system

Abstract

Provide a system. A voice input means; a speech recognition means for converting speech input data into character data; an analysis means for analyzing character data using generative artificial intelligence to generate an answer; a speech synthesis means for converting the generated answer into speech; an output means for reproducing the voice generated by the voice synthesis means; A system including:

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 recent years, there has been an increasing demand for systems that allow users to ask questions by voice and automatically provide answers by voice. However, current systems often lack the accuracy of question analysis and the appropriateness of answers, resulting in a poor user experience. Generating answers to complex questions is particularly difficult when specific product knowledge must be provided quickly and accurately. Another problem is the low accuracy of speech recognition, which can lead to misrecognition and reduced reliability of answers. In these circumstances, there is a need for a system that can consistently perform everything from speech recognition to providing appropriate information with high accuracy. [Means for solving the problem]

[0005] The present invention provides a system including a voice input means, a voice recognition means for converting voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence to generate an answer, a voice synthesis means for converting the generated answer into voice, and an output means for playing back the voice generated by the voice synthesis means. With this system, when a user voice-asks a question, the voice recognition means converts the voice into text data, and the analysis means analyzes the question and generates an appropriate answer. The generated answer is then converted into voice by the voice synthesis means and read aloud to the user by the output means. In this way, the user can receive information quickly and accurately. Furthermore, by using the intention recognition means described in claim 2, the user's intention can be accurately estimated and a more appropriate answer can be generated. Furthermore, by using the search means described in claim 3, necessary information can be efficiently obtained from a product knowledge database, improving the appropriateness of the answer.

[0006] A "voice input means" is a device or system that receives voice from a user.

[0007] "Speech recognition means" refers to a technology or system that converts received voice data into text data.

[0008] "Generative AI" is an AI technology that analyzes input data and generates appropriate answers and information.

[0009] "Analysis means" refers to a technology or system that analyzes text data, understands the intent of the question, and generates an appropriate answer.

[0010] "Speech synthesis means" refers to a technology or system that converts character data into voice data.

[0011] The "output means" is a device or system for providing the user with the voice data generated by the voice synthesis means.

[0012] "Intention recognition means" is a technology or system that estimates the background intention of a user's question or statement and uses this information to analyze the question or statement appropriately and generate an answer.

[0013] A "search means" is a technology or system for searching and retrieving necessary information from a database.

[0014] A "product knowledge database" is a database in which detailed knowledge and information about products is registered. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[0037] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server converts the voice into text data using voice recognition means. Then, analysis means using generative artificial intelligence analyzes the text data and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by voice synthesis means, and finally provided to the user through output means (e.g., a speaker).

[0038] Below, the specific program processing will be explained in natural language.

[0039] Program processing explanation

[0040] 1. Voice input

[0041] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0042] Device: Captures the user's voice through a microphone.

[0043] 2. Voice Recognition

[0044] Device: Sends captured audio data to the server.

[0045] Server: Uses speech recognition means to convert the received voice data into text data.

[0046] Server: For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0047] 3. Question Analysis and Answer Generation

[0048] Server: Analyzes the received text data using generative AI. The analysis means understands the meaning of the question and searches for information from a related product knowledge database.

[0049] Server: For example, generate information such as "New smartphone plans start at 5,000 yen per month" as an answer to a question.

[0050] 4. Speech Synthesis

[0051] Server: Sends the generated answer text to the speech synthesis means.

[0052] Server: The speech synthesis means converts the text data into speech data, generating speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0053] 5. Audio Output

[0054] Server: Sends the generated audio data to the device.

[0055] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to questions.

[0056] Specific examples

[0057] User Question: What is the pricing plan for my new smartphone?

[0058] Capture audio input: The device captures audio and sends it to the server.

[0059] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0060] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[0061] Speech synthesis: Converting the generated text into speech.

[0062] Audio output: Provides answers to the user through the device's speaker.

[0063] In this way, the system based on the present invention can provide a prompt and accurate answer to a user's voice question. Furthermore, by linking with an intent recognition means and a product knowledge database, it can generate even more appropriate answers, improving the user experience. This system is particularly useful in situations where detailed information about a product is to be provided, enabling efficient customer service.

[0064] The processing flow will be explained below.

[0065] Step 1:

[0066] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0067] Step 2:

[0068] Device: Captures the user's voice through a microphone.

[0069] Step 3:

[0070] Device: Sends captured audio data to the server.

[0071] Step 4:

[0072] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0073] Step 5:

[0074] Server: Analyzes the received text data using generative AI. The analysis means analyzes the text data and understands the intent of the question.

[0075] Step 6:

[0076] Server: Searches for information corresponding to the question from a product knowledge database. For example, retrieves information on "new smartphone pricing plans" from the database.

[0077] Step 7:

[0078] Server: Generates an answer to the question based on the search results. For example, it generates the text "New smartphone plans start at 5,000 yen per month" as an answer to the user's question.

[0079] Step 8:

[0080] Server: Sends the generated answer text to the speech synthesis means.

[0081] Step 9:

[0082] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0083] Step 10:

[0084] Server: Sends the generated audio data to the device.

[0085] Step 11:

[0086] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0087] This series of processes allows the user to ask a question by voice and quickly receive an appropriate answer by voice.

[0088] Example 1

[0089] 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."

[0090] Modern information provision systems require a smooth process for users to ask questions by voice and receive appropriate answers. However, existing systems often have problems with speech recognition accuracy and answer generation and reproduction. In particular, they can be difficult to provide accurate and appropriate answers in real time. They also lack the ability to accurately understand user intent and generate the optimal answer. This can lead to a poor user experience.

[0091] 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.

[0092] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence to generate an answer, a voice synthesis means for converting the generated answer into speech, a means for transmitting the generated speech data to an output means, and an output means for playing back the speech generated by the voice synthesis means. This allows a user to obtain a highly accurate answer in real time when asking a question by voice. Furthermore, by linking with an intent recognition means and a product knowledge database, it is possible to generate more appropriate answers and improve the user experience.

[0093] "Voice input means" refers to a device or software that captures a user's speech or voice data and records it as digital data.

[0094] "Speech recognition means" refers to technology that analyzes voice data and converts the voice into corresponding text data.

[0095] "Generative AI" refers to an AI technology that analyzes given text or data and creates optimal answers or products based on that content.

[0096] "Analysis means" refers to a process or device that analyzes received text data, understands the user's intent, and generates an appropriate response.

[0097] "Speech synthesis means" refers to technology that converts text data into voice data.

[0098] "Output means" refers to a device or system for playing back the generated audio data.

[0099] "Intention recognition means" refers to the technology or process for inferring a user's intentions or desires from the data entered by the user.

[0100] A "product knowledge database" refers to a database that stores detailed information such as the features, prices, and specifications of specific products.

[0101] "Search methods" refer to the techniques and processes used to search for and retrieve data or information based on specific criteria.

[0102] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[0103] When a user asks a question, their voice is captured through a voice input means (a microphone or the built-in microphone of a smart device). This voice data is sent to a server by the device. For example, when using a smartphone or tablet, the voice data is sent via the Internet.

[0104] The server uses a speech recognition means to convert the received voice data into text data. For example, Google® Cloud Speech-to-Text API can be used. The voice data is converted into text data such as "Please tell me the price plan for my new smartphone."

[0105] Next, the text data is analyzed using generative artificial intelligence (e.g., OpenAI (registered trademark) GPT-4 (registered trademark)). This analysis means understands the meaning of the user's question and generates an answer based on related information. Specifically, based on the context and keywords of the question, an answer appropriate to the question content is searched from a product knowledge database and an appropriate answer text is generated.

[0106] The server sends the generated response text to a speech synthesis means (e.g., Amazon Polly), which converts the text data into speech data. The speech data generated is, "The new smartphone plan starts at 5,000 yen per month."

[0107] Finally, the generated voice data is sent to the terminal and played back through the terminal's output means (speaker or headset), allowing the user to receive the answer to their question by voice.

[0108] For example, if a user asks, "What is the pricing plan for my new smartphone?", the following prompt is input to the generative AI:

[0109] Example prompt sentence:

[0110] User Question: What is the pricing plan for my new smartphone?

[0111] Generative AI answer: New smartphone plans start at 5,000 yen per month.

[0112] Based on this prompt, the generative artificial intelligence generates an appropriate response, which is then converted into voice data by a speech synthesis means and finally provided to the user. This system allows users to receive fast and accurate information by voice. It is particularly useful in situations where detailed information about products is to be provided, and realizes efficient customer service.

[0113] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0114] Step 1: Voice Input

[0115] User: Asks the device, "What is the price plan for my new smartphone?" The user's input is voice data.

[0116] Device: Uses a built-in microphone to capture the user's voice, which is then stored digitally.

[0117] Output: The captured audio data is generated.

[0118] Step 2: Sending audio data

[0119] On your device: The captured audio data is sent to the server using the HTTPS protocol, which sends the data securely over your internet connection.

[0120] Input: Audio data

[0121] Output: Audio data sent to the server

[0122] Step 3: Voice Recognition

[0123] Server: Calls the API of the speech recognition method (for example, Google Cloud Speech-to-Text API) and converts the received voice data into text data.

[0124] Input: Audio data sent to the server

[0125] Data processing: Converting audio data into text using speech recognition tools. This process involves analyzing the audio data and generating a corresponding string of characters.

[0126] Output: Converted text data (e.g., "What is the price plan for my new smartphone?")

[0127] Step 4: Parsing the question and generating an answer

[0128] Server: Input the text data into a generative artificial intelligence (e.g., OpenAI GPT-4) and analyze it.

[0129] Input: Converted character data

[0130] Data calculation: The generative AI model generates appropriate answers to questions based on prompts, searching for information from a product knowledge database based on the context and keywords of the question.

[0131] Output: Generated answer text (e.g., "New smartphone plans start at 5,000 yen per month.")

[0132] Step 5: Text-to-speech synthesis of the answer text

[0133] Server: The generated answer text is input into a speech synthesis means (e.g., Amazon Polly) and converted into voice data.

[0134] Input: Generated answer text

[0135] Data processing: Converting text into audio data using speech synthesis tools.

[0136] Output: Generated speech data (e.g., "New smartphone plans start at 5,000 yen per month.")

