software module for converting voice data into text data.

[0616] A "speech synthesis means" is a software module and hardware device for converting text data into speech data.

[0617] A "terminal device" is a hardware device that captures voice data from a user and transmits it to a server.

[0618] The "server" is a central processing unit that processes voice data and performs emotion recognition and response generation.

[0619] A "prompt" is an instruction or question to be input into a generative AI model.

[0620] This invention is a system that receives voice data and uses a generative AI model and an emotion recognition engine to engage in natural dialogue with the user. It also recognizes the user's emotions and provides appropriate responses based on those emotions.

[0621] Specifically, the following hardware and software are used.

[0622] Hardware:

[0623] 1. Terminal equipment:

[0624] Microphone: Captures the user's voice.

[0625] Speaker: Provides a spoken response to the user.

[0626] Communication module: Performs data communication with the server.

[0627] 2. Server:

[0628] Central processing unit: processes voice data, recognizes emotions, and generates responses.

[0629] software:

[0630] 1. Speech recognition engine (e.g. Google Speech-to-Text API)

[0631] Convert the audio data into text data.

[0632] 2. Emotion recognition engine (e.g., Microsoft Azure Text Analytics)

[0633] Analyze text data to identify user sentiment.

[0634] 3. Generative AI models (e.g., OpenAI GPT-4)

[0635] Generate appropriate answers based on text data and emotional information.

[0636] 4. Speech synthesis engine (e.g. IBM Watson Text-to-Speech)

[0637] Convert text data into audio data.

[0638] Detailed instructions on how to do this:

[0639] The server uses a voice recognition engine to convert the voice data sent from the terminal device into text data. The converted text data is analyzed by an emotion recognition engine to identify the user's emotion. This text data and emotion information are then input into a generative AI model to generate an appropriate answer. This answer is formatted as text data and sent from the server to the terminal device. The terminal device then converts the received text data into voice data using a voice synthesis engine and responds to the user audibly using a speaker.

[0640] Examples:

[0641] When a user asks, "What's the weather today?", the terminal device captures the voice and sends it to the server. The server uses a speech recognition engine to convert this voice data into text data, generating the text "What's the weather today?". It then uses an emotion recognition engine to identify the user's emotion (e.g., "interested") from this text and provides that emotion information to the generative AI model. The generative AI model then generates an appropriate answer based on the following prompt sentence, for example:

[0642] Example prompt sentence:

[0643] "A user asks, 'What's the weather like today?' The user's emotion is 'interested.' Generate an appropriate answer based on the user's emotion."

[0644] The generative AI model generates a response such as "Today's weather is sunny," which the server sends to the terminal device as text data. The terminal device then uses a speech synthesis engine to convert the text data into voice data, responding to the user in a bright tone: "Today's weather is sunny."

[0645] This system allows users to enjoy a natural, emotion-sensitive conversational experience, helping them to obtain information on a daily basis and alleviating the loneliness of living alone. Furthermore, feedback on emotion recognition results allows users to confirm that their emotions are being understood by the system.

[0646] Furthermore, the server regularly updates the generative AI model and emotion recognition engine, allowing it to continually provide the latest information and advanced functionality, thereby increasing the long-term utility value of the system and improving user satisfaction.

[0647] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0648] Step 1:

[0649] The user provides voice input.

[0650] Specifically, the user speaks a question such as "What's the weather like today?" This speech is input into the system. The input data is the user's speech, and the output is voice data.

[0651] Step 2:

[0652] The device captures the audio data.

[0653] The device's microphone captures the user's voice and records it as digital voice data. During this process, voice data is input and the voice data is output as is. For example, the voice saying "What's the weather like today?" is captured in digital form.

[0654] Step 3:

[0655] The terminal transmits the voice data to the server.

[0656] The communication module of the device sends the captured audio data to the server. The input here is the audio data of the device, and the output is the transmission of audio data to the server. Specifically, the device sends the audio data to the server using an HTTP request.

[0657] Step 4:

[0658] The server converts the voice data into text data.

[0659] The server converts the received voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). The input of this step is voice data, and the output is text data. For example, the voice data "What's the weather like today?" is converted into the text data "What's the weather like today?"

[0660] Step 5:

[0661] The server analyzes emotions using an emotion recognition engine.

[0662] The server inputs the text data into an emotion recognition engine (e.g., Microsoft Azure Text Analytics) to analyze the user's emotions. The input is text data, and the output is emotion information. Specifically, the emotion "interest" is identified from the text data "What's the weather like today?"

[0663] Step 6:

[0664] The server inputs a prompt into the generative AI model and generates an appropriate answer.

[0665] The server inputs a prompt based on text data and emotional information into a generative AI model (e.g., OpenAI GPT-4) and generates an appropriate answer. The input for this step is text data and emotional information, and the output is the generated answer text data. As a specific example, the prompt "The user asked, 'What's the weather like today?' The user's emotion at that time is 'interested.' Please generate an appropriate answer based on the user's emotion." is input, and the AI ​​model generates the text "The weather is sunny today."

[0666] Step 7:

[0667] The server transmits the generated text data to the terminal.

[0668] The server sends the generated answer to the terminal as text data. The input is the generated text data, and the output is the transmission of the text data to the terminal. Specifically, the text data "Today's weather is sunny" is sent to the terminal.

[0669] Step 8:

[0670] The terminal converts the text data into audio data.

[0671] The device uses a speech synthesis engine (e.g., IBM Watson Text-to-Speech) to convert text data into voice data. The input is text data and the output is voice data. For example, the text data "The weather is sunny today" is converted into voice data.

[0672] Step 9:

[0673] The device responds to the user with voice and also provides feedback on the emotional results.

[0674] The device plays the converted voice data through a speaker and also feeds back the emotional result to the user. The input is the voice data and the emotional result, and the output is a voice response to the user and emotional feedback. As a concrete example, the device tells the user, "Today's weather is sunny," in a bright tone.

[0675] Step 10:

[0676] The server periodically updates the generative AI model and emotion recognition engine.

[0677] The server regularly updates the generative AI model and emotion recognition engine to maintain the system's accuracy and response quality. The input is a new dataset and algorithm, and the output is an updated generative AI model and emotion recognition engine. Specifically, the server trains the generative AI model using a new dataset once a month and updates the emotion recognition engine algorithm to the latest version.

[0678] (Application example 2)

[0679] 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."

[0680] Conventional voice dialogue systems simply convert voice data into text and generate responses using a generative AI model, limiting their ability to provide personalized responses that take the user's emotions into account. In brick-and-mortar stores in particular, appropriate responses based on customer emotions are often required, and in this regard, existing technologies are inadequate. Furthermore, delays in information delivery during the voice recognition and response generation process, as well as inaccurate emotion recognition, are factors that reduce the quality of customer service. It is necessary to solve these problems.

[0681] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion engine to recognize the user's emotion, means for inputting the text data and emotion information into a generative AI model to generate a response in an appropriate tone, means for transmitting the generated response as text data, voice synthesis means for converting the text data into voice data, and means for playing back the voice data. This makes it possible to quickly provide an appropriate response according to the user's emotion and improve customer satisfaction.

[0682] "Audio data" refers to data in which audio is recorded in digital format.

[0683] "Text data" is data expressed in the form of a string of characters.

[0684] A "generative AI model" is an artificial intelligence model that analyzes user input data and generates appropriate answers based on it.

[0685] An "emotion engine" is a system that estimates a user's emotions from text and voice data.

[0686] "Appropriate tone" refers to the appropriate speaking style and tone of voice that corresponds to the user's emotions.

[0687] A "voice recognition engine" is a system that analyzes voice data and converts it into text data.

[0688] "Speech synthesis means" is a technology that converts text data into voice data.

[0689] "User" refers to an individual or group that uses the system.

[0690] A "server" is a central computing device that performs various processes for the system.

[0691] A "terminal device" is a device operated by a user that captures and plays audio.

[0692] System Configuration

[0693] The embodiment of the invention employs the following system configuration: The system is mainly composed of three entities: a server, a terminal device, and a user, and each function operates in cooperation with the others.

[0694] Hardware Configuration

[0695] Server: A high-performance computing device that utilizes cloud or on-premise servers.

[0696] Terminal Device: A user interface device such as a smartphone, smart glasses, or tablet.

[0697] Microphone: A device for inputting the user's voice.

[0698] Speaker: A device for playing back generated audio data.

[0699] Software Configuration

[0700] Voice capture module: Installed on the terminal device, it captures the user's voice data.

[0701] Speech recognition engine: Used to convert voice data into text data (e.g., Google Cloud Speech-to-Text).

[0702] Emotion engine: Recognizes user emotions from text data (e.g., IBM Watson Tone Analyzer).

[0703] Generative AI models: Generate appropriate answers based on user input and emotional information (e.g., OpenAI GPT-4).

[0704] Speech synthesis engine: Converts the generated text data into speech data (e.g., Amazon Polly).

[0705] System Operation Overview

[0706] Step 1: Capture audio

[0707] The voice capture module of the terminal device uses a microphone to capture the user's voice data, which is then transmitted to the server in real time.

[0708] Step 2: Voice Recognition

[0709] The server converts the received voice data into text data using a voice recognition engine. For example, if a user says, "I want new shoes," the voice data is converted into text data saying, "I want new shoes."

[0710] Step 3: Emotion Recognition

[0711] The converted text data is then fed into an emotion engine, which analyzes the user's emotions. For example, if the user's comments are recognized as "excited," that emotional information is provided to the generative AI model.

[0712] Step 4: Answer Generation

[0713] The generative AI model generates an appropriate response based on the text data and emotional information. For example, based on the text data "I want new shoes" and the emotional information "I'm excited," the generative AI model generates the following response: "Thank you for visiting! New shoes, then let me show you our latest collection!"

[0714] Step 5: Text-to-Speech

[0715] A speech synthesis engine is used to convert the generated response text data into speech data, which generates a cheerful voice saying, "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!"

[0716] Step 6: Play back the response

[0717] The terminal device plays back the voice data generated by the voice synthesis engine, and the response is sent to the user through the speaker.

[0718] Prompt Sentence Examples

[0719] The customer says, "I want new shoes." The customer is now excited. An appropriate response would be:

[0720] This allows users to receive appropriate responses based on their emotions, improving the customer service experience in physical stores.

[0721] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0722] Step 1:

[0723] The terminal uses the voice capture module to capture the user's voice data through the microphone. The input is the user's voice, and the output is the voice data. This voice data is sent to the server in real time.

[0724] Step 2:

[0725] The voice data received by the server is converted into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text). The input is voice data and the output is text data. For example, voice data saying "I want new shoes" is converted into text data saying "I want new shoes."

[0726] Step 3:

[0727] The converted text data is input into the server's emotion engine (e.g., IBM Watson Tone Analyzer), which analyzes the user's emotions. The input is text data, and the output is emotional information. For example, the emotional information "excited" can be obtained from the text data "I want new shoes."

[0728] Step 4:

[0729] The server inputs text data and emotional information into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates an appropriate answer based on this information. The input is text data and emotional information, and the output is response text data. For example, from the information "I want new shoes" and "Excited," the response text data "Thank you for visiting! New shoes, let me show you our latest collection!" is generated.

[0730] Step 5:

[0731] The generated response text data is input to a speech synthesis engine (e.g., Amazon Polly) and converted into voice data. The input is the response text data, and the output is the response voice data. The speech synthesis engine converts the text data into voice data with a bright tone.

[0732] Step 6:

[0733] The device plays back the response voice data generated by the speech synthesis engine. The input is the response voice data, and the output is a voice response that the user can hear. For example, a voice saying "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!" is played through the device's speaker.

[0734] The above is the specific processing flow of the program of the system that realizes the application example.

[0735] 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.

[0736] 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.

[0737] 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.

[0738] [Third embodiment]

[0739] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0740] 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.

[0741] 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).

[0742] 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.

[0743] 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.

[0744] 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).

[0745] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0746] 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.

[0747] 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.

[0748] 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.

[0749] 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.

[0750] 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."

[0751] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[0752] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This text data is then input into a generative AI model to generate appropriate responses to the user's questions and inquiries. These generated responses are then sent back to the device as text data. The server regularly updates the generative AI model to always provide the latest information and functions.

[0753] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends it to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. Finally, it plays this voice data back to the user through the speaker. This allows the user to receive answers to their questions and inquiries via voice.

[0754] Users can operate the system with simple speech. For example, if they ask, "What's the weather like today?", the device captures the speech and sends it to the server. The server converts this speech into text data and uses a generative AI model to generate an answer based on the weather information. Based on this, the server sends text data such as "The weather is sunny today" back to the device. The device then converts this text data into voice data and tells the user, "The weather is sunny today" over the speaker.

[0755] The system is intended to be installed in the living room for use by the whole family. Its simple and intuitive operation makes it suitable for a wide range of users, from children to the elderly. Regular updates ensure the system always provides the latest information and functions, ensuring long-term use.

[0756] By using this system, users can alleviate the loneliness of living alone and easily obtain information in their daily lives. They can also enjoy new forms of entertainment through dialogue. This system will provide users with new value and experiences, and will greatly enrich their lives in the future.

[0757] The processing flow will be explained below.

[0758] Step 1:

[0759] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[0760] Step 2:

[0761] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[0762] Step 3:

[0763] The device sends the audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[0764] Step 4:

[0765] The server receives the voice data sent from the device and temporarily stores it in the server's storage.

[0766] Step 5:

[0767] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[0768] Step 6:

[0769] The server inputs the text data generated by the speech recognition engine into the generative AI model, which then generates an appropriate answer based on the text data.

[0770] Step 7:

[0771] The server then reformats the text data of the answer returned by the generative AI model and prepares it for transmission to the terminal.

[0772] Step 8:

[0773] The server sends the formatted text data to the device, again via an HTTP response or a WebSocket message.

[0774] Step 9:

[0775] The terminal receives the text data from the server and temporarily stores it.

[0776] Step 10:

[0777] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[0778] Step 11:

[0779] The device plays the generated voice data over a speaker, allowing the user to hear the answer to their question aloud.

[0780] Step 12:

[0781] The user can obtain information by listening to the voice response played back from the terminal. This system allows users to easily obtain information and enjoy dialogue.

[0782] Example 1

[0783] 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."

[0784] Existing speech recognition systems and AI dialogue systems face several challenges in accurately recognizing speech data and generating natural dialogue. In particular, there are issues with misrecognition during the process of converting speech data into text data and the quality of the answers generated by the AI ​​model. Furthermore, the lack of regular model updates makes it difficult to provide appropriate answers based on the latest information. Furthermore, there is a need to efficiently transmit user speech data to a server and provide appropriate answers in real time.

[0785] 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.

[0786] In this invention, the server includes a speech recognition means, a means for inputting text data into a generative AI model to generate an answer, and a means for transmitting the generated answer as text data. This enables accurate conversion of speech data into text and generation of natural dialogue using the generative AI model. Furthermore, the generative AI model generates appropriate answers based on prompts from external input, improving the quality of dialogue with the user. Furthermore, the server periodically updates the generative AI model, allowing it to always provide answers based on the latest information.

[0787] A "means for receiving voice data" is a device or method that captures voice input from a user as a digital signal and receives it in a form that can be processed within the system.

[0788] "Speech recognition means" refers to a device or method that analyzes received voice data and converts it into text format data.

[0789] "Means for inputting text data into a generative AI model to generate an appropriate answer" refers to a device or method for inputting text data into an artificial intelligence model to generate an answer to a user's question or inquiry.

[0790] "Means for transmitting generated answers as text data" refers to a device or method for transmitting answers generated by a generative AI model to another device as text data.

[0791] "Speech synthesis means" refers to a device or method that converts text data into speech data and provides it to the user in a form that can be understood audibly.

[0792] "Means for reproducing audio data" refers to a device or method for reproducing audio data through an output device such as a speaker, and providing information to the user by voice.

[0793] "Means for a terminal device to communicate with a server, transmit voice data to the server, and receive text data from the server" refers to a device or method for communicating to transmit a user's voice data to the server and receive text data from the server.

[0794] "Means for regularly updating a generative AI model" means a device or method for regularly updating a generative AI model based on the latest information and technology, in order to always provide optimal performance and the latest information.

[0795] A "prompt" is an instruction or question input to a generative AI model, and is the input data that enables the model to generate an appropriate answer.

[0796] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[0797] The system of the present invention performs data processing and data calculation using the following hardware and software.

[0798] Hardware and Software Configuration

[0799] 1. Terminal equipment:

[0800] Microphone: A device for capturing audio data from a user.

[0801] Speaker: A device for playing back generated audio data to the user.

[0802] Network interface: A device for communication between a terminal device and a server.

[0803] Speech recognition software: Software that converts voice data into text data. For example, Google Cloud Speech-to-Text API is used.

[0804] 2. Server Device:

[0805] Generative AI model: An artificial intelligence model that inputs text data and generates appropriate answers to user questions or inquiries. For example, OpenAI GPT-4 is used.

[0806] Database: A database for managing update information for generative AI models and speech recognition engines.

[0807] Network interface: A device for communication between a server device and a terminal.

[0808] Specific processing flow

[0809] 1. Capture audio data:

[0810] When a user speaks to the system, "What's the weather like today?", the device's microphone converts the voice signal into digital data, which is then stored in buffer memory.

[0811] 2. Sending audio data:

[0812] The device sends the captured audio data to the server using an HTTP POST request. This audio data is encoded in, for example, WAV format.

[0813] 3. Audio to text conversion:

[0814] The server inputs the received voice data into a voice recognition engine (such as Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, the voice data is converted into text data such as "What's the weather like today?"

[0815] 4. Answer generation using generative AI models:

[0816] The server inputs the generated text data into a generative AI model (OpenAI GPT-4) to generate an appropriate answer. For example, in response to the input "What's the weather like today?", the answer generated is "Today's weather is sunny."

[0817] 5. Submit the generated answer:

[0818] The server sends the generated text data to the terminal via an HTTP POST request.

[0819] 6. Text-to-speech conversion:

[0820] The device converts the received text data into voice data using a speech synthesis engine (such as Amazon Polly), which is then converted into a format that the user can understand.

[0821] 7. Playing audio data:

[0822] The device's speaker will play a voice message saying "Today's weather is sunny" and return the answer to the user.

[0823] Example prompt sentence:

[0824] A user asks about today's weather. Please respond in the following format:

[0825] User asks: "What's the weather like today?"

[0826] answer:

[0827] As described above, this invention allows users to easily obtain information through voice and enjoy natural dialogue. By regularly updating the generative AI model, it is possible to always provide the latest information and improve the quality of the dialogue.

[0828] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0829] Step 1:

[0830] The user speaks to the system, "What's the weather like today?" The device's microphone captures this voice in real time. The voice signal is converted into digital data and stored in buffer memory. The input here is the user's voice, and the output is digital voice data.

[0831] Step 2:

[0832] The device sends the captured audio data to the server in fixed buffer size increments. The audio data (e.g., WAV format) is transferred to the server using an HTTP POST request. The input is the digital audio data, and the output is a transmission request to the server.

[0833] Step 3:

[0834] The server inputs the received voice data into a voice recognition engine (e.g., Google Cloud Speech-to-Text API) and converts it into text data. Specifically, the voice data is analyzed and recognized as text, such as "What's the weather like today?" The input here is digital voice data, and the output is text data.

[0835] Step 4:

[0836] The server inputs the text data obtained from the speech recognition engine into a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate answer to the user's question. The prompt text used is "The user asked about today's weather. Please answer in the following format: User question: 'What's the weather like today?' Answer:". Based on this prompt, the generative AI model generates an answer such as "The weather is sunny today." The input is text data and the prompt text, and the output is the answer text data.