[0137] Step 6: Sending audio data

[0138] Server: Sends the generated audio data to the device using the HTTP / HTTPS protocol to ensure secure communication.

[0139] Input: Generated audio data

[0140] Output: Audio data sent to the device

[0141] Step 7: Audio Output

[0142] Terminal: Passes the audio data received by the terminal to the output means (speaker or headset).

[0143] Input: Audio data sent to the device

[0144] What it does: Plays audio data through a speaker and provides answers to the user.

[0145] Output: A spoken response to the user (e.g., "New smartphone plans start at ¥5,000 per month.")

[0146] By going through the above processing steps, when a user asks a question by voice, the user can quickly and accurately receive a response by voice.

[0147] (Application example 1)

[0148] 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."

[0149] Conventional voice recognition systems and automated response systems focus only on simple voice input and conversion to text data, and lack the ability to respond to detailed user needs and complex questions. Furthermore, especially in new digital markets such as virtual stores, there is a growing need for systems that allow users to quickly and accurately obtain product information and engage in real-time voice interaction. Therefore, it is necessary to provide a new system that meets these needs and improves the user experience.

[0150] 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.

[0151] In this invention, the server includes a speech recognition unit that converts voice input data into text data, an analysis unit that uses generative artificial intelligence to analyze the text data and generate a response, and a speech synthesis unit that converts the generated response into speech. This allows the server to capture voice data when a user asks a question by voice in a virtual store, execute processing using cloud computing technology, and provide the user with a quick and accurate response. Furthermore, by obtaining information from a product knowledge database and accurately recognizing the user's intent, the server can generate a more appropriate response.

[0152] "Audio input means" refers to a device and method for capturing user-uttered speech as a digital signal.

[0153] "Speech recognition means" refers to the technology and devices for converting voice input data into text data.

[0154] "Analysis means" refers to technology and devices that have the function of analyzing received text data using generative artificial intelligence and generating answers to users' questions.

[0155] "Speech synthesis means" refers to the technology and device that converts the generated response text into voice data.

[0156] "Output means" refers to a device that reproduces the voice generated by the voice synthesis means.

[0157] "Cloud computing technology" refers to technology that performs distributed processing over the Internet and efficiently executes large-scale data analysis and calculations.

[0158] A "virtual store" refers to a virtual shopping space built on the Internet where users can browse and purchase products.

[0159] "Apparatus" means a machine or mechanism designed for a specific purpose.

[0160] "Generative AI" refers to artificial intelligence technology that has the ability to learn large amounts of data and generate new information and answers based on human language and knowledge.

[0161] A "product knowledge database" refers to a database that stores detailed information related to products.

[0162] The present invention relates to a system in which a user can ask a question by voice and a response to the question is provided by voice, and the system uses a voice input means, a voice recognition means, an analysis means, a voice synthesis means, an output means, cloud computing technology, and a voice data capture device in a virtual store.

[0163] When a user asks a question through a voice input device in the virtual store, the voice is captured as a digital signal through a microphone. This digital signal is voice input data, and the terminal transmits the voice input data to a cloud server.

[0164] The cloud server uses the Google Cloud Speech-to-Text API to convert this voice input data into text data. This text data represents the user's question. For example, if a user asks, "Tell me about this product," the cloud server converts this into text data: "Tell me about this product."

[0165] Next, this text data is analyzed using generative artificial intelligence. OpenAI's GPT-3 (registered trademark) 5 is used as the analysis method. The cloud server analyzes the text data using the analysis method and generates the optimal answer to the user's question. The analysis method understands the meaning of the question and extracts the necessary information from the product knowledge database.

[0166] The answer text generated by the analysis means contains specific information such as "This product is the latest model, priced at 50,000 yen, and in stock." This answer text is converted into audio data using the Google Cloud Text-to-Speech API. This audio data is provided in an audio format that is easy for the user to understand.

[0167] Finally, the generated voice data is played back by a voice synthesis means, and the speaker of the smart glasses or head-mounted display is used as an output means, allowing the user to receive a voice response to their question.

[0168] Specific examples are shown below.

[0169] When a user asks, "Tell me about this product," the smart glasses' microphone captures the voice and sends it to a cloud server. The cloud server converts the text data using the Google Cloud Speech-to-Text API, analyzes it using GPT-3.5, and generates an answer. The text data is then converted back into audio using the Google Cloud Text-to-Speech API, and the audio is played through the smart glasses' speaker.

[0170] An example of a prompt is as follows:

[0171] User Asks: "Tell me about this product"

[0172] Data to send to Google Cloud Speech-to-Text API: Audio data

[0173] Prompt GPT-3.5: "The user says, 'Tell me about this product.' Extract information from the corresponding product database and generate an appropriate answer."

[0174] Data to send to Google Cloud Text-to-Speech API: Generated response text

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

[0176] Step 1:

[0177] Voice input

[0178] A user asks a question to a terminal in a virtual store, such as "Tell me about this product." The terminal uses a voice input means to capture the user's voice as a digital signal. This captured voice data is the input. This input data is ready to be sent to the subsequent processing steps.

[0179] Step 2:

[0180] Voice Recognition

[0181] The device sends the captured voice data to a cloud server, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. During this conversion process, the voice waveform is analyzed and converted into the corresponding text, "Tell me about this product." This text data is the output. The phonemes of the voice are analyzed, and the most appropriate text is generated using a language model.

[0182] Step 3:

[0183] Question analysis and answer generation

[0184] The cloud server analyzes the text data using generative artificial intelligence (OpenAI's GPT-3.5). This analysis method understands the user's question and extracts appropriate information from a product knowledge database. The input is the converted text data "Tell me about this product," and the output is the answer text generated based on the analysis: "This product is the latest model, priced at 50,000 yen. In stock." The generative AI model analyzes the question and generates relevant information by filtering it from the database.

[0185] Step 4:

[0186] Speech synthesis

[0187] The cloud server sends the generated answer text to the Google Cloud Text-to-Speech API, which converts it into audio data. The input is the answer text, and the output is the corresponding audio data. In this process, the answer text is synthesized into natural-sounding audio and provided in a format that is easy for the user to understand. The generated audio data is prepared here.

[0188] Step 5:

[0189] Audio Output

[0190] The cloud server sends the generated voice data to the device. The device uses an output means (speakers in smart glasses or a head-mounted display) to play this voice to the user. The input is the generated voice data, and the output is a voice response played to the user. The device outputs the voice data through a designated speaker and provides it to the user.

[0191] These steps enable the system to quickly and accurately provide a spoken answer to a user's spoken question.

[0192] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0193] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. This system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit. It also includes a configuration that further combines an emotion engine that recognizes the user's emotions and provides appropriate answers based on those emotions.

[0194] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server then converts the voice into text data using a voice recognition means. The server then uses an emotion engine to recognize the user's emotion from the text data and provides the emotion information to the analysis means. Next, the analysis means using generative artificial intelligence analyzes the text data and emotion information and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by a voice synthesis means and finally provided to the user through an output means (e.g., a speaker).

[0195] Below, the specific program processing will be explained in natural language.

[0196] Program processing explanation

[0197] 1. Voice input

[0198] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0199] Device: Captures the user's voice through a microphone.

[0200] 2. Voice Recognition

[0201] Device: Sends captured audio data to the server.

[0202] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0203] 3. Emotion recognition

[0204] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it recognizes whether the user is excited or confused from a text such as "New smartphone pricing plan."

[0205] Server: Provides the recognized emotion information to the analysis means.

[0206] 4. Question Analysis and Answer Generation

[0207] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[0208] 5. Adjusting your answers

[0209] Server: Adjust the answer appropriately based on the perceived emotion information, for example adding more detailed explanation if the user is confused.

[0210] 6. Speech Synthesis

[0211] Server: Sends the generated answer text to the speech synthesis means.

[0212] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0213] 7. Audio Output

[0214] Server: Sends the generated audio data to the device.

[0215] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0216] Specific examples

[0217] User Question: What is the pricing plan for my new smartphone?

[0218] Capture audio input: The device captures audio and sends it to the server.

[0219] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0220] Emotion recognition using an emotion engine: Recognizes when a user is confused from text data.

[0221] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[0222] Tailor your response: For confused users, add an additional "Would you like more options?"

[0223] Speech synthesis: Converting the generated text into speech.

[0224] Audio output: Provides answers to the user through the device's speaker.

[0225] In this way, the system according to the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is particularly useful in situations where detailed information about products is provided, realizing efficient customer service.

[0226] The processing flow will be explained below.

[0227] Step 1:

[0228] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0229] Step 2:

[0230] Device: Captures the user's voice through a microphone.

[0231] Step 3:

[0232] Device: Sends captured audio data to the server.

[0233] Step 4:

[0234] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0235] Step 5:

[0236] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it analyzes the text data and determines whether the user is interested, confused, or in a hurry.

[0237] Step 6:

[0238] Server: Provides the emotion information analyzed by the emotion engine to the analysis means.

[0239] Step 7:

[0240] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[0241] Step 8:

[0242] Server: Adjust the answer appropriately based on the perceived emotion information. For example, if the user is confused, add an additional explanation: "Would you like more options?"

[0243] Step 9:

[0244] Server: Sends the generated answer text to the speech synthesis means.

[0245] Step 10:

[0246] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0247] Step 11:

[0248] Server: Sends the generated audio data to the device.

[0249] Step 12:

[0250] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0251] This specific process allows the user to ask a question by voice and not only receive an appropriate answer quickly by voice, but also obtain a more friendly and understandable answer based on emotion recognition by the emotion engine.

[0252] Example 2

[0253] 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."

[0254] Conventional voice response systems can automatically generate and provide voice responses to user questions, but they have the problem of being unable to recognize the user's emotions and provide appropriate responses based on those emotions. This can lead to a poor user experience, especially when the user is confused or expecting something, as the system is unable to respond in accordance with that emotion. Furthermore, the responses often do not match the user's intentions, making it necessary to improve user satisfaction.

[0255] 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.

[0256] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into speech, an output means for playing back the generated speech, and an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the recognized emotion. This makes it possible to provide an appropriate answer according to the user's emotion, thereby improving the user experience and increasing user satisfaction. Furthermore, by using an intent recognition means, the server can accurately understand the user's intent and generate a more accurate answer.

[0257] "Audio input means" is a device or software for capturing speech produced by a user and transmitting that data to a processing system.