[0837] Step 5:

[0838] The server sends the generated answer text data to the terminal via an HTTP POST request. The input is the generated answer text data, and the output is a request to send to the terminal.

[0839] Step 6:

[0840] The device inputs the received text data into a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. Specifically, it converts the text "Today's weather is sunny" into voice data. The input is text data, and the output is voice data.

[0841] Step 7:

[0842] The terminal speaker plays the converted voice data, and the answer "Today's weather is sunny" is spoken to the user. The input is voice data, and the output is voice playback.

[0843] Through the above steps, the user can have a natural voice interaction with the system.

[0844] (Application example 1)

[0845] 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."

[0846] Conventional food delivery systems have the drawback of being difficult to use, requiring users to input their order details and questions. Furthermore, there is a demand for systems that can respond quickly and appropriately to special requests and questions. Therefore, there is a need for the development of a system that allows users to order food delivery via intuitive voice commands, providing a comfortable user experience.

[0847] 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.

[0848] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for inputting the text data into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis for converting the text data into voice data, means for playing the voice data, means for receiving order details from a user via voice and confirming questions or special requests through natural dialogue, and means for placing a food delivery order based on the voice instructions. This allows users to place food delivery orders and ask questions through intuitive voice operations, improving the user experience.

[0849] "Means for receiving voice data" refers to equipment or software for capturing the voice spoken by the user as digital data.

[0850] The "means for converting voice data into text data" refers to a voice recognition technology that analyzes received voice data and expresses it as text information.

[0851] "Means of inputting into a generative AI model to generate an appropriate response" refers to the process by which the generative AI uses text data obtained through speech recognition to create an appropriate response.

[0852] "Means for transmitting the generated answer as text data" refers to means for transmitting the text-formatted response created by the generation AI to another device or server via communication.

[0853] "Speech synthesis means for converting text data into voice data" is a technology for artificially generating voice using text data.

[0854] "Means for reproducing voice data" refers to a device or software that allows the user to hear the voice data generated by voice synthesis through a speaker.

[0855] "A means of receiving order details via user voice and confirming questions or special requests through natural dialogue" refers to technology that interactively confirms the order details or questions communicated by the user and obtains additional information as needed.

[0856] "Means for ordering food delivery based on voice instructions" refers to technology that automates the ordering process by sending order information to a food delivery service based on the user's voice instructions.

[0857] This invention provides a system that allows users to intuitively order food delivery and ask questions through voice. To implement the invention, a system configuration using mainly a server and a terminal is required.

[0858] Terminal configuration and operation

[0859] The terminal is equipped with a microphone for capturing the user's voice data, a speaker for playing back the synthesized voice data, and a network interface for communicating with the server. Specifically, a mobile terminal such as a smartphone is assumed. The application installed on the terminal has the following functions:

[0860] 1. Audio capture function:

[0861] Use the device's microphone to receive user voice input.

[0862] 2. Voice recognition function:

[0863] The received voice data is converted into text data by a voice recognition engine.

[0864] Software used: Google Speech Recognition API

[0865] 3. Communication functions:

[0866] The converted text data is sent to a server and a response is received from the server.

[0867] Software used: Communication library using HTTP / HTTPS protocol

[0868] Server configuration and operation

[0869] The server analyzes the received text data and generates an appropriate answer using a generative AI model. The generated answer is sent to the device as text data. The server has the following functions implemented:

[0870] 1. Data receiving function:

[0871] Receive text data sent from the terminal.

[0872] 2. Generative AI Model:

[0873] The received text data is used as input to generate an appropriate answer.

[0874] Software used: OpenAI GPT-3

[0875] 3. Data transmission function:

[0876] The generated answer is sent to the terminal as text data.

[0877] Speech synthesis and output

[0878] The device converts the text data received from the server into voice data using a speech synthesis engine and plays a voice response to the user. The software used is Google Text-to-Speech (gTTS).

[0879] Specific examples

[0880] For example, if a user says, "I'd like to order pizza. What's your recommendation?", the process will proceed as follows:

[0881] 1. The device captures the audio data.

[0882] 2. The voice data is converted into text data by a voice recognition engine.

[0883] 3. The text data is sent to the server.

[0884] 4. The server generates an appropriate answer using the generative AI model. Example prompt:

[0885] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[0886] Please write an appropriate answer in Japanese.

[0887] 5. The server generates a response and sends it to the terminal as text data.

[0888] 6. The text data received by the device is converted into voice data using a voice synthesis engine.

[0889] 7. Play the answer back to the user through the speaker.

[0890] In this way, users can smoothly place food delivery orders and ask questions through voice interaction, improving convenience.

[0891] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0892] Step 1:

[0893] Users can order food delivery or ask questions by speaking into the microphone on their smartphone.

[0894] Input: User's voice data

[0895] Output: Audio data captured by the smartphone microphone

[0896] What happens: The user says, "I'd like to order a pizza. What do you recommend?"

[0897] Step 2:

[0898] The device sends the captured voice data to a voice recognition engine, which converts it into text data.

[0899] Input: Audio data

[0900] Output: Text data as a string

[0901] Specific operation: The voice recognition software (Google Speech Recognition API) on the smartphone analyzes the voice data and generates the corresponding text data: "I'd like to order a pizza. What do you recommend?"

[0902] Step 3:

[0903] The terminal transmits the text data to the server.

[0904] Input: Text data

[0905] Output: Text data sent to the server

[0906] Specific operation: The smartphone sends the generated text data to the server using the HTTP / HTTPS protocol.

[0907] Step 4:

[0908] The server inputs the received text data into a generative AI model to generate an appropriate answer.

[0909] Input: Text data

[0910] Output: Text data of generated answers

[0911] Specific operation: The generative AI model (OpenAI GPT-3) on the server generates an appropriate answer based on the following prompt.

[0912] Example prompt sentence:

[0913] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[0914] Please write an appropriate answer in Japanese.

[0915] Sample generated answer: "Recommended pizzas are the Margherita and seafood pizza."

[0916] Step 5:

[0917] The server sends the generated response to the terminal as text data.

[0918] Input: Text data of generated answers

[0919] Output: Text data sent to the terminal

[0920] Specific operation: The server uses the HTTP / HTTPS protocol to send the generated response to the terminal as text data.

[0921] Step 6:

[0922] The terminal inputs the received text data into a speech synthesis engine and converts it into voice data.

[0923] Input: Text data of generated answers

[0924] Output: Audio data

[0925] Specific operation: The speech synthesis software (Google Text-to-Speech, gTTS) on the smartphone converts the text data "Recommended pizza menu items are Margherita and seafood pizza" into speech data.

[0926] Step 7:

[0927] The device plays the converted audio data to the user through the speaker.

[0928] Input: Audio data

[0929] Output: The audio the user hears

[0930] Specific operation: The smartphone speaker plays audio data, telling the user, "Our recommended pizza menu items are Margherita and seafood pizza."

[0931] 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.

[0932] This invention is a system that receives voice data and uses a speech recognition engine and a generative AI model to have a natural dialogue with the user. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides appropriate responses according to the user's emotions.

[0933] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This converted text data is also input into an emotion engine to recognize the user's emotions. Once the emotion engine identifies the user's emotions (e.g., joy, sadness, anger, etc.), it provides that information to a generative AI model. The generative AI model generates an appropriate response based on the user's emotions based on the text data and emotion information. This generated response is formatted as text data and sent to the device. The server also regularly updates the generative AI model and emotion engine to provide the latest information and functions.

[0934] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends the acquired voice data to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. The device plays the converted voice data through a speaker and also provides feedback to the user on the emotions recognized by the emotion engine. This allows the user to confirm that their emotions are being understood by the system.

[0935] Users can operate the system with simple speech. For example, if a user asks, "What's the weather like today?", the device captures the voice and sends it to the server. The server converts the voice data into text and uses an emotion engine to analyze the user's emotions. Based on the analysis results, the generative AI model generates an appropriate answer and returns text data such as "The weather is sunny today." The device then converts this text data into speech and tells the user, along with the emotion result, "The weather is sunny today" in a bright tone.

[0936] This system is installed in the living room and is designed to blend into the user's daily life. Its simple operation makes it easy to use for a wide range of users, from children to the elderly. Regular updates ensure that the latest information and functions are always provided, ensuring long-term usability. Using this system can alleviate the loneliness of living alone and allow users to easily access information in their daily lives. Furthermore, emotion-based responses allow users to enjoy a richer interactive experience. In this way, the present invention provides users with new value and experiences, supporting a richer lifestyle.

[0937] The processing flow will be explained below.

[0938] Step 1:

[0939] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[0940] Step 2:

[0941] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[0942] Step 3:

[0943] The device sends the captured audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[0944] Step 4:

[0945] The server receives the voice data sent from the terminal and temporarily stores it in the server's storage.

[0946] Step 5:

[0947] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[0948] Step 6:

[0949] The server sends the converted text data to the emotion engine, which analyzes the text data and recognizes the user's emotions (e.g., joy, sadness, anger, etc.).

[0950] Step 7:

[0951] The server inputs the emotional information and text data obtained from the emotion engine into the generative AI model, which then generates an appropriate answer based on the user's emotions and text data.

[0952] Step 8:

[0953] The server formats the answer obtained from the generative AI model as text data and prepares it to be sent to the device.

[0954] Step 9:

[0955] The server sends the formatted text data and emotion information to the device, again via HTTP responses or WebSocket messages.

[0956] Step 10:

[0957] The device receives the text data and emotion information from the server and temporarily stores them.

[0958] Step 11:

[0959] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[0960] Step 12:

[0961] The device then plays the generated voice data over a speaker, allowing the user to hear the answer to their question. The tone and intonation of the voice may be adjusted based on the emotional information.

[0962] Step 13:

[0963] The user listens to the voice response played back from the device. The emotional feedback confirms that their feelings are understood. This system allows the user to feel emotional satisfaction through the interaction.

[0964] Example 2

[0965] 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."