[0258] "Speech recognition means" means technology or equipment for converting voice data into text data, including algorithms and services used for that purpose.

[0259] "Generative AI" is an AI model used to analyze given input data and generate answers in a natural language format.

[0260] "Analysis means" refers to technology or devices that use generative artificial intelligence to analyze input data, understand the intent of the question, and generate an appropriate answer.

[0261] "Speech synthesis means" refers to a technique or device for converting generated character data into voice data.

[0262] "Output means" refers to hardware or software for reproducing audio data and providing information to the user by audio.

[0263] An "emotion recognition means" is a technology or device that identifies emotions from user input data and appropriately adjusts the system's response based on those emotions.

[0264] "Intention recognition means" refers to technology or devices that infer a user's intention from input data and generate appropriate analysis and responses.

[0265] A "product knowledge database" is a database that stores detailed information about products and serves as the basis for the system to search for information and generate answers.

[0266] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, an output unit, and an emotion recognition unit.

[0267] Specifically, when a user speaks a question through the microphone of a device on which the software is installed, the voice input means receives the voice and sends it to the server. For example, consider a situation where a user asks, "What is the price plan for my new smartphone?" The device's microphone captures the voice and transfers the data to the server. The server then uses the voice recognition means to convert the voice data into text data. This process can be performed using voice recognition services such as Google Cloud Speech-to-Text or Amazon Transcribe.

[0268] The emotion recognition means then analyzes the text data and recognizes the user's emotion. For example, it can recognize whether the user is excited or confused from the text "New smartphone pricing plan." This process can be performed using a Tone Analyzer by IBM Watson (registered trademark). The recognized emotion information is provided to the analysis means.

[0269] Next, an analysis means using generative AI (e.g., GPT-4) analyzes the text data and emotional information to generate an appropriate answer to the user's question. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month." This process is performed using a generative AI model.

[0270] The generated answer is sent to a speech synthesis means and converted into voice data. This voice synthesis can be performed using speech synthesis services such as Amazon Polly or Google Text-to-Speech. Finally, the generated voice data is sent to the device and played back to the user through the speaker.

[0271] The answer is also adjusted based on the emotion information recognized by the emotion recognition means. For example, if the user is confused, an explanation such as "Would you like more options?" is added to the answer.

[0272] Specific examples

[0273] User asks: "What plan is available for my new smartphone?"

[0274] Device audio input: Captures audio through the microphone and sends it to the server.

[0275] Server speech recognition: Converts speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0276] Emotion recognition: Analyzes text data to recognize when a user is confused.

[0277] Analysis and Answer Generation: Generate the answer "New smartphone plans start at 5,000 yen per month."

[0278] Tailor your response: For confused users, add the explanation "Would you like more options?"

[0279] Speech synthesis: Converting the generated text into audio data.

[0280] Speech output: Provides answers to the user through a speaker.

[0281] In this way, the system based on the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is useful in situations where detailed information about products is provided, realizing efficient customer service.

[0282] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0283] Step 1:

[0284] Voice input

[0285] User: The user asks the device, "What is the pricing plan for my new smartphone?"

[0286] On the device: The built-in microphone is used to capture the user's voice, and the voice data is generated and temporarily stored in the device's memory.

[0287] Step 2:

[0288] Sending audio data

[0289] Terminal: The captured audio data is compressed and sent to the server via the network using a transfer protocol such as HTTP or HTTPS.

[0290] Input: User's voice data

[0291] Output: Audio data sent to the server

[0292] Step 3:

[0293] Voice Recognition

[0294] Server: Passes the received voice data to the voice recognition means.

[0295] Server: Converts voice data into text data using a speech recognition method. Calls a speech recognition API (e.g., Google Cloud Speech-to-Text) and converts voice data into text data.

[0296] Input: Audio data

[0297] Output: The result of converting speech to text data, e.g. "What is the price plan for my new smartphone?"

[0298] Step 4:

[0299] emotion recognition

[0300] Server: Passes the text data to the emotion recognition means.

[0301] Server: Uses emotion recognition means to recognize the user's emotion from the text data. Calls an emotion recognition API (e.g., IBM Watson Tone Analyzer) to obtain emotion information from the text data.

[0302] Input: Character data

[0303] Output: Emotional information such as user confusion or expectation

[0304] Step 5:

[0305] Question analysis and answer generation

[0306] Server: Passes text data and emotion information to the analysis means.

[0307] Server: Uses generative AI (e.g., GPT-4) to analyze text data and emotional information and generate appropriate answers. Enter prompt sentences into the generative AI model to generate answers.

[0308] Input: Text data, emotion information

[0309] Output: Answer text "New smartphone plans start at 5,000 yen per month."

[0310] Step 6:

[0311] Adjusting your answers

[0312] Server: Adjust the generated answer based on sentiment information, for example, adding additional explanation to the answer if the user is confused.

[0313] Input: Answer text, emotion information

[0314] Output: Tailored response text, e.g. "New smartphone plans start at ¥5,000 / month. Would you like more options?"

[0315] Step 7:

[0316] Speech synthesis

[0317] Server: Pass the adjusted answer text to the speech synthesis means.

[0318] Server: Convert the response text into audio data using a speech synthesis tool. Call a speech synthesis API (e.g., Amazon Polly).

[0319] Input: Adjusted answer text

[0320] Output: Voice data, for example, "New smartphone plans start at 5,000 yen per month. Would you like more options?"

[0321] Step 8:

[0322] Audio Output

[0323] Server: Sends the generated audio data to the device.

[0324] Terminal: The generated audio data is played back through a speaker and provided to the user.

[0325] Input: Audio data

[0326] Output: The answer provided to the user verbally

[0327] (Application example 2)

[0328] 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."

[0329] In factories, it is important for workers to quickly and accurately obtain the information they need. However, conventional methods require workers to manually search for information, which is time-consuming and labor-intensive and inefficient. In addition, it is difficult to respond appropriately while taking into account the emotions and fatigue levels of workers, which can affect the quality of work. This poses the problem of reduced work efficiency and an increased risk of errors.

[0330] 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 a voice input means, a voice recognition means for converting voice input data into character data, an analysis means for analyzing the character data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into voice, an output means for playing back the voice generated by the voice synthesis means, an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the emotion, and a means for providing voice instructions in cooperation with the device to support work in the factory. As a result, when a worker asks a question by voice, appropriate information can be provided in real time, and a response can be made that takes into account the worker's emotion and fatigue level.

[0331] "Voice input means" refers to a device or function that captures the user's voice.

[0332] "Speech recognition means" refers to software or a device that converts voice data captured by a voice input means into text data.

[0333] "Generative AI" is a type of AI that analyzes text data and generates appropriate answers.

[0334] The "analysis means" is a function or device that uses generative artificial intelligence to analyze character data and generate answers to user questions.

[0335] The "voice synthesis means" is software or a device that converts the answer generated by the analysis means into voice.

[0336] The "output means" is a device or function for reproducing the voice generated by the voice synthesis means to the user.

[0337] "Emotion recognition means" refers to a device or function that recognizes the user's emotions and adjusts the content of the response based on those emotions.

[0338] "Devices for supporting work within a factory" refers to systems and devices that are intended to support work within a factory.

[0339] The present invention is a system that responds to factory workers' voice questions by providing appropriate information in real time and taking into consideration the worker's emotions and fatigue level. The system includes a voice input means, a voice recognition means, an analysis means using generative artificial intelligence, a voice synthesis means, an output means, an emotion recognition means, and a means for linking with devices for supporting factory work.

[0340] Program processing explanation

[0341] 1. Voice input

[0342] User: A worker in a factory asks the robot, "What should I do next?"

[0343] Robot: Captures the worker's voice through a built-in microphone.

[0344] 2. Voice Recognition

[0345] Robot: Sends captured audio data to the server.

[0346] Server: Uses a speech recognition method (for example, Google Speech-to-Text API) to convert the received voice data into text data.

[0347] 3. Emotion recognition

[0348] Server: The emotion recognition means receives the text data and analyzes the worker's emotions. For example, it recognizes that the worker is tired.

[0349] Server: Provides the recognized emotion information to the analysis means.

[0350] 4. Question Analysis and Answer Generation

[0351] Server: Using generative artificial intelligence (e.g., GPT-4), it analyzes the received text data and emotional information, understands the intent of the question, and then generates an appropriate response.

[0352] Example: Generate the answer "The next thing to do is check the tool kit."

[0353] 5. Adjusting your answers

[0354] Server: Based on the recognized emotional information, the server adjusts the response appropriately, for example adding an instruction to a tired worker saying, "Take it easy and take a short break."

[0355] 6. Speech Synthesis

[0356] Server: The generated answer text is sent to a speech synthesis means (e.g., Google Text-to-Speech API) and converted into voice data.

[0357] 7. Audio Output

[0358] Robot: The generated voice data is played back through the robot's speaker to provide a response to the worker.

[0359] Specific examples

[0360] Scenario 1: Dealing with fatigued workers

[0361] User Question: "What do I do next?"

[0362] Voice input capture: The robot captures the voice and sends it to the server.

[0363] Server speech recognition processing: Converts the speech into text data such as "What should I do next?"

[0364] Emotion recognition using an emotion engine: Recognizing that a worker is tired from text data.

[0365] Analysis and answer generation using generative artificial intelligence: The answer generated is, "The next thing you should do is check your tool kit."

[0366] Response adjustment: For tired workers, add the instruction, "Take it easy and take a short break."

[0367] Speech synthesis: Converting the generated text into speech.

[0368] Voice output: Provides answers to the worker through the robot's speaker.

[0369] Prompt Sentence Examples

[0370] When asked, "What's next?"

[0371] Recognize when a worker is tired and take that into account when generating text that answers the question "What task needs to be done next?"

[0372] example:

[0373] Question: "What should I do next?"

[0374] (Recognize fatigue): The next thing to do is check your tool kit. Take a short break and don't push yourself too hard.

[0375] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0376] Step 1:

[0377] Voice Input Processing

[0378] Subject: User

[0379] Specific operation: A worker in a factory asks the robot, "What should I do next?"

[0380] Input: Worker's voice

[0381] Data processing or data calculation: Audio is captured by a microphone.