[0966] Conventional voice recognition systems do not take user emotions into account, making dialogue mechanical and difficult to achieve natural communication. Furthermore, the lack of regular updates to generative AI models and emotion recognition engines has led to problems with the system's accuracy and response quality declining. Furthermore, users are unable to confirm whether their emotions are being understood, which leads to a decline in dialogue satisfaction.

[0967] 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.

[0968] In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion recognition engine to analyze the user's emotions, means for inputting the text data and emotion information into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis to convert the text data into voice data, means for playing the voice data, and means for feeding back the emotion recognition results to the user. This enables natural dialogue according to the user's emotions, and regular updates can maintain the accuracy and quality of the system's responses, improving user satisfaction.

[0969] "Voice Data" means a digital representation of voice information uttered by a User.

[0970] "Text data" is digital information that is audio data converted into text format.

[0971] An "emotion recognition engine" is a software module that analyzes text data to identify a user's emotions.

[0972] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers based on input text data and emotional information.

[0973] A "voice recognition engine" is a software module for converting voice data into text data.

[0974] A "speech synthesis means" is a software module and hardware device for converting text data into speech data.

[0975] A "terminal device" is a hardware device that captures voice data from a user and transmits it to a server.

[0976] The "server" is a central processing unit that processes voice data and performs emotion recognition and response generation.

[0977] A "prompt" is an instruction or question to be input into a generative AI model.

[0978] This invention is a system that receives voice data and uses a generative AI model and an emotion recognition engine to engage in natural dialogue with the user. It also recognizes the user's emotions and provides appropriate responses based on those emotions.

[0979] Specifically, the following hardware and software are used.

[0980] Hardware:

[0981] 1. Terminal equipment:

[0982] Microphone: Captures the user's voice.

[0983] Speaker: Provides a spoken response to the user.

[0984] Communication module: Performs data communication with the server.

[0985] 2. Server:

[0986] Central processing unit: processes voice data, recognizes emotions, and generates responses.

[0987] software:

[0988] 1. Speech recognition engine (e.g. Google Speech-to-Text API)

[0989] Convert the audio data into text data.

[0990] 2. Emotion recognition engine (e.g., Microsoft Azure Text Analytics)

[0991] Analyze text data to identify user sentiment.

[0992] 3. Generative AI models (e.g., OpenAI GPT-4)

[0993] Generate appropriate answers based on text data and emotional information.

[0994] 4. Speech synthesis engine (e.g. IBM Watson Text-to-Speech)

[0995] Convert text data into audio data.

[0996] Detailed instructions on how to do this:

[0997] The server uses a voice recognition engine to convert the voice data sent from the terminal device into text data. The converted text data is analyzed by an emotion recognition engine to identify the user's emotion. This text data and emotion information are then input into a generative AI model to generate an appropriate answer. This answer is formatted as text data and sent from the server to the terminal device. The terminal device then converts the received text data into voice data using a voice synthesis engine and responds to the user audibly using a speaker.

[0998] Examples:

[0999] When a user asks, "What's the weather today?", the terminal device captures the voice and sends it to the server. The server uses a speech recognition engine to convert this voice data into text data, generating the text "What's the weather today?". It then uses an emotion recognition engine to identify the user's emotion (e.g., "interested") from this text and provides that emotion information to the generative AI model. The generative AI model then generates an appropriate answer based on the following prompt sentence, for example:

[1000] Example prompt sentence:

[1001] "A user asks, 'What's the weather like today?' The user's emotion is 'interested.' Generate an appropriate answer based on the user's emotion."

[1002] The generative AI model generates a response such as "Today's weather is sunny," which the server sends to the terminal device as text data. The terminal device then uses a speech synthesis engine to convert the text data into voice data, responding to the user in a bright tone: "Today's weather is sunny."

[1003] This system allows users to enjoy a natural, emotion-sensitive conversational experience, helping them to obtain information on a daily basis and alleviating the loneliness of living alone. Furthermore, feedback on emotion recognition results allows users to confirm that their emotions are being understood by the system.

[1004] Furthermore, the server regularly updates the generative AI model and emotion recognition engine, allowing it to continually provide the latest information and advanced functionality, thereby increasing the long-term utility value of the system and improving user satisfaction.

[1005] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1006] Step 1:

[1007] The user provides voice input.

[1008] Specifically, the user speaks a question such as "What's the weather like today?" This speech is input into the system. The input data is the user's speech, and the output is voice data.

[1009] Step 2:

[1010] The device captures the audio data.

[1011] The device's microphone captures the user's voice and records it as digital voice data. During this process, voice data is input and the voice data is output as is. For example, the voice saying "What's the weather like today?" is captured in digital form.

[1012] Step 3:

[1013] The terminal transmits the voice data to the server.

[1014] The communication module of the device sends the captured audio data to the server. The input here is the audio data of the device, and the output is the transmission of audio data to the server. Specifically, the device sends the audio data to the server using an HTTP request.

[1015] Step 4:

[1016] The server converts the voice data into text data.

[1017] The server converts the received voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). The input of this step is voice data, and the output is text data. For example, the voice data "What's the weather like today?" is converted into the text data "What's the weather like today?"

[1018] Step 5:

[1019] The server analyzes emotions using an emotion recognition engine.

[1020] The server inputs the text data into an emotion recognition engine (e.g., Microsoft Azure Text Analytics) to analyze the user's emotions. The input is text data, and the output is emotion information. Specifically, the emotion "interest" is identified from the text data "What's the weather like today?"

[1021] Step 6:

[1022] The server inputs a prompt into the generative AI model and generates an appropriate answer.

[1023] The server inputs a prompt based on text data and emotional information into a generative AI model (e.g., OpenAI GPT-4) and generates an appropriate answer. The input for this step is text data and emotional information, and the output is the generated answer text data. As a specific example, the prompt "The user asked, 'What's the weather like today?' The user's emotion at that time is 'interested.' Please generate an appropriate answer based on the user's emotion." is input, and the AI ​​model generates the text "The weather is sunny today."

[1024] Step 7:

[1025] The server transmits the generated text data to the terminal.

[1026] The server sends the generated answer to the terminal as text data. The input is the generated text data, and the output is the transmission of the text data to the terminal. Specifically, the text data "Today's weather is sunny" is sent to the terminal.

[1027] Step 8:

[1028] The terminal converts the text data into audio data.

[1029] The device uses a speech synthesis engine (e.g., IBM Watson Text-to-Speech) to convert text data into voice data. The input is text data and the output is voice data. For example, the text data "The weather is sunny today" is converted into voice data.

[1030] Step 9:

[1031] The device responds to the user with voice and also provides feedback on the emotional results.

[1032] The device plays the converted voice data through a speaker and also feeds back the emotional result to the user. The input is the voice data and the emotional result, and the output is a voice response to the user and emotional feedback. As a concrete example, the device tells the user, "Today's weather is sunny," in a bright tone.

[1033] Step 10:

[1034] The server periodically updates the generative AI model and emotion recognition engine.

[1035] The server regularly updates the generative AI model and emotion recognition engine to maintain the system's accuracy and response quality. The input is a new dataset and algorithm, and the output is an updated generative AI model and emotion recognition engine. Specifically, the server trains the generative AI model using a new dataset once a month and updates the emotion recognition engine algorithm to the latest version.

[1036] (Application example 2)

[1037] 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."

[1038] Conventional voice dialogue systems simply convert voice data into text and generate responses using a generative AI model, limiting their ability to provide personalized responses that take the user's emotions into account. In brick-and-mortar stores in particular, appropriate responses based on customer emotions are often required, and in this regard, existing technologies are inadequate. Furthermore, delays in information delivery during the voice recognition and response generation process, as well as inaccurate emotion recognition, are factors that reduce the quality of customer service. It is necessary to solve these problems.

[1039] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion engine to recognize the user's emotion, means for inputting the text data and emotion information into a generative AI model to generate a response in an appropriate tone, means for transmitting the generated response as text data, voice synthesis means for converting the text data into voice data, and means for playing back the voice data. This makes it possible to quickly provide an appropriate response according to the user's emotion and improve customer satisfaction.

[1040] "Audio data" refers to data in which audio is recorded in digital format.

[1041] "Text data" is data expressed in the form of a string of characters.

[1042] A "generative AI model" is an artificial intelligence model that analyzes user input data and generates appropriate answers based on it.

[1043] An "emotion engine" is a system that estimates a user's emotions from text and voice data.

[1044] "Appropriate tone" refers to the appropriate speaking style and tone of voice that corresponds to the user's emotions.

[1045] A "voice recognition engine" is a system that analyzes voice data and converts it into text data.

[1046] "Speech synthesis means" is a technology that converts text data into voice data.

[1047] "User" refers to an individual or group that uses the system.

[1048] A "server" is a central computing device that performs various processes for the system.

[1049] A "terminal device" is a device operated by a user that captures and plays audio.

[1050] System Configuration

[1051] The embodiment of the invention employs the following system configuration: The system is mainly composed of three entities: a server, a terminal device, and a user, and each function operates in cooperation with the others.

[1052] Hardware Configuration

[1053] Server: A high-performance computing device that utilizes cloud or on-premise servers.

[1054] Terminal Device: A user interface device such as a smartphone, smart glasses, or tablet.

[1055] Microphone: A device for inputting the user's voice.

[1056] Speaker: A device for playing back generated audio data.

[1057] Software Configuration

[1058] Voice capture module: Installed on the terminal device, it captures the user's voice data.

[1059] Speech recognition engine: Used to convert voice data into text data (e.g., Google Cloud Speech-to-Text).

[1060] Emotion engine: Recognizes user emotions from text data (e.g., IBM Watson Tone Analyzer).

[1061] Generative AI models: Generate appropriate answers based on user input and emotional information (e.g., OpenAI GPT-4).

[1062] Speech synthesis engine: Converts the generated text data into speech data (e.g., Amazon Polly).

[1063] System Operation Overview

[1064] Step 1: Capture audio

[1065] The voice capture module of the terminal device uses a microphone to capture the user's voice data, which is then transmitted to the server in real time.