[0382] Output: Captured audio data

[0383] Step 2:

[0384] Sending audio data

[0385] Subject: Terminal (robot)

[0386] Specific operation: The robot sends the captured voice data to the server.

[0387] Input: Audio data

[0388] Data processing or data calculation: The voice data is packetized and sent to the server via the network.

[0389] Output: Audio data sent to the server

[0390] Step 3:

[0391] Speech Recognition Processing

[0392] Subject: Server

[0393] Specific operation: The server converts the voice data into text data using a voice recognition method (for example, Google Speech-to-Text API).

[0394] Input: Audio data

[0395] Data processing or data calculation: Analyzing voice data and converting it into corresponding text data.

[0396] Output: Converted character data

[0397] Step 4:

[0398] Emotion Recognition Processing

[0399] Subject: Server

[0400] Specific operation: The server uses emotion recognition means to analyze the worker's emotions from the text data.

[0401] Input: Character data

[0402] Data processing or data calculation: Analyze emotions using natural language processing techniques and generate emotional information.

[0403] Output: Emotional information

[0404] Step 5:

[0405] Question analysis and answer generation processing

[0406] Subject: Server

[0407] Specific operation: The server uses generative artificial intelligence (e.g., GPT-4) to analyze text data and emotional information and generate an appropriate response.

[0408] Input: Text data and emotion information

[0409] Data processing or data computation: Generate prompts for a generative artificial intelligence model, which then analyzes and generates an answer.

[0410] Output: Generated answer text

[0411] Step 6:

[0412] Response adjustment process

[0413] Subject: Server

[0414] Specific operation: The server adjusts the generated answer appropriately based on the recognized emotion information.

[0415] Input: Generated answer text and sentiment information

[0416] Data processing or data calculation: Refer to the emotional information and add supplements or corrections to the answer text.

[0417] Output: Adjusted answer text

[0418] Step 7:

[0419] Speech synthesis processing

[0420] Subject: Server

[0421] Specific operation: The server converts the adjusted answer text into audio data using a speech synthesis means (e.g., Google Text-to-Speech API).

[0422] Input: Adjusted answer text

[0423] Data processing or data calculation: Text data is input into a speech synthesis engine to generate corresponding speech data.

[0424] Output: Generated audio data

[0425] Step 8:

[0426] Sending audio data

[0427] Subject: Server

[0428] Specific operation: The server sends the generated voice data to the robot.

[0429] Input: Generated audio data

[0430] Data processing or data calculation: The voice data is packetized and sent to the robot via the network.

[0431] Output: Audio data sent to the robot

[0432] Step 9:

[0433] Audio Output Processing

[0434] Subject: Terminal (robot)

[0435] Specific operation: The robot plays back the audio data through a speaker and provides a response to the worker.

[0436] Input: Transmitted audio data

[0437] Data processing or data calculation: Deserializing the audio data and converting it into a format that can be played by the speaker.

[0438] Output: Audio response that can be heard by the worker

[0439] 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.

[0440] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0441] 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.

[0442] [Second embodiment]

[0443] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0444] 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.

[0445] 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).

[0446] 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.

[0447] 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.

[0448] 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).

[0449] 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.

[0450] 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.

[0451] 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.

[0452] 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.

[0453] 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.

[0454] 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."

[0455] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[0456] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server converts the voice into text data using voice recognition means. Then, analysis means using generative artificial intelligence analyzes the text data and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by voice synthesis means, and finally provided to the user through output means (e.g., a speaker).

[0457] Below, the specific program processing will be explained in natural language.

[0458] Program processing explanation

[0459] 1. Voice input

[0460] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0461] Device: Captures the user's voice through a microphone.

[0462] 2. Voice Recognition

[0463] Device: Sends captured audio data to the server.

[0464] Server: Uses speech recognition means to convert the received voice data into text data.

[0465] Server: For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0466] 3. Question Analysis and Answer Generation

[0467] Server: Analyzes the received text data using generative AI. The analysis means understands the meaning of the question and searches for information from a related product knowledge database.

[0468] Server: For example, generate information such as "New smartphone plans start at 5,000 yen per month" as an answer to a question.

[0469] 4. Speech Synthesis

[0470] Server: Sends the generated answer text to the speech synthesis means.

[0471] Server: The speech synthesis means converts the text data into speech data, generating speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0472] 5. Audio Output

[0473] Server: Sends the generated audio data to the device.

[0474] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to questions.

[0475] Specific examples

[0476] User Question: What is the pricing plan for my new smartphone?

[0477] Capture audio input: The device captures audio and sends it to the server.

[0478] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0479] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[0480] Speech synthesis: Converting the generated text into speech.

[0481] Audio output: Provides answers to the user through the device's speaker.

[0482] In this way, the system based on the present invention can provide a prompt and accurate answer to a user's voice question. Furthermore, by linking with an intent recognition means and a product knowledge database, it can generate even more appropriate answers, improving the user experience. This system is particularly useful in situations where detailed information about a product is to be provided, enabling efficient customer service.

[0483] The processing flow will be explained below.

[0484] Step 1:

[0485] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0486] Step 2:

[0487] Device: Captures the user's voice through a microphone.

[0488] Step 3:

[0489] Device: Sends captured audio data to the server.

[0490] Step 4:

[0491] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0492] Step 5:

[0493] Server: Analyzes the received text data using generative AI. The analysis means analyzes the text data and understands the intent of the question.

[0494] Step 6:

[0495] Server: Searches for information corresponding to the question from a product knowledge database. For example, retrieves information on "new smartphone pricing plans" from the database.

[0496] Step 7:

[0497] Server: Generates an answer to the question based on the search results. For example, it generates the text "New smartphone plans start at 5,000 yen per month" as an answer to the user's question.

[0498] Step 8:

[0499] Server: Sends the generated answer text to the speech synthesis means.

[0500] Step 9:

[0501] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0502] Step 10:

[0503] Server: Sends the generated audio data to the device.

[0504] Step 11:

[0505] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0506] This series of processes allows the user to ask a question by voice and quickly receive an appropriate answer by voice.

[0507] Example 1

[0508] 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."

[0509] Modern information provision systems require a smooth process for users to ask questions by voice and receive appropriate answers. However, existing systems often have problems with speech recognition accuracy and answer generation and reproduction. In particular, they can be difficult to provide accurate and appropriate answers in real time. They also lack the ability to accurately understand user intent and generate the optimal answer. This can lead to a poor user experience.

[0510] 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.

[0511] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence to generate an answer, a voice synthesis means for converting the generated answer into speech, a means for transmitting the generated speech data to an output means, and an output means for playing back the speech generated by the voice synthesis means. This allows a user to obtain a highly accurate answer in real time when asking a question by voice. Furthermore, by linking with an intent recognition means and a product knowledge database, it is possible to generate more appropriate answers and improve the user experience.

[0512] "Voice input means" refers to a device or software that captures a user's speech or voice data and records it as digital data.

[0513] "Speech recognition means" refers to technology that analyzes voice data and converts the voice into corresponding text data.

[0514] "Generative AI" refers to an AI technology that analyzes given text or data and creates optimal answers or products based on that content.

[0515] "Analysis means" refers to a process or device that analyzes received text data, understands the user's intent, and generates an appropriate response.

[0516] "Speech synthesis means" refers to technology that converts text data into voice data.

[0517] "Output means" refers to a device or system for playing back the generated audio data.

[0518] "Intention recognition means" refers to the technology or process for inferring a user's intentions or desires from the data entered by the user.

[0519] A "product knowledge database" refers to a database that stores detailed information such as the features, prices, and specifications of specific products.

[0520] "Search methods" refer to the techniques and processes used to search for and retrieve data or information based on specific criteria.

[0521] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[0522] When a user asks a question, their voice is captured through a voice input means (a microphone or the built-in microphone of a smart device). This voice data is sent to a server by the device. For example, when using a smartphone or tablet, the voice data is sent via the Internet.

[0523] The server uses a speech recognition tool to convert the received voice data into text data. For example, Google Cloud Speech-to-Text API can be used. The voice data is converted into text data such as "Please tell me the price plan for my new smartphone."

[0524] Next, the text data is analyzed using generative artificial intelligence (e.g., OpenAI GPT-4). This analysis method understands the meaning of the user's question and generates an answer based on related information. Specifically, based on the context and keywords of the question, an answer appropriate to the question is searched from a product knowledge database and an appropriate answer text is generated.

[0525] The server sends the generated response text to a speech synthesis means (e.g., Amazon Polly), which converts the text data into speech data. The speech data generated is, "The new smartphone plan starts at 5,000 yen per month."

[0526] Finally, the generated voice data is sent to the terminal and played back through the terminal's output means (speaker or headset), allowing the user to receive the answer to their question by voice.

[0527] For example, if a user asks, "What is the pricing plan for my new smartphone?", the following prompt is input to the generative AI:

[0528] Example prompt sentence:

[0529] User Question: What is the pricing plan for my new smartphone?

[0530] Generative AI answer: New smartphone plans start at 5,000 yen per month.

[0531] Based on this prompt, the generative artificial intelligence generates an appropriate response, which is then converted into voice data by a speech synthesis means and finally provided to the user. This system allows users to receive fast and accurate information by voice. It is particularly useful in situations where detailed information about products is to be provided, and realizes efficient customer service.

[0532] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0533] Step 1: Voice Input

[0534] User: Asks the device, "What is the price plan for my new smartphone?" The user's input is voice data.

[0535] Device: Uses a built-in microphone to capture the user's voice, which is then stored digitally.

[0536] Output: The captured audio data is generated.

[0537] Step 2: Sending audio data

[0538] On your device: The captured audio data is sent to the server using the HTTPS protocol, which sends the data securely over your internet connection.

[0539] Input: Audio data

[0540] Output: Audio data sent to the server

[0541] Step 3: Voice Recognition

[0542] Server: Calls the API of the speech recognition method (for example, Google Cloud Speech-to-Text API) and converts the received voice data into text data.

[0543] Input: Audio data sent to the server

[0544] Data processing: Converting audio data into text using speech recognition tools. This process involves analyzing the audio data and generating a corresponding string of characters.

[0545] Output: Converted text data (e.g., "What is the price plan for my new smartphone?")

[0546] Step 4: Parsing the question and generating an answer

[0547] Server: Input the text data into a generative artificial intelligence (e.g., OpenAI GPT-4) and analyze it.

[0548] Input: Converted character data

[0549] Data calculation: The generative AI model generates appropriate answers to questions based on prompts, searching for information from a product knowledge database based on the context and keywords of the question.

[0550] Output: Generated answer text (e.g., "New smartphone plans start at 5,000 yen per month.")

[0551] Step 5: Text-to-speech synthesis of the answer text

[0552] Server: The generated answer text is input into a speech synthesis means (e.g., Amazon Polly) and converted into voice data.

[0553] Input: Generated answer text

[0554] Data processing: Converting text into audio data using speech synthesis tools.

[0555] Output: Generated speech data (e.g., "New smartphone plans start at 5,000 yen per month.")

[0556] Step 6: Sending audio data

[0557] Server: Sends the generated audio data to the device using the HTTP / HTTPS protocol to ensure secure communication.

[0558] Input: Generated audio data

[0559] Output: Audio data sent to the device

[0560] Step 7: Audio Output

[0561] Terminal: Passes the audio data received by the terminal to the output means (speaker or headset).

[0562] Input: Audio data sent to the device

[0563] What it does: Plays audio data through a speaker and provides answers to the user.

[0564] Output: A spoken response to the user (e.g., "New smartphone plans start at ¥5,000 per month.")

[0565] By going through the above processing steps, when a user asks a question by voice, the user can quickly and accurately receive a response by voice.

[0566] (Application example 1)

[0567] 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."

[0568] Conventional voice recognition systems and automated response systems focus only on simple voice input and conversion to text data, and lack the ability to respond to detailed user needs and complex questions. Furthermore, especially in new digital markets such as virtual stores, there is a growing need for systems that allow users to quickly and accurately obtain product information and engage in real-time voice interaction. Therefore, it is necessary to provide a new system that meets these needs and improves the user experience.

[0569] 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.

[0570] In this invention, the server includes a speech recognition unit that converts voice input data into text data, an analysis unit that uses generative artificial intelligence to analyze the text data and generate a response, and a speech synthesis unit that converts the generated response into speech. This allows the server to capture voice data when a user asks a question by voice in a virtual store, execute processing using cloud computing technology, and provide the user with a quick and accurate response. Furthermore, by obtaining information from a product knowledge database and accurately recognizing the user's intent, the server can generate a more appropriate response.

[0571] "Audio input means" refers to a device and method for capturing user-uttered speech as a digital signal.

[0572] "Speech recognition means" refers to the technology and devices for converting voice input data into text data.

[0573] "Analysis means" refers to technology and devices that have the function of analyzing received text data using generative artificial intelligence and generating answers to users' questions.

[0574] "Speech synthesis means" refers to the technology and device that converts the generated response text into voice data.

[0575] "Output means" refers to a device that reproduces the voice generated by the voice synthesis means.

[0576] "Cloud computing technology" refers to technology that performs distributed processing over the Internet and efficiently executes large-scale data analysis and calculations.

[0577] A "virtual store" refers to a virtual shopping space built on the Internet where users can browse and purchase products.

[0578] "Apparatus" means a machine or mechanism designed for a specific purpose.

[0579] "Generative AI" refers to artificial intelligence technology that has the ability to learn large amounts of data and generate new information and answers based on human language and knowledge.

[0580] A "product knowledge database" refers to a database that stores detailed information related to products.

[0581] The present invention relates to a system in which a user can ask a question by voice and a response to the question is provided by voice, and the system uses a voice input means, a voice recognition means, an analysis means, a voice synthesis means, an output means, cloud computing technology, and a voice data capture device in a virtual store.

[0582] When a user asks a question through a voice input device in the virtual store, the voice is captured as a digital signal through a microphone. This digital signal is voice input data, and the terminal transmits the voice input data to a cloud server.

[0583] The cloud server uses the Google Cloud Speech-to-Text API to convert this voice input data into text data. This text data represents the user's question. For example, if a user asks, "Tell me about this product," the cloud server converts this into text data: "Tell me about this product."

[0584] Next, this text data is analyzed using generative artificial intelligence. OpenAI's GPT-3.5 is used as the analysis method. The cloud server uses the analysis method to analyze the text data and generate the optimal answer to the user's question. The analysis method understands the meaning of the question and extracts the necessary information from the product knowledge database.

[0585] The answer text generated by the analysis means contains specific information such as "This product is the latest model, priced at 50,000 yen, and in stock." This answer text is converted into audio data using the Google Cloud Text-to-Speech API. This audio data is provided in an audio format that is easy for the user to understand.

[0586] Finally, the generated voice data is played back by a voice synthesis means, and the speaker of the smart glasses or head-mounted display is used as an output means, allowing the user to receive a voice response to their question.

[0587] Specific examples are shown below.

[0588] When a user asks, "Tell me about this product," the smart glasses' microphone captures the voice and sends it to a cloud server. The cloud server converts the text data using the Google Cloud Speech-to-Text API, analyzes it using GPT-3.5, and generates an answer. The text data is then converted back into audio using the Google Cloud Text-to-Speech API, and the audio is played through the smart glasses' speaker.

[0589] An example of a prompt is as follows:

[0590] User Asks: "Tell me about this product"

[0591] Data to send to Google Cloud Speech-to-Text API: Audio data

[0592] Prompt GPT-3.5: "The user says, 'Tell me about this product.' Extract information from the corresponding product database and generate an appropriate answer."

[0593] Data to send to Google Cloud Text-to-Speech API: Generated response text

[0594] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0595] Step 1:

[0596] Voice input

[0597] A user asks a question to a terminal in a virtual store, such as "Tell me about this product." The terminal uses a voice input means to capture the user's voice as a digital signal. This captured voice data is the input. This input data is ready to be sent to the subsequent processing steps.

[0598] Step 2:

[0599] Voice Recognition

[0600] The device sends the captured voice data to a cloud server, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. During this conversion process, the voice waveform is analyzed and converted into the corresponding text, "Tell me about this product." This text data is the output. The phonemes of the voice are analyzed, and the most appropriate text is generated using a language model.

[0601] Step 3:

[0602] Question analysis and answer generation

[0603] The cloud server analyzes the text data using generative artificial intelligence (OpenAI's GPT-3.5). This analysis method understands the user's question and extracts appropriate information from a product knowledge database. The input is the converted text data "Tell me about this product," and the output is the answer text generated based on the analysis: "This product is the latest model, priced at 50,000 yen. In stock." The generative AI model analyzes the question and generates relevant information by filtering it from the database.

[0604] Step 4:

[0605] Speech synthesis

[0606] The cloud server sends the generated answer text to the Google Cloud Text-to-Speech API, which converts it into audio data. The input is the answer text, and the output is the corresponding audio data. In this process, the answer text is synthesized into natural-sounding audio and provided in a format that is easy for the user to understand. The generated audio data is prepared here.

[0607] Step 5:

[0608] Audio Output

[0609] The cloud server sends the generated voice data to the device. The device uses an output means (speakers in smart glasses or a head-mounted display) to play this voice to the user. The input is the generated voice data, and the output is a voice response played to the user. The device outputs the voice data through a designated speaker and provides it to the user.

[0610] These steps enable the system to quickly and accurately provide a spoken answer to a user's spoken question.

[0611] 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.

[0612] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. This system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit. It also includes a configuration that further combines an emotion engine that recognizes the user's emotions and provides appropriate answers based on those emotions.

[0613] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server then converts the voice into text data using a voice recognition means. The server then uses an emotion engine to recognize the user's emotion from the text data and provides the emotion information to the analysis means. Next, the analysis means using generative artificial intelligence analyzes the text data and emotion information and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by a voice synthesis means and finally provided to the user through an output means (e.g., a speaker).

[0614] Below, the specific program processing will be explained in natural language.

[0615] Program processing explanation

[0616] 1. Voice input

[0617] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0618] Device: Captures the user's voice through a microphone.

[0619] 2. Voice Recognition

[0620] Device: Sends captured audio data to the server.

[0621] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0622] 3. Emotion recognition

[0623] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it recognizes whether the user is excited or confused from a text such as "New smartphone pricing plan."

[0624] Server: Provides the recognized emotion information to the analysis means.

[0625] 4. Question Analysis and Answer Generation

[0626] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[0627] 5. Adjusting your answers

[0628] Server: Adjust the answer appropriately based on the perceived emotion information, for example adding more detailed explanation if the user is confused.

[0629] 6. Speech Synthesis

[0630] Server: Sends the generated answer text to the speech synthesis means.

[0631] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0632] 7. Audio Output

[0633] Server: Sends the generated audio data to the device.

[0634] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0635] Specific examples

[0636] User Question: What is the pricing plan for my new smartphone?

[0637] Capture audio input: The device captures audio and sends it to the server.

[0638] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0639] Emotion recognition using an emotion engine: Recognizes when a user is confused from text data.

[0640] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[0641] Tailor your response: For confused users, add an additional "Would you like more options?"

[0642] Speech synthesis: Converting the generated text into speech.

[0643] Audio output: Provides answers to the user through the device's speaker.

[0644] In this way, the system according to the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is particularly useful in situations where detailed information about products is provided, realizing efficient customer service.

[0645] The processing flow will be explained below.

[0646] Step 1:

[0647] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0648] Step 2:

[0649] Device: Captures the user's voice through a microphone.

[0650] Step 3:

[0651] Device: Sends captured audio data to the server.

[0652] Step 4:

[0653] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0654] Step 5:

[0655] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it analyzes the text data and determines whether the user is interested, confused, or in a hurry.

[0656] Step 6:

[0657] Server: Provides the emotion information analyzed by the emotion engine to the analysis means.

[0658] Step 7:

[0659] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[0660] Step 8:

[0661] Server: Adjust the answer appropriately based on the perceived emotion information. For example, if the user is confused, add an additional explanation: "Would you like more options?"

[0662] Step 9:

[0663] Server: Sends the generated answer text to the speech synthesis means.

[0664] Step 10:

[0665] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0666] Step 11:

[0667] Server: Sends the generated audio data to the device.

[0668] Step 12:

[0669] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0670] This specific process allows the user to ask a question by voice and not only receive an appropriate answer quickly by voice, but also obtain a more friendly and understandable answer based on emotion recognition by the emotion engine.