[1066] Step 2: Voice Recognition

[1067] The server converts the received voice data into text data using a voice recognition engine. For example, if a user says, "I want new shoes," the voice data is converted into text data saying, "I want new shoes."

[1068] Step 3: Emotion Recognition

[1069] The converted text data is then fed into an emotion engine, which analyzes the user's emotions. For example, if the user's comments are recognized as "excited," that emotional information is provided to the generative AI model.

[1070] Step 4: Answer Generation

[1071] The generative AI model generates an appropriate response based on the text data and emotional information. For example, based on the text data "I want new shoes" and the emotional information "I'm excited," the generative AI model generates the following response: "Thank you for visiting! New shoes, then let me show you our latest collection!"

[1072] Step 5: Text-to-Speech

[1073] A speech synthesis engine is used to convert the generated response text data into speech data, which generates a cheerful voice saying, "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!"

[1074] Step 6: Play back the response

[1075] The terminal device plays back the voice data generated by the voice synthesis engine, and the response is sent to the user through the speaker.

[1076] Prompt Sentence Examples

[1077] The customer says, "I want new shoes." The customer is now excited. An appropriate response would be:

[1078] This allows users to receive appropriate responses based on their emotions, improving the customer service experience in physical stores.

[1079] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1080] Step 1:

[1081] The terminal uses the voice capture module to capture the user's voice data through the microphone. The input is the user's voice, and the output is the voice data. This voice data is sent to the server in real time.

[1082] Step 2:

[1083] The voice data received by the server is converted into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text). The input is voice data and the output is text data. For example, voice data saying "I want new shoes" is converted into text data saying "I want new shoes."

[1084] Step 3:

[1085] The converted text data is input into the server's emotion engine (e.g., IBM Watson Tone Analyzer), which analyzes the user's emotions. The input is text data, and the output is emotional information. For example, the emotional information "excited" can be obtained from the text data "I want new shoes."

[1086] Step 4:

[1087] The server inputs text data and emotional information into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates an appropriate answer based on this information. The input is text data and emotional information, and the output is response text data. For example, from the information "I want new shoes" and "Excited," the response text data "Thank you for visiting! New shoes, let me show you our latest collection!" is generated.

[1088] Step 5:

[1089] The generated response text data is input to a speech synthesis engine (e.g., Amazon Polly) and converted into voice data. The input is the response text data, and the output is the response voice data. The speech synthesis engine converts the text data into voice data with a bright tone.

[1090] Step 6:

[1091] The device plays back the response voice data generated by the speech synthesis engine. The input is the response voice data, and the output is a voice response that the user can hear. For example, a voice saying "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!" is played through the device's speaker.

[1092] The above is the specific processing flow of the program of the system that realizes the application example.

[1093] 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.

[1094] 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.

[1095] 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.

[1096] [Fourth embodiment]

[1097] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1098] 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.

[1099] 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).

[1100] 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.

[1101] 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.

[1102] 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).

[1103] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1104] 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.

[1105] 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.

[1106] 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.

[1107] 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.

[1108] 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.

[1109] 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."

[1110] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[1111] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This text data is then input into a generative AI model to generate appropriate responses to the user's questions and inquiries. These generated responses are then sent back to the device as text data. The server regularly updates the generative AI model to always provide the latest information and functions.

[1112] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends it to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. Finally, it plays this voice data back to the user through the speaker. This allows the user to receive answers to their questions and inquiries via voice.

[1113] Users can operate the system with simple speech. For example, if they ask, "What's the weather like today?", the device captures the speech and sends it to the server. The server converts this speech into text data and uses a generative AI model to generate an answer based on the weather information. Based on this, the server sends text data such as "The weather is sunny today" back to the device. The device then converts this text data into voice data and tells the user, "The weather is sunny today" over the speaker.

[1114] The system is intended to be installed in the living room for use by the whole family. Its simple and intuitive operation makes it suitable for a wide range of users, from children to the elderly. Regular updates ensure the system always provides the latest information and functions, ensuring long-term use.

[1115] By using this system, users can alleviate the loneliness of living alone and easily obtain information in their daily lives. They can also enjoy new forms of entertainment through dialogue. This system will provide users with new value and experiences, and will greatly enrich their lives in the future.

[1116] The processing flow will be explained below.

[1117] Step 1:

[1118] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[1119] Step 2:

[1120] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[1121] Step 3:

[1122] The device sends the audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[1123] Step 4:

[1124] The server receives the voice data sent from the device and temporarily stores it in the server's storage.

[1125] Step 5:

[1126] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[1127] Step 6:

[1128] The server inputs the text data generated by the speech recognition engine into the generative AI model, which then generates an appropriate answer based on the text data.

[1129] Step 7:

[1130] The server then reformats the text data of the answer returned by the generative AI model and prepares it for transmission to the terminal.

[1131] Step 8:

[1132] The server sends the formatted text data to the device, again via an HTTP response or a WebSocket message.

[1133] Step 9:

[1134] The terminal receives the text data from the server and temporarily stores it.

[1135] Step 10:

[1136] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[1137] Step 11:

[1138] The device plays the generated voice data over a speaker, allowing the user to hear the answer to their question aloud.

[1139] Step 12:

[1140] The user can obtain information by listening to the voice response played back from the terminal. This system allows users to easily obtain information and enjoy dialogue.

[1141] Example 1

[1142] 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."

[1143] Existing speech recognition systems and AI dialogue systems face several challenges in accurately recognizing speech data and generating natural dialogue. In particular, there are issues with misrecognition during the process of converting speech data into text data and the quality of the answers generated by the AI ​​model. Furthermore, the lack of regular model updates makes it difficult to provide appropriate answers based on the latest information. Furthermore, there is a need to efficiently transmit user speech data to a server and provide appropriate answers in real time.

[1144] 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.

[1145] In this invention, the server includes a speech recognition means, a means for inputting text data into a generative AI model to generate an answer, and a means for transmitting the generated answer as text data. This enables accurate conversion of speech data into text and generation of natural dialogue using the generative AI model. Furthermore, the generative AI model generates appropriate answers based on prompts from external input, improving the quality of dialogue with the user. Furthermore, the server periodically updates the generative AI model, allowing it to always provide answers based on the latest information.

[1146] A "means for receiving voice data" is a device or method that captures voice input from a user as a digital signal and receives it in a form that can be processed within the system.

[1147] "Speech recognition means" refers to a device or method that analyzes received voice data and converts it into text format data.

[1148] "Means for inputting text data into a generative AI model to generate an appropriate answer" refers to a device or method for inputting text data into an artificial intelligence model to generate an answer to a user's question or inquiry.

[1149] "Means for transmitting generated answers as text data" refers to a device or method for transmitting answers generated by a generative AI model to another device as text data.

[1150] "Speech synthesis means" refers to a device or method that converts text data into speech data and provides it to the user in a form that can be understood audibly.

[1151] "Means for reproducing audio data" refers to a device or method for reproducing audio data through an output device such as a speaker, and providing information to the user by voice.

[1152] "Means for a terminal device to communicate with a server, transmit voice data to the server, and receive text data from the server" refers to a device or method for communicating to transmit a user's voice data to the server and receive text data from the server.

[1153] "Means for regularly updating a generative AI model" means a device or method for regularly updating a generative AI model based on the latest information and technology, in order to always provide optimal performance and the latest information.

[1154] A "prompt" is an instruction or question input to a generative AI model, and is the input data that enables the model to generate an appropriate answer.

[1155] The present invention is a system for receiving voice data and engaging in natural dialogue with a user using a speech recognition engine and a generative AI model. The system of the present invention is composed of a server and a terminal, and is designed to be easy for users to operate.

[1156] The system of the present invention performs data processing and data calculation using the following hardware and software.

[1157] Hardware and Software Configuration

[1158] 1. Terminal equipment:

[1159] Microphone: A device for capturing audio data from a user.

[1160] Speaker: A device for playing back generated audio data to the user.

[1161] Network interface: A device for communication between a terminal device and a server.

[1162] Speech recognition software: Software that converts voice data into text data. For example, Google Cloud Speech-to-Text API is used.

[1163] 2. Server Device:

[1164] Generative AI model: An artificial intelligence model that inputs text data and generates appropriate answers to user questions or inquiries. For example, OpenAI GPT-4 is used.

[1165] Database: A database for managing update information for generative AI models and speech recognition engines.

[1166] Network interface: A device for communication between a server device and a terminal.

[1167] Specific processing flow

[1168] 1. Capture audio data:

[1169] When a user speaks to the system, "What's the weather like today?", the device's microphone converts the voice signal into digital data, which is then stored in buffer memory.

[1170] 2. Sending audio data:

[1171] The device sends the captured audio data to the server using an HTTP POST request. This audio data is encoded in, for example, WAV format.

[1172] 3. Audio to text conversion:

[1173] The server inputs the received voice data into a voice recognition engine (such as Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, the voice data is converted into text data such as "What's the weather like today?"

[1174] 4. Answer generation using generative AI models:

[1175] The server inputs the generated text data into a generative AI model (OpenAI GPT-4) to generate an appropriate answer. For example, in response to the input "What's the weather like today?", the answer generated is "Today's weather is sunny."

[1176] 5. Submit the generated answer:

[1177] The server sends the generated text data to the terminal via an HTTP POST request.

[1178] 6. Text-to-speech conversion:

[1179] The device converts the received text data into voice data using a speech synthesis engine (such as Amazon Polly), which is then converted into a format that the user can understand.

[1180] 7. Playing audio data:

[1181] The device's speaker will play a voice message saying "Today's weather is sunny" and return the answer to the user.

[1182] Example prompt sentence:

[1183] A user asks about today's weather. Please respond in the following format:

[1184] User asks: "What's the weather like today?"

[1185] answer:

[1186] As described above, this invention allows users to easily obtain information through voice and enjoy natural dialogue. By regularly updating the generative AI model, it is possible to always provide the latest information and improve the quality of the dialogue.

[1187] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1188] Step 1:

[1189] The user speaks to the system, "What's the weather like today?" The device's microphone captures this voice in real time. The voice signal is converted into digital data and stored in buffer memory. The input here is the user's voice, and the output is digital voice data.

[1190] Step 2:

[1191] The device sends the captured audio data to the server in fixed buffer size increments. The audio data (e.g., WAV format) is transferred to the server using an HTTP POST request. The input is the digital audio data, and the output is a transmission request to the server.

[1192] Step 3:

[1193] The server inputs the received voice data into a voice recognition engine (e.g., Google Cloud Speech-to-Text API) and converts it into text data. Specifically, the voice data is analyzed and recognized as text, such as "What's the weather like today?" The input here is digital voice data, and the output is text data.

[1194] Step 4:

[1195] The server inputs the text data obtained from the speech recognition engine into a generative AI model (e.g., OpenAI GPT-4) to generate an appropriate answer to the user's question. The prompt text used is "The user asked about today's weather. Please answer in the following format: User question: 'What's the weather like today?' Answer:". Based on this prompt, the generative AI model generates an answer such as "The weather is sunny today." The input is text data and the prompt text, and the output is the answer text data.

[1196] Step 5:

[1197] The server sends the generated answer text data to the terminal via an HTTP POST request. The input is the generated answer text data, and the output is a request to send to the terminal.

[1198] Step 6:

[1199] The device inputs the received text data into a speech synthesis engine (e.g., Amazon Polly) and converts it into voice data. Specifically, it converts the text "Today's weather is sunny" into voice data. The input is text data, and the output is voice data.

[1200] Step 7:

[1201] The terminal speaker plays the converted voice data, and the answer "Today's weather is sunny" is spoken to the user. The input is voice data, and the output is voice playback.

[1202] Through the above steps, the user can have a natural voice interaction with the system.

[1203] (Application example 1)

[1204] 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."

[1205] Conventional food delivery systems have the drawback of being difficult to use, requiring users to input their order details and questions. Furthermore, there is a demand for systems that can respond quickly and appropriately to special requests and questions. Therefore, there is a need for the development of a system that allows users to order food delivery via intuitive voice commands, providing a comfortable user experience.

[1206] 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.

[1207] In this invention, the server includes means for receiving voice data, means for converting the voice data into text data, means for inputting the text data into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis for converting the text data into voice data, means for playing the voice data, means for receiving order details from a user via voice and confirming questions or special requests through natural dialogue, and means for placing a food delivery order based on the voice instructions. This allows users to place food delivery orders and ask questions through intuitive voice operations, improving the user experience.

[1208] "Means for receiving voice data" refers to equipment or software for capturing the voice spoken by the user as digital data.

[1209] The "means for converting voice data into text data" refers to a voice recognition technology that analyzes received voice data and expresses it as text information.

[1210] "Means of inputting into a generative AI model to generate an appropriate response" refers to the process by which the generative AI uses text data obtained through speech recognition to create an appropriate response.

[1211] "Means for transmitting the generated answer as text data" refers to means for transmitting the text-formatted response created by the generation AI to another device or server via communication.

[1212] "Speech synthesis means for converting text data into voice data" is a technology for artificially generating voice using text data.

[1213] "Means for reproducing voice data" refers to a device or software that allows the user to hear the voice data generated by voice synthesis through a speaker.

[1214] "A means of receiving order details via user voice and confirming questions or special requests through natural dialogue" refers to technology that interactively confirms the order details or questions communicated by the user and obtains additional information as needed.

[1215] "Means for ordering food delivery based on voice instructions" refers to technology that automates the ordering process by sending order information to a food delivery service based on the user's voice instructions.

[1216] This invention provides a system that allows users to intuitively order food delivery and ask questions through voice. To implement the invention, a system configuration using mainly a server and a terminal is required.

[1217] Terminal configuration and operation

[1218] The terminal is equipped with a microphone for capturing the user's voice data, a speaker for playing back the synthesized voice data, and a network interface for communicating with the server. Specifically, a mobile terminal such as a smartphone is assumed. The application installed on the terminal has the following functions:

[1219] 1. Audio capture function:

[1220] Use the device's microphone to receive user voice input.

[1221] 2. Voice recognition function:

[1222] The received voice data is converted into text data by a voice recognition engine.

[1223] Software used: Google Speech Recognition API

[1224] 3. Communication functions:

[1225] The converted text data is sent to a server and a response is received from the server.

[1226] Software used: Communication library using HTTP / HTTPS protocol

[1227] Server configuration and operation

[1228] The server analyzes the received text data and generates an appropriate answer using a generative AI model. The generated answer is sent to the device as text data. The server has the following functions implemented:

[1229] 1. Data receiving function:

[1230] Receive text data sent from the terminal.

[1231] 2. Generative AI Model:

[1232] The received text data is used as input to generate an appropriate answer.

[1233] Software used: OpenAI GPT-3

[1234] 3. Data transmission function:

[1235] The generated answer is sent to the terminal as text data.

[1236] Speech synthesis and output

[1237] The device converts the text data received from the server into voice data using a speech synthesis engine and plays a voice response to the user. The software used is Google Text-to-Speech (gTTS).

[1238] Specific examples

[1239] For example, if a user says, "I'd like to order pizza. What's your recommendation?", the process will proceed as follows:

[1240] 1. The device captures the audio data.

[1241] 2. The voice data is converted into text data by a voice recognition engine.

[1242] 3. The text data is sent to the server.

[1243] 4. The server generates an appropriate answer using the generative AI model. Example prompt:

[1244] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[1245] Please write an appropriate answer in Japanese.

[1246] 5. The server generates a response and sends it to the terminal as text data.

[1247] 6. The text data received by the device is converted into voice data using a voice synthesis engine.

[1248] 7. Play the answer back to the user through the speaker.

[1249] In this way, users can smoothly place food delivery orders and ask questions through voice interaction, improving convenience.

[1250] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1251] Step 1:

[1252] Users can order food delivery or ask questions by speaking into the microphone on their smartphone.

[1253] Input: User's voice data

[1254] Output: Audio data captured by the smartphone microphone

[1255] What happens: The user says, "I'd like to order a pizza. What do you recommend?"

[1256] Step 2:

[1257] The device sends the captured voice data to a voice recognition engine, which converts it into text data.

[1258] Input: Audio data

[1259] Output: Text data as a string

[1260] Specific operation: The voice recognition software (Google Speech Recognition API) on the smartphone analyzes the voice data and generates the corresponding text data: "I'd like to order a pizza. What do you recommend?"

[1261] Step 3:

[1262] The terminal transmits the text data to the server.

[1263] Input: Text data

[1264] Output: Text data sent to the server

[1265] Specific operation: The smartphone sends the generated text data to the server using the HTTP / HTTPS protocol.

[1266] Step 4:

[1267] The server inputs the received text data into a generative AI model to generate an appropriate answer.

[1268] Input: Text data

[1269] Output: Text data of generated answers

[1270] Specific operation: The generative AI model (OpenAI GPT-3) on the server generates an appropriate answer based on the following prompt.

[1271] Example prompt sentence:

[1272] A user provided the following order: "I'd like to order a pizza. What do you recommend?"

[1273] Please write an appropriate answer in Japanese.

[1274] Sample generated answer: "Recommended pizzas are the Margherita and seafood pizza."

[1275] Step 5:

[1276] The server sends the generated response to the terminal as text data.

[1277] Input: Text data of generated answers

[1278] Output: Text data sent to the terminal

[1279] Specific operation: The server uses the HTTP / HTTPS protocol to send the generated response to the terminal as text data.

[1280] Step 6:

[1281] The terminal inputs the received text data into a speech synthesis engine and converts it into voice data.

[1282] Input: Text data of generated answers

[1283] Output: Audio data

[1284] Specific operation: The speech synthesis software (Google Text-to-Speech, gTTS) on the smartphone converts the text data "Recommended pizza menu items are Margherita and seafood pizza" into speech data.

[1285] Step 7:

[1286] The device plays the converted audio data to the user through the speaker.

[1287] Input: Audio data

[1288] Output: The audio the user hears

[1289] Specific operation: The smartphone speaker plays audio data, telling the user, "Our recommended pizza menu items are Margherita and seafood pizza."

[1290] 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.

[1291] This invention is a system that receives voice data and uses a speech recognition engine and a generative AI model to have a natural dialogue with the user. In addition, by combining it with an emotion engine that recognizes the user's emotions, it provides appropriate responses according to the user's emotions.

[1292] The server's primary role is to process voice data and generate appropriate responses. The server receives voice data from the device and converts it into text data using a speech recognition engine. This converted text data is also input into an emotion engine to recognize the user's emotions. Once the emotion engine identifies the user's emotions (e.g., joy, sadness, anger, etc.), it provides that information to a generative AI model. The generative AI model generates an appropriate response based on the user's emotions based on the text data and emotion information. This generated response is formatted as text data and sent to the device. The server also regularly updates the generative AI model and emotion engine to provide the latest information and functions.

[1293] The main role of the device is to capture voice data from the user and communicate with the server. The device uses a microphone to capture the user's voice input and sends the acquired voice data to the server. It also receives text data returned from the server and converts it back into voice data using a speech synthesis engine. The device plays the converted voice data through a speaker and also provides feedback to the user on the emotions recognized by the emotion engine. This allows the user to confirm that their emotions are being understood by the system.

[1294] Users can operate the system with simple speech. For example, if a user asks, "What's the weather like today?", the device captures the voice and sends it to the server. The server converts the voice data into text and uses an emotion engine to analyze the user's emotions. Based on the analysis results, the generative AI model generates an appropriate answer and returns text data such as "The weather is sunny today." The device then converts this text data into speech and tells the user, along with the emotion result, "The weather is sunny today" in a bright tone.

[1295] This system is installed in the living room and is designed to blend into the user's daily life. Its simple operation makes it easy to use for a wide range of users, from children to the elderly. Regular updates ensure that the latest information and functions are always provided, ensuring long-term usability. Using this system can alleviate the loneliness of living alone and allow users to easily access information in their daily lives. Furthermore, emotion-based responses allow users to enjoy a richer interactive experience. In this way, the present invention provides users with new value and experiences, supporting a richer lifestyle.