[0671] Example 2

[0672] 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."

[0673] Conventional voice response systems can automatically generate and provide voice responses to user questions, but they have the problem of being unable to recognize the user's emotions and provide appropriate responses based on those emotions. This can lead to a poor user experience, especially when the user is confused or expecting something, as the system is unable to respond in accordance with that emotion. Furthermore, the responses often do not match the user's intentions, making it necessary to improve user satisfaction.

[0674] 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.

[0675] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into speech, an output means for playing back the generated speech, and an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the recognized emotion. This makes it possible to provide an appropriate answer according to the user's emotion, thereby improving the user experience and increasing user satisfaction. Furthermore, by using an intent recognition means, the server can accurately understand the user's intent and generate a more accurate answer.

[0676] "Audio input means" is a device or software for capturing speech produced by a user and transmitting that data to a processing system.

[0677] "Speech recognition means" means technology or equipment for converting voice data into text data, including algorithms and services used for that purpose.

[0678] "Generative AI" is an AI model used to analyze given input data and generate answers in a natural language format.

[0679] "Analysis means" refers to technology or devices that use generative artificial intelligence to analyze input data, understand the intent of the question, and generate an appropriate answer.

[0680] "Speech synthesis means" refers to a technique or device for converting generated character data into voice data.

[0681] "Output means" refers to hardware or software for reproducing audio data and providing information to the user by audio.

[0682] An "emotion recognition means" is a technology or device that identifies emotions from user input data and appropriately adjusts the system's response based on those emotions.

[0683] "Intention recognition means" refers to technology or devices that infer a user's intention from input data and generate appropriate analysis and responses.

[0684] A "product knowledge database" is a database that stores detailed information about products and serves as the basis for the system to search for information and generate answers.

[0685] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, an output unit, and an emotion recognition unit.

[0686] Specifically, when a user speaks a question through the microphone of a device on which the software is installed, the voice input means receives the voice and sends it to the server. For example, consider a situation where a user asks, "What is the price plan for my new smartphone?" The device's microphone captures the voice and transfers the data to the server. The server then uses the voice recognition means to convert the voice data into text data. This process can be performed using voice recognition services such as Google Cloud Speech-to-Text or Amazon Transcribe.

[0687] The emotion recognition means then analyzes the text data and recognizes the user's emotions. For example, it can recognize whether the user is excited or confused from the text "New smartphone pricing plan." This process can be performed using IBM Watson's Tone Analyzer, among other tools. The recognized emotion information is provided to the analysis means.

[0688] Next, an analysis means using generative AI (e.g., GPT-4) analyzes the text data and emotional information to generate an appropriate answer to the user's question. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month." This process is performed using a generative AI model.

[0689] The generated answer is sent to a speech synthesis means and converted into voice data. This voice synthesis can be performed using speech synthesis services such as Amazon Polly or Google Text-to-Speech. Finally, the generated voice data is sent to the device and played back to the user through the speaker.

[0690] The answer is also adjusted based on the emotion information recognized by the emotion recognition means. For example, if the user is confused, an explanation such as "Would you like more options?" is added to the answer.

[0691] Specific examples

[0692] User asks: "What plan is available for my new smartphone?"

[0693] Device audio input: Captures audio through the microphone and sends it to the server.

[0694] Server speech recognition: Converts speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0695] Emotion recognition: Analyzes text data to recognize when a user is confused.

[0696] Analysis and Answer Generation: Generate the answer "New smartphone plans start at 5,000 yen per month."

[0697] Tailor your response: For confused users, add the explanation "Would you like more options?"

[0698] Speech synthesis: Converting the generated text into audio data.

[0699] Speech output: Provides answers to the user through a speaker.

[0700] In this way, the system based on the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is useful in situations where detailed information about products is provided, realizing efficient customer service.

[0701] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0702] Step 1:

[0703] Voice input

[0704] User: The user asks the device, "What is the pricing plan for my new smartphone?"

[0705] On the device: The built-in microphone is used to capture the user's voice, and the voice data is generated and temporarily stored in the device's memory.

[0706] Step 2:

[0707] Sending audio data

[0708] Terminal: The captured audio data is compressed and sent to the server via the network using a transfer protocol such as HTTP or HTTPS.

[0709] Input: User's voice data

[0710] Output: Audio data sent to the server

[0711] Step 3:

[0712] Voice Recognition

[0713] Server: Passes the received voice data to the voice recognition means.

[0714] Server: Converts voice data into text data using a speech recognition method. Calls a speech recognition API (e.g., Google Cloud Speech-to-Text) and converts voice data into text data.

[0715] Input: Audio data

[0716] Output: The result of converting speech to text data, e.g. "What is the price plan for my new smartphone?"

[0717] Step 4:

[0718] emotion recognition

[0719] Server: Passes the text data to the emotion recognition means.

[0720] Server: Uses emotion recognition means to recognize the user's emotion from the text data. Calls an emotion recognition API (e.g., IBM Watson Tone Analyzer) to obtain emotion information from the text data.

[0721] Input: Character data

[0722] Output: Emotional information such as user confusion or expectation

[0723] Step 5:

[0724] Question analysis and answer generation

[0725] Server: Passes text data and emotion information to the analysis means.

[0726] Server: Uses generative AI (e.g., GPT-4) to analyze text data and emotional information and generate appropriate answers. Enter prompt sentences into the generative AI model to generate answers.

[0727] Input: Text data, emotion information

[0728] Output: Answer text "New smartphone plans start at 5,000 yen per month."

[0729] Step 6:

[0730] Adjusting your answers

[0731] Server: Adjust the generated answer based on sentiment information, for example, adding additional explanation to the answer if the user is confused.

[0732] Input: Answer text, emotion information

[0733] Output: Tailored response text, e.g. "New smartphone plans start at ¥5,000 / month. Would you like more options?"

[0734] Step 7:

[0735] Speech synthesis

[0736] Server: Pass the adjusted answer text to the speech synthesis means.

[0737] Server: Convert the response text into audio data using a speech synthesis tool. Call a speech synthesis API (e.g., Amazon Polly).

[0738] Input: Adjusted answer text

[0739] Output: Voice data, for example, "New smartphone plans start at 5,000 yen per month. Would you like more options?"

[0740] Step 8:

[0741] Audio Output

[0742] Server: Sends the generated audio data to the device.

[0743] Terminal: The generated audio data is played back through a speaker and provided to the user.

[0744] Input: Audio data

[0745] Output: The answer provided to the user verbally

[0746] (Application example 2)

[0747] 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."

[0748] In factories, it is important for workers to quickly and accurately obtain the information they need. However, conventional methods require workers to manually search for information, which is time-consuming and labor-intensive and inefficient. In addition, it is difficult to respond appropriately while taking into account the emotions and fatigue levels of workers, which can affect the quality of work. This poses the problem of reduced work efficiency and an increased risk of errors.

[0749] 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 a voice input means, a voice recognition means for converting voice input data into character data, an analysis means for analyzing the character data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into voice, an output means for playing back the voice generated by the voice synthesis means, an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the emotion, and a means for providing voice instructions in cooperation with the device to support work in the factory. As a result, when a worker asks a question by voice, appropriate information can be provided in real time, and a response can be made that takes into account the worker's emotion and fatigue level.

[0750] "Voice input means" refers to a device or function that captures the user's voice.

[0751] "Speech recognition means" refers to software or a device that converts voice data captured by a voice input means into text data.

[0752] "Generative AI" is a type of AI that analyzes text data and generates appropriate answers.

[0753] The "analysis means" is a function or device that uses generative artificial intelligence to analyze character data and generate answers to user questions.

[0754] The "voice synthesis means" is software or a device that converts the answer generated by the analysis means into voice.

[0755] The "output means" is a device or function for reproducing the voice generated by the voice synthesis means to the user.

[0756] "Emotion recognition means" refers to a device or function that recognizes the user's emotions and adjusts the content of the response based on those emotions.

[0757] "Devices for supporting work within a factory" refers to systems and devices that are intended to support work within a factory.

[0758] The present invention is a system that responds to factory workers' voice questions by providing appropriate information in real time and taking into consideration the worker's emotions and fatigue level. The system includes a voice input means, a voice recognition means, an analysis means using generative artificial intelligence, a voice synthesis means, an output means, an emotion recognition means, and a means for linking with devices for supporting factory work.

[0759] Program processing explanation

[0760] 1. Voice input

[0761] User: A worker in a factory asks the robot, "What should I do next?"

[0762] Robot: Captures the worker's voice through a built-in microphone.

[0763] 2. Voice Recognition

[0764] Robot: Sends captured audio data to the server.

[0765] Server: Uses a speech recognition method (for example, Google Speech-to-Text API) to convert the received voice data into text data.

[0766] 3. Emotion recognition

[0767] Server: The emotion recognition means receives the text data and analyzes the worker's emotions. For example, it recognizes that the worker is tired.

[0768] Server: Provides the recognized emotion information to the analysis means.

[0769] 4. Question Analysis and Answer Generation

[0770] Server: Using generative artificial intelligence (e.g., GPT-4), it analyzes the received text data and emotional information, understands the intent of the question, and then generates an appropriate response.

[0771] Example: Generate the answer "The next thing to do is check the tool kit."

[0772] 5. Adjusting your answers

[0773] Server: Based on the recognized emotional information, the server adjusts the response appropriately, for example adding an instruction to a tired worker saying, "Take it easy and take a short break."

[0774] 6. Speech Synthesis

[0775] Server: The generated answer text is sent to a speech synthesis means (e.g., Google Text-to-Speech API) and converted into voice data.

[0776] 7. Audio Output

[0777] Robot: The generated voice data is played back through the robot's speaker to provide a response to the worker.

[0778] Specific examples

[0779] Scenario 1: Dealing with fatigued workers

[0780] User Question: "What do I do next?"

[0781] Voice input capture: The robot captures the voice and sends it to the server.

[0782] Server speech recognition processing: Converts the speech into text data such as "What should I do next?"

[0783] Emotion recognition using an emotion engine: Recognizing that a worker is tired from text data.

[0784] Analysis and answer generation using generative artificial intelligence: The answer generated is, "The next thing you should do is check your tool kit."

[0785] Response adjustment: For tired workers, add the instruction, "Take it easy and take a short break."