[1296] The processing flow will be explained below.

[1297] Step 1:

[1298] The user speaks a question or request to the robot, for example, "What's the weather like today?"

[1299] Step 2:

[1300] The device uses a microphone to capture the user's voice data, which is then temporarily stored digitally within the device.

[1301] Step 3:

[1302] The device sends the captured audio data to the server using a communication protocol such as an HTTP request or WebSocket.

[1303] Step 4:

[1304] The server receives the voice data sent from the terminal and temporarily stores it in the server's storage.

[1305] Step 5:

[1306] The server converts the voice data into text data using a speech recognition engine, which analyzes the voice data and generates a corresponding string of characters.

[1307] Step 6:

[1308] The server sends the converted text data to the emotion engine, which analyzes the text data and recognizes the user's emotions (e.g., joy, sadness, anger, etc.).

[1309] Step 7:

[1310] The server inputs the emotional information and text data obtained from the emotion engine into the generative AI model, which then generates an appropriate answer based on the user's emotions and text data.

[1311] Step 8:

[1312] The server formats the answer obtained from the generative AI model as text data and prepares it to be sent to the device.

[1313] Step 9:

[1314] The server sends the formatted text data and emotion information to the device, again via HTTP responses or WebSocket messages.

[1315] Step 10:

[1316] The device receives the text data and emotion information from the server and temporarily stores them.

[1317] Step 11:

[1318] The device converts the text data into speech data using a speech synthesis engine, which analyzes the text data and generates a corresponding speech signal.

[1319] Step 12:

[1320] The device then plays the generated voice data over a speaker, allowing the user to hear the answer to their question. The tone and intonation of the voice may be adjusted based on the emotional information.

[1321] Step 13:

[1322] The user listens to the voice response played back from the device. The emotional feedback confirms that their feelings are understood. This system allows the user to feel emotional satisfaction through the interaction.

[1323] Example 2

[1324] 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."

[1325] Conventional voice recognition systems do not take user emotions into account, making dialogue mechanical and difficult to achieve natural communication. Furthermore, the lack of regular updates to generative AI models and emotion recognition engines has led to problems with the system's accuracy and response quality declining. Furthermore, users are unable to confirm whether their emotions are being understood, which leads to a decline in dialogue satisfaction.

[1326] 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.

[1327] In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion recognition engine to analyze the user's emotions, means for inputting the text data and emotion information into a generative AI model to generate an appropriate answer, means for transmitting the generated answer as text data, means for speech synthesis to convert the text data into voice data, means for playing the voice data, and means for feeding back the emotion recognition results to the user. This enables natural dialogue according to the user's emotions, and regular updates can maintain the accuracy and quality of the system's responses, improving user satisfaction.

[1328] "Voice Data" means a digital representation of voice information uttered by a User.

[1329] "Text data" is digital information that is audio data converted into text format.

[1330] An "emotion recognition engine" is a software module that analyzes text data to identify a user's emotions.

[1331] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers based on input text data and emotional information.

[1332] A "voice recognition engine" is a software module for converting voice data into text data.

[1333] A "speech synthesis means" is a software module and hardware device for converting text data into speech data.

[1334] A "terminal device" is a hardware device that captures voice data from a user and transmits it to a server.

[1335] The "server" is a central processing unit that processes voice data and performs emotion recognition and response generation.

[1336] A "prompt" is an instruction or question to be input into a generative AI model.

[1337] This invention is a system that receives voice data and uses a generative AI model and an emotion recognition engine to engage in natural dialogue with the user. It also recognizes the user's emotions and provides appropriate responses based on those emotions.

[1338] Specifically, the following hardware and software are used.

[1339] Hardware:

[1340] 1. Terminal equipment:

[1341] Microphone: Captures the user's voice.

[1342] Speaker: Provides a spoken response to the user.

[1343] Communication module: Performs data communication with the server.

[1344] 2. Server:

[1345] Central processing unit: processes voice data, recognizes emotions, and generates responses.

[1346] software:

[1347] 1. Speech recognition engine (e.g. Google Speech-to-Text API)

[1348] Convert the audio data into text data.

[1349] 2. Emotion recognition engine (e.g., Microsoft Azure Text Analytics)

[1350] Analyze text data to identify user sentiment.

[1351] 3. Generative AI models (e.g., OpenAI GPT-4)

[1352] Generate appropriate answers based on text data and emotional information.

[1353] 4. Speech synthesis engine (e.g. IBM Watson Text-to-Speech)

[1354] Convert text data into audio data.

[1355] Detailed instructions on how to do this:

[1356] The server uses a voice recognition engine to convert the voice data sent from the terminal device into text data. The converted text data is analyzed by an emotion recognition engine to identify the user's emotion. This text data and emotion information are then input into a generative AI model to generate an appropriate answer. This answer is formatted as text data and sent from the server to the terminal device. The terminal device then converts the received text data into voice data using a voice synthesis engine and responds to the user audibly using a speaker.

[1357] Examples:

[1358] When a user asks, "What's the weather today?", the terminal device captures the voice and sends it to the server. The server uses a speech recognition engine to convert this voice data into text data, generating the text "What's the weather today?". It then uses an emotion recognition engine to identify the user's emotion (e.g., "interested") from this text and provides that emotion information to the generative AI model. The generative AI model then generates an appropriate answer based on the following prompt sentence, for example:

[1359] Example prompt sentence:

[1360] "A user asks, 'What's the weather like today?' The user's emotion is 'interested.' Generate an appropriate answer based on the user's emotion."

[1361] The generative AI model generates a response such as "Today's weather is sunny," which the server sends to the terminal device as text data. The terminal device then uses a speech synthesis engine to convert the text data into voice data, responding to the user in a bright tone: "Today's weather is sunny."

[1362] This system allows users to enjoy a natural, emotion-sensitive conversational experience, helping them to obtain information on a daily basis and alleviating the loneliness of living alone. Furthermore, feedback on emotion recognition results allows users to confirm that their emotions are being understood by the system.

[1363] Furthermore, the server regularly updates the generative AI model and emotion recognition engine, allowing it to continually provide the latest information and advanced functionality, thereby increasing the long-term utility value of the system and improving user satisfaction.

[1364] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1365] Step 1:

[1366] The user provides voice input.

[1367] Specifically, the user speaks a question such as "What's the weather like today?" This speech is input into the system. The input data is the user's speech, and the output is voice data.

[1368] Step 2:

[1369] The device captures the audio data.

[1370] The device's microphone captures the user's voice and records it as digital voice data. During this process, voice data is input and the voice data is output as is. For example, the voice saying "What's the weather like today?" is captured in digital form.

[1371] Step 3:

[1372] The terminal transmits the voice data to the server.

[1373] The communication module of the device sends the captured audio data to the server. The input here is the audio data of the device, and the output is the transmission of audio data to the server. Specifically, the device sends the audio data to the server using an HTTP request.

[1374] Step 4:

[1375] The server converts the voice data into text data.

[1376] The server converts the received voice data into text data using a voice recognition engine (e.g., Google Speech-to-Text API). The input of this step is voice data, and the output is text data. For example, the voice data "What's the weather like today?" is converted into the text data "What's the weather like today?"

[1377] Step 5:

[1378] The server analyzes emotions using an emotion recognition engine.

[1379] The server inputs the text data into an emotion recognition engine (e.g., Microsoft Azure Text Analytics) to analyze the user's emotions. The input is text data, and the output is emotion information. Specifically, the emotion "interest" is identified from the text data "What's the weather like today?"

[1380] Step 6:

[1381] The server inputs a prompt into the generative AI model and generates an appropriate answer.

[1382] The server inputs a prompt based on text data and emotional information into a generative AI model (e.g., OpenAI GPT-4) and generates an appropriate answer. The input for this step is text data and emotional information, and the output is the generated answer text data. As a specific example, the prompt "The user asked, 'What's the weather like today?' The user's emotion at that time is 'interested.' Please generate an appropriate answer based on the user's emotion." is input, and the AI ​​model generates the text "The weather is sunny today."

[1383] Step 7:

[1384] The server transmits the generated text data to the terminal.

[1385] The server sends the generated answer to the terminal as text data. The input is the generated text data, and the output is the transmission of the text data to the terminal. Specifically, the text data "Today's weather is sunny" is sent to the terminal.

[1386] Step 8:

[1387] The terminal converts the text data into audio data.

[1388] The device uses a speech synthesis engine (e.g., IBM Watson Text-to-Speech) to convert text data into voice data. The input is text data and the output is voice data. For example, the text data "The weather is sunny today" is converted into voice data.

[1389] Step 9:

[1390] The device responds to the user with voice and also provides feedback on the emotional results.

[1391] The device plays the converted voice data through a speaker and also feeds back the emotional result to the user. The input is the voice data and the emotional result, and the output is a voice response to the user and emotional feedback. As a concrete example, the device tells the user, "Today's weather is sunny," in a bright tone.

[1392] Step 10:

[1393] The server periodically updates the generative AI model and emotion recognition engine.

[1394] The server regularly updates the generative AI model and emotion recognition engine to maintain the system's accuracy and response quality. The input is a new dataset and algorithm, and the output is an updated generative AI model and emotion recognition engine. Specifically, the server trains the generative AI model using a new dataset once a month and updates the emotion recognition engine algorithm to the latest version.

[1395] (Application example 2)

[1396] 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."

[1397] Conventional voice dialogue systems simply convert voice data into text and generate responses using a generative AI model, limiting their ability to provide personalized responses that take the user's emotions into account. In brick-and-mortar stores in particular, appropriate responses based on customer emotions are often required, and in this regard, existing technologies are inadequate. Furthermore, delays in information delivery during the voice recognition and response generation process, as well as inaccurate emotion recognition, are factors that reduce the quality of customer service. It is necessary to solve these problems.

[1398] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice data, means for converting voice data into text data, means for inputting the text data into an emotion engine to recognize the user's emotion, means for inputting the text data and emotion information into a generative AI model to generate a response in an appropriate tone, means for transmitting the generated response as text data, voice synthesis means for converting the text data into voice data, and means for playing back the voice data. This makes it possible to quickly provide an appropriate response according to the user's emotion and improve customer satisfaction.

[1399] "Audio data" refers to data in which audio is recorded in digital format.

[1400] "Text data" is data expressed in the form of a string of characters.

[1401] A "generative AI model" is an artificial intelligence model that analyzes user input data and generates appropriate answers based on it.

[1402] An "emotion engine" is a system that estimates a user's emotions from text and voice data.

[1403] "Appropriate tone" refers to the appropriate speaking style and tone of voice that corresponds to the user's emotions.

[1404] A "voice recognition engine" is a system that analyzes voice data and converts it into text data.

[1405] "Speech synthesis means" is a technology that converts text data into voice data.

[1406] "User" refers to an individual or group that uses the system.

[1407] A "server" is a central computing device that performs various processes for the system.

[1408] A "terminal device" is a device operated by a user that captures and plays audio.

[1409] System Configuration

[1410] The embodiment of the invention employs the following system configuration: The system is mainly composed of three entities: a server, a terminal device, and a user, and each function operates in cooperation with the others.

[1411] Hardware Configuration

[1412] Server: A high-performance computing device that utilizes cloud or on-premise servers.

[1413] Terminal Device: A user interface device such as a smartphone, smart glasses, or tablet.

[1414] Microphone: A device for inputting the user's voice.

[1415] Speaker: A device for playing back generated audio data.

[1416] Software Configuration

[1417] Voice capture module: Installed on the terminal device, it captures the user's voice data.

[1418] Speech recognition engine: Used to convert voice data into text data (e.g., Google Cloud Speech-to-Text).

[1419] Emotion engine: Recognizes user emotions from text data (e.g., IBM Watson Tone Analyzer).

[1420] Generative AI models: Generate appropriate answers based on user input and emotional information (e.g., OpenAI GPT-4).

[1421] Speech synthesis engine: Converts the generated text data into speech data (e.g., Amazon Polly).

[1422] System Operation Overview

[1423] Step 1: Capture audio

[1424] The voice capture module of the terminal device uses a microphone to capture the user's voice data, which is then transmitted to the server in real time.

[1425] Step 2: Voice Recognition

[1426] The server converts the received voice data into text data using a voice recognition engine. For example, if a user says, "I want new shoes," the voice data is converted into text data saying, "I want new shoes."

[1427] Step 3: Emotion Recognition

[1428] The converted text data is then fed into an emotion engine, which analyzes the user's emotions. For example, if the user's comments are recognized as "excited," that emotional information is provided to the generative AI model.

[1429] Step 4: Answer Generation

[1430] The generative AI model generates an appropriate response based on the text data and emotional information. For example, based on the text data "I want new shoes" and the emotional information "I'm excited," the generative AI model generates the following response: "Thank you for visiting! New shoes, then let me show you our latest collection!"

[1431] Step 5: Text-to-Speech

[1432] A speech synthesis engine is used to convert the generated response text data into speech data, which generates a cheerful voice saying, "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!"

[1433] Step 6: Play back the response

[1434] The terminal device plays back the voice data generated by the voice synthesis engine, and the response is sent to the user through the speaker.

[1435] Prompt Sentence Examples

[1436] The customer says, "I want new shoes." The customer is now excited. An appropriate response would be:

[1437] This allows users to receive appropriate responses based on their emotions, improving the customer service experience in physical stores.

[1438] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1439] Step 1:

[1440] The terminal uses the voice capture module to capture the user's voice data through the microphone. The input is the user's voice, and the output is the voice data. This voice data is sent to the server in real time.

[1441] Step 2:

[1442] The voice data received by the server is converted into text data using a voice recognition engine (e.g., Google Cloud Speech-to-Text). The input is voice data and the output is text data. For example, voice data saying "I want new shoes" is converted into text data saying "I want new shoes."

[1443] Step 3:

[1444] The converted text data is input into the server's emotion engine (e.g., IBM Watson Tone Analyzer), which analyzes the user's emotions. The input is text data, and the output is emotional information. For example, the emotional information "excited" can be obtained from the text data "I want new shoes."

[1445] Step 4:

[1446] The server inputs text data and emotional information into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates an appropriate answer based on this information. The input is text data and emotional information, and the output is response text data. For example, from the information "I want new shoes" and "Excited," the response text data "Thank you for visiting! New shoes, let me show you our latest collection!" is generated.

[1447] Step 5:

[1448] The generated response text data is input to a speech synthesis engine (e.g., Amazon Polly) and converted into voice data. The input is the response text data, and the output is the response voice data. The speech synthesis engine converts the text data into voice data with a bright tone.

[1449] Step 6:

[1450] The device plays back the response voice data generated by the speech synthesis engine. The input is the response voice data, and the output is a voice response that the user can hear. For example, a voice saying "Thank you for visiting our store! These are new shoes. Let us show you our latest collection!" is played through the device's speaker.

[1451] The above is the specific processing flow of the program of the system that realizes the application example.

[1452] 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.

[1453] 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.

[1454] 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.

[1455] 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.

[1456] 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.

[1457] 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.

[1458] 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).

[1459] 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.

[1460] 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."

[1461] 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.

[1462] 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).

[1463] 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.

[1464] 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.

[1465] 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.

[1466] 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.

[1467] 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.

[1468] 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.

[1469] 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.

[1470] 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.

[1471] 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.

[1472] 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.

[1473] The following is further disclosed regarding the above embodiment.

[1474] (Claim 1)

[1475] means for receiving audio data;

[1476] means for converting the voice data into text data;

[1477] A means for inputting the text data into a generative AI model to generate an appropriate answer;

[1478] means for transmitting the generated answer as text data;

[1479] a voice synthesis means for converting the text data into voice data;

[1480] means for reproducing the audio data;

[1481] A system including:

[1482] (Claim 2)

[1483] 10. The system of claim 1, wherein the generative AI model is periodically updated.

[1484] (Claim 3)

[1485] 2. The system according to claim 1, further comprising a terminal device that transmits the voice data to a server and receives text data from the server.

[1486] "Example 1"

[1487] (Claim 1)

[1488] means for receiving audio data;

[1489] a voice recognition means for converting the voice data into text data;

[1490] A means for inputting the text data into a generative AI model to generate an appropriate answer;

[1491] means for transmitting the generated answer as text data;

[1492] a voice synthesis means for converting the text data into voice data;

[1493] means for reproducing the audio data;

[1494] a means for the terminal device to communicate with a server, transmit the voice data to the server, and receive text data from the server;

[1495] A means for the server to periodically update the generative AI model;

[1496] A system including:

[1497] (Claim 2)

[1498] The system according to claim 1, characterized in that the generative AI model generates an appropriate answer based on a prompt sentence from an external input.

[1499] (Claim 3)

[1500] 2. The system of claim 1, wherein the terminal device captures voice data from the user and communicates with the server to transmit the voice data.

[1501] "Application Example 1"

[1502] (Claim 1)

[1503] means for receiving audio data;

[1504] means for converting the voice data into text data;

[1505] A means for inputting the text data into a generative AI model to generate an appropriate answer;

[1506] means for transmitting the generated answer as text data;

[1507] a voice synthesis means for converting the text data into voice data;

[1508] means for reproducing the audio data;

[1509] A way to receive orders via user voice and confirm questions or special requests through natural dialogue.

[1510] means for placing a food delivery order based on voice instructions;

[1511] A system including:

[1512] (Claim 2)

[1513] 10. The system of claim 1, wherein the generative AI model is periodically updated.

[1514] (Claim 3)

[1515] 2. The system according to claim 1, further comprising a terminal device that transmits the voice data to a server and receives text data from the server.

[1516] "Example 2: Combining Emotion Engines"

[1517] (Claim 1)

[1518] means for receiving audio data;

[1519] means for converting the voice data into text data;

[1520] means for inputting the text data into an emotion recognition engine to analyze the emotion of the user;

[1521] A means for inputting the text data and emotion information into a generative AI model to generate an appropriate answer;

[1522] means for transmitting the generated answer as text data;

[1523] a voice synthesis means for converting the text data into voice data;

[1524] means for reproducing the audio data;

[1525] a means for feeding back the emotion recognition result to a user;

[1526] A system including:

[1527] (Claim 2)

[1528] 10. The system of claim 1, wherein the generative AI model and emotion recognition engine are periodically updated.

[1529] (Claim 3)

[1530] 2. The system according to claim 1, further comprising a terminal device that transmits the voice data to a server and receives text data from the server.

[1531] "Application example 2 when combining emotion engines"

[1532] (Claim 1)

[1533] means for receiving audio data;

[1534] means for converting the voice data into text data;

[1535] A means for inputting the text data into a generative AI model to generate an appropriate answer;

[1536] means for transmitting the generated answer as text data;

[1537] a voice synthesis means for converting the text data into voice data;

[1538] means for reproducing the audio data;

[1539] means for inputting the text data into an emotion engine to recognize a user's emotion;

[1540] A means for the generative AI model to generate an answer in an appropriate tone based on the emotion information;

[1541] A system including:

[1542] (Claim 2)

[1543] 10. The system of claim 1, wherein the generative AI model is periodically updated.

[1544] (Claim 3)

[1545] 2. The system according to claim 1, further comprising a terminal device that transmits the voice data to a server and receives text data from the server. [Explanation of symbols]

[1546] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving audio data; means for converting the voice data into text data; A means for inputting the text data into a generative AI model to generate an appropriate answer; means for transmitting the generated answer as text data; a voice synthesis means for converting the text data into voice data; means for reproducing the audio data; A system including:

2. 10. The system of claim 1, wherein the generative AI model is periodically updated.

3. 2. The system according to claim 1, further comprising a terminal device for transmitting said voice data to a server and receiving text data from said server.

Images