[0786] Speech synthesis: Converting the generated text into speech.

[0787] Voice output: Provides answers to the worker through the robot's speaker.

[0788] Prompt Sentence Examples

[0789] When asked, "What's next?"

[0790] Recognize when a worker is tired and take that into account when generating text that answers the question "What task needs to be done next?"

[0791] example:

[0792] Question: "What should I do next?"

[0793] (Recognize fatigue): The next thing to do is check your tool kit. Take a short break and don't push yourself too hard.

[0794] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0795] Step 1:

[0796] Voice Input Processing

[0797] Subject: User

[0798] Specific operation: A worker in a factory asks the robot, "What should I do next?"

[0799] Input: Worker's voice

[0800] Data processing or data calculation: Audio is captured by a microphone.

[0801] Output: Captured audio data

[0802] Step 2:

[0803] Sending audio data

[0804] Subject: Terminal (robot)

[0805] Specific operation: The robot sends the captured voice data to the server.

[0806] Input: Audio data

[0807] Data processing or data calculation: The voice data is packetized and sent to the server via the network.

[0808] Output: Audio data sent to the server

[0809] Step 3:

[0810] Speech Recognition Processing

[0811] Subject: Server

[0812] Specific operation: The server converts the voice data into text data using a voice recognition method (for example, Google Speech-to-Text API).

[0813] Input: Audio data

[0814] Data processing or data calculation: Analyzing voice data and converting it into corresponding text data.

[0815] Output: Converted character data

[0816] Step 4:

[0817] Emotion Recognition Processing

[0818] Subject: Server

[0819] Specific operation: The server uses emotion recognition means to analyze the worker's emotions from the text data.

[0820] Input: Character data

[0821] Data processing or data calculation: Analyze emotions using natural language processing techniques and generate emotional information.

[0822] Output: Emotional information

[0823] Step 5:

[0824] Question analysis and answer generation processing

[0825] Subject: Server

[0826] Specific operation: The server uses generative artificial intelligence (e.g., GPT-4) to analyze text data and emotional information and generate an appropriate response.

[0827] Input: Text data and emotion information

[0828] Data processing or data computation: Generate prompts for a generative artificial intelligence model, which then analyzes and generates an answer.

[0829] Output: Generated answer text

[0830] Step 6:

[0831] Response adjustment process

[0832] Subject: Server

[0833] Specific operation: The server adjusts the generated answer appropriately based on the recognized emotion information.

[0834] Input: Generated answer text and sentiment information

[0835] Data processing or data calculation: Refer to the emotional information and add supplements or corrections to the answer text.

[0836] Output: Adjusted answer text

[0837] Step 7:

[0838] Speech synthesis processing

[0839] Subject: Server

[0840] Specific operation: The server converts the adjusted answer text into audio data using a speech synthesis means (e.g., Google Text-to-Speech API).

[0841] Input: Adjusted answer text

[0842] Data processing or data calculation: Text data is input into a speech synthesis engine to generate corresponding speech data.

[0843] Output: Generated audio data

[0844] Step 8:

[0845] Sending audio data

[0846] Subject: Server

[0847] Specific operation: The server sends the generated voice data to the robot.

[0848] Input: Generated audio data

[0849] Data processing or data calculation: The voice data is packetized and sent to the robot via the network.

[0850] Output: Audio data sent to the robot

[0851] Step 9:

[0852] Audio Output Processing

[0853] Subject: Terminal (robot)

[0854] Specific operation: The robot plays back the audio data through a speaker and provides a response to the worker.

[0855] Input: Transmitted audio data

[0856] Data processing or data calculation: Deserializing the audio data and converting it into a format that can be played by the speaker.

[0857] Output: Audio response that can be heard by the worker

[0858] 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.

[0859] 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.

[0860] 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.

[0861] [Third embodiment]

[0862] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0863] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0864] 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).

[0865] 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.

[0866] 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.

[0867] 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).

[0868] 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.

[0869] 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.

[0870] 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.

[0871] 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.

[0872] 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.

[0873] 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."

[0874] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[0875] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server converts the voice into text data using voice recognition means. Then, analysis means using generative artificial intelligence analyzes the text data and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by voice synthesis means, and finally provided to the user through output means (e.g., a speaker).

[0876] Below, the specific program processing will be explained in natural language.

[0877] Program processing explanation

[0878] 1. Voice input

[0879] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0880] Device: Captures the user's voice through a microphone.

[0881] 2. Voice Recognition

[0882] Device: Sends captured audio data to the server.

[0883] Server: Uses speech recognition means to convert the received voice data into text data.

[0884] Server: For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0885] 3. Question Analysis and Answer Generation

[0886] Server: Analyzes the received text data using generative AI. The analysis means understands the meaning of the question and searches for information from a related product knowledge database.

[0887] Server: For example, generate information such as "New smartphone plans start at 5,000 yen per month" as an answer to a question.

[0888] 4. Speech Synthesis

[0889] Server: Sends the generated answer text to the speech synthesis means.

[0890] Server: The speech synthesis means converts the text data into speech data, generating speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0891] 5. Audio Output

[0892] Server: Sends the generated audio data to the device.

[0893] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to questions.

[0894] Specific examples

[0895] User Question: What is the pricing plan for my new smartphone?

[0896] Capture audio input: The device captures audio and sends it to the server.

[0897] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[0898] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[0899] Speech synthesis: Converting the generated text into speech.

[0900] Audio output: Provides answers to the user through the device's speaker.

[0901] In this way, the system based on the present invention can provide a prompt and accurate answer to a user's voice question. Furthermore, by linking with an intent recognition means and a product knowledge database, it can generate even more appropriate answers, improving the user experience. This system is particularly useful in situations where detailed information about a product is to be provided, enabling efficient customer service.

[0902] The processing flow will be explained below.

[0903] Step 1:

[0904] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[0905] Step 2:

[0906] Device: Captures the user's voice through a microphone.

[0907] Step 3:

[0908] Device: Sends captured audio data to the server.

[0909] Step 4:

[0910] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[0911] Step 5:

[0912] Server: Analyzes the received text data using generative AI. The analysis means analyzes the text data and understands the intent of the question.

[0913] Step 6:

[0914] Server: Searches for information corresponding to the question from a product knowledge database. For example, retrieves information on "new smartphone pricing plans" from the database.

[0915] Step 7:

[0916] Server: Generates an answer to the question based on the search results. For example, it generates the text "New smartphone plans start at 5,000 yen per month" as an answer to the user's question.

[0917] Step 8:

[0918] Server: Sends the generated answer text to the speech synthesis means.

[0919] Step 9:

[0920] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[0921] Step 10:

[0922] Server: Sends the generated audio data to the device.

[0923] Step 11:

[0924] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[0925] This series of processes allows the user to ask a question by voice and quickly receive an appropriate answer by voice.

[0926] Example 1

[0927] 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."

[0928] Modern information provision systems require a smooth process for users to ask questions by voice and receive appropriate answers. However, existing systems often have problems with speech recognition accuracy and answer generation and reproduction. In particular, they can be difficult to provide accurate and appropriate answers in real time. They also lack the ability to accurately understand user intent and generate the optimal answer. This can lead to a poor user experience.

[0929] 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.

[0930] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence to generate an answer, a voice synthesis means for converting the generated answer into speech, a means for transmitting the generated speech data to an output means, and an output means for playing back the speech generated by the voice synthesis means. This allows a user to obtain a highly accurate answer in real time when asking a question by voice. Furthermore, by linking with an intent recognition means and a product knowledge database, it is possible to generate more appropriate answers and improve the user experience.

[0931] "Voice input means" refers to a device or software that captures a user's speech or voice data and records it as digital data.

[0932] "Speech recognition means" refers to technology that analyzes voice data and converts the voice into corresponding text data.

[0933] "Generative AI" refers to an AI technology that analyzes given text or data and creates optimal answers or products based on that content.

[0934] "Analysis means" refers to a process or device that analyzes received text data, understands the user's intent, and generates an appropriate response.

[0935] "Speech synthesis means" refers to technology that converts text data into voice data.

[0936] "Output means" refers to a device or system for playing back the generated audio data.

[0937] "Intention recognition means" refers to the technology or process for inferring a user's intentions or desires from the data entered by the user.

[0938] A "product knowledge database" refers to a database that stores detailed information such as the features, prices, and specifications of specific products.

[0939] "Search methods" refer to the techniques and processes used to search for and retrieve data or information based on specific criteria.

[0940] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[0941] When a user asks a question, their voice is captured through a voice input means (a microphone or the built-in microphone of a smart device). This voice data is sent to a server by the device. For example, when using a smartphone or tablet, the voice data is sent via the Internet.

[0942] The server uses a speech recognition tool to convert the received voice data into text data. For example, Google Cloud Speech-to-Text API can be used. The voice data is converted into text data such as "Please tell me the price plan for my new smartphone."

[0943] Next, the text data is analyzed using generative artificial intelligence (e.g., OpenAI GPT-4). This analysis method understands the meaning of the user's question and generates an answer based on related information. Specifically, based on the context and keywords of the question, an answer appropriate to the question is searched from a product knowledge database and an appropriate answer text is generated.

[0944] The server sends the generated response text to a speech synthesis means (e.g., Amazon Polly), which converts the text data into speech data. The speech data generated is, "The new smartphone plan starts at 5,000 yen per month."

[0945] Finally, the generated voice data is sent to the terminal and played back through the terminal's output means (speaker or headset), allowing the user to receive the answer to their question by voice.

[0946] For example, if a user asks, "What is the pricing plan for my new smartphone?", the following prompt is input to the generative AI:

[0947] Example prompt sentence:

[0948] User Question: What is the pricing plan for my new smartphone?

[0949] Generative AI answer: New smartphone plans start at 5,000 yen per month.

[0950] Based on this prompt, the generative artificial intelligence generates an appropriate response, which is then converted into voice data by a speech synthesis means and finally provided to the user. This system allows users to receive fast and accurate information by voice. It is particularly useful in situations where detailed information about products is to be provided, and realizes efficient customer service.

[0951] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0952] Step 1: Voice Input

[0953] User: Asks the device, "What is the price plan for my new smartphone?" The user's input is voice data.

[0954] Device: Uses a built-in microphone to capture the user's voice, which is then stored digitally.

[0955] Output: The captured audio data is generated.

[0956] Step 2: Sending audio data

[0957] On your device: The captured audio data is sent to the server using the HTTPS protocol, which sends the data securely over your internet connection.

[0958] Input: Audio data

[0959] Output: Audio data sent to the server

[0960] Step 3: Voice Recognition

[0961] Server: Calls the API of the speech recognition method (for example, Google Cloud Speech-to-Text API) and converts the received voice data into text data.

[0962] Input: Audio data sent to the server

[0963] Data processing: Converting audio data into text using speech recognition tools. This process involves analyzing the audio data and generating a corresponding string of characters.

[0964] Output: Converted text data (e.g., "What is the price plan for my new smartphone?")

[0965] Step 4: Parsing the question and generating an answer

[0966] Server: Input the text data into a generative artificial intelligence (e.g., OpenAI GPT-4) and analyze it.

[0967] Input: Converted character data

[0968] Data calculation: The generative AI model generates appropriate answers to questions based on prompts, searching for information from a product knowledge database based on the context and keywords of the question.

[0969] Output: Generated answer text (e.g., "New smartphone plans start at 5,000 yen per month.")

[0970] Step 5: Text-to-speech synthesis of the answer text

[0971] Server: The generated answer text is input into a speech synthesis means (e.g., Amazon Polly) and converted into voice data.

[0972] Input: Generated answer text

[0973] Data processing: Converting text into audio data using speech synthesis tools.

[0974] Output: Generated speech data (e.g., "New smartphone plans start at 5,000 yen per month.")

[0975] Step 6: Sending audio data

[0976] Server: Sends the generated audio data to the device using the HTTP / HTTPS protocol to ensure secure communication.

[0977] Input: Generated audio data

[0978] Output: Audio data sent to the device

[0979] Step 7: Audio Output

[0980] Terminal: Passes the audio data received by the terminal to the output means (speaker or headset).

[0981] Input: Audio data sent to the device

[0982] What it does: Plays audio data through a speaker and provides answers to the user.

[0983] Output: A spoken response to the user (e.g., "New smartphone plans start at ¥5,000 per month.")

[0984] By going through the above processing steps, when a user asks a question by voice, the user can quickly and accurately receive a response by voice.

[0985] (Application example 1)

[0986] 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."

[0987] Conventional voice recognition systems and automated response systems focus only on simple voice input and conversion to text data, and lack the ability to respond to detailed user needs and complex questions. Furthermore, especially in new digital markets such as virtual stores, there is a growing need for systems that allow users to quickly and accurately obtain product information and engage in real-time voice interaction. Therefore, it is necessary to provide a new system that meets these needs and improves the user experience.

[0988] 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.

[0989] In this invention, the server includes a speech recognition unit that converts voice input data into text data, an analysis unit that uses generative artificial intelligence to analyze the text data and generate a response, and a speech synthesis unit that converts the generated response into speech. This allows the server to capture voice data when a user asks a question by voice in a virtual store, execute processing using cloud computing technology, and provide the user with a quick and accurate response. Furthermore, by obtaining information from a product knowledge database and accurately recognizing the user's intent, the server can generate a more appropriate response.

[0990] "Audio input means" refers to a device and method for capturing user-uttered speech as a digital signal.

[0991] "Speech recognition means" refers to the technology and devices for converting voice input data into text data.

[0992] "Analysis means" refers to technology and devices that have the function of analyzing received text data using generative artificial intelligence and generating answers to users' questions.

[0993] "Speech synthesis means" refers to the technology and device that converts the generated response text into voice data.

[0994] "Output means" refers to a device that reproduces the voice generated by the voice synthesis means.

[0995] "Cloud computing technology" refers to technology that performs distributed processing over the Internet and efficiently executes large-scale data analysis and calculations.

[0996] A "virtual store" refers to a virtual shopping space built on the Internet where users can browse and purchase products.

[0997] "Apparatus" means a machine or mechanism designed for a specific purpose.

[0998] "Generative AI" refers to artificial intelligence technology that has the ability to learn large amounts of data and generate new information and answers based on human language and knowledge.

[0999] A "product knowledge database" refers to a database that stores detailed information related to products.

[1000] The present invention relates to a system in which a user can ask a question by voice and a response to the question is provided by voice, and the system uses a voice input means, a voice recognition means, an analysis means, a voice synthesis means, an output means, cloud computing technology, and a voice data capture device in a virtual store.

[1001] When a user asks a question through a voice input device in the virtual store, the voice is captured as a digital signal through a microphone. This digital signal is voice input data, and the terminal transmits the voice input data to a cloud server.

[1002] The cloud server uses the Google Cloud Speech-to-Text API to convert this voice input data into text data. This text data represents the user's question. For example, if a user asks, "Tell me about this product," the cloud server converts this into text data: "Tell me about this product."

[1003] Next, this text data is analyzed using generative artificial intelligence. OpenAI's GPT-3.5 is used as the analysis method. The cloud server uses the analysis method to analyze the text data and generate the optimal answer to the user's question. The analysis method understands the meaning of the question and extracts the necessary information from the product knowledge database.

[1004] The answer text generated by the analysis means contains specific information such as "This product is the latest model, priced at 50,000 yen, and in stock." This answer text is converted into audio data using the Google Cloud Text-to-Speech API. This audio data is provided in an audio format that is easy for the user to understand.

[1005] Finally, the generated voice data is played back by a voice synthesis means, and the speaker of the smart glasses or head-mounted display is used as an output means, allowing the user to receive a voice response to their question.

[1006] Specific examples are shown below.

[1007] When a user asks, "Tell me about this product," the smart glasses' microphone captures the voice and sends it to a cloud server. The cloud server converts the text data using the Google Cloud Speech-to-Text API, analyzes it using GPT-3.5, and generates an answer. The text data is then converted back into audio using the Google Cloud Text-to-Speech API, and the audio is played through the smart glasses' speaker.

[1008] An example of a prompt is as follows:

[1009] User Asks: "Tell me about this product"

[1010] Data to send to Google Cloud Speech-to-Text API: Audio data

[1011] Prompt GPT-3.5: "The user says, 'Tell me about this product.' Extract information from the corresponding product database and generate an appropriate answer."

[1012] Data to send to Google Cloud Text-to-Speech API: Generated response text

[1013] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1014] Step 1:

[1015] Voice input

[1016] A user asks a question to a terminal in a virtual store, such as "Tell me about this product." The terminal uses a voice input means to capture the user's voice as a digital signal. This captured voice data is the input. This input data is ready to be sent to the subsequent processing steps.

[1017] Step 2:

[1018] Voice Recognition

[1019] The device sends the captured voice data to a cloud server, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. During this conversion process, the voice waveform is analyzed and converted into the corresponding text, "Tell me about this product." This text data is the output. The phonemes of the voice are analyzed, and the most appropriate text is generated using a language model.

[1020] Step 3:

[1021] Question analysis and answer generation

[1022] The cloud server analyzes the text data using generative artificial intelligence (OpenAI's GPT-3.5). This analysis method understands the user's question and extracts appropriate information from a product knowledge database. The input is the converted text data "Tell me about this product," and the output is the answer text generated based on the analysis: "This product is the latest model, priced at 50,000 yen. In stock." The generative AI model analyzes the question and generates relevant information by filtering it from the database.

[1023] Step 4:

[1024] Speech synthesis

[1025] The cloud server sends the generated answer text to the Google Cloud Text-to-Speech API, which converts it into audio data. The input is the answer text, and the output is the corresponding audio data. In this process, the answer text is synthesized into natural-sounding audio and provided in a format that is easy for the user to understand. The generated audio data is prepared here.

[1026] Step 5:

[1027] Audio Output

[1028] The cloud server sends the generated voice data to the device. The device uses an output means (speakers in smart glasses or a head-mounted display) to play this voice to the user. The input is the generated voice data, and the output is a voice response played to the user. The device outputs the voice data through a designated speaker and provides it to the user.

[1029] These steps enable the system to quickly and accurately provide a spoken answer to a user's spoken question.

[1030] 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.

[1031] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. This system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit. It also includes a configuration that further combines an emotion engine that recognizes the user's emotions and provides appropriate answers based on those emotions.

[1032] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server then converts the voice into text data using a voice recognition means. The server then uses an emotion engine to recognize the user's emotion from the text data and provides the emotion information to the analysis means. Next, the analysis means using generative artificial intelligence analyzes the text data and emotion information and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by a voice synthesis means and finally provided to the user through an output means (e.g., a speaker).

[1033] Below, the specific program processing will be explained in natural language.

[1034] Program processing explanation

[1035] 1. Voice input

[1036] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[1037] Device: Captures the user's voice through a microphone.

[1038] 2. Voice Recognition

[1039] Device: Sends captured audio data to the server.

[1040] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[1041] 3. Emotion recognition

[1042] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it recognizes whether the user is excited or confused from a text such as "New smartphone pricing plan."

[1043] Server: Provides the recognized emotion information to the analysis means.

[1044] 4. Question Analysis and Answer Generation

[1045] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[1046] 5. Adjusting your answers

[1047] Server: Adjust the answer appropriately based on the perceived emotion information, for example adding more detailed explanation if the user is confused.

[1048] 6. Speech Synthesis

[1049] Server: Sends the generated answer text to the speech synthesis means.

[1050] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[1051] 7. Audio Output

[1052] Server: Sends the generated audio data to the device.

[1053] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[1054] Specific examples

[1055] User Question: What is the pricing plan for my new smartphone?

[1056] Capture audio input: The device captures audio and sends it to the server.

[1057] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[1058] Emotion recognition using an emotion engine: Recognizes when a user is confused from text data.

[1059] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[1060] Tailor your response: For confused users, add an additional "Would you like more options?"

[1061] Speech synthesis: Converting the generated text into speech.

[1062] Audio output: Provides answers to the user through the device's speaker.

[1063] In this way, the system according to the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is particularly useful in situations where detailed information about products is provided, realizing efficient customer service.

[1064] The processing flow will be explained below.

[1065] Step 1:

[1066] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[1067] Step 2:

[1068] Device: Captures the user's voice through a microphone.

[1069] Step 3:

[1070] Device: Sends captured audio data to the server.

[1071] Step 4:

[1072] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[1073] Step 5:

[1074] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it analyzes the text data and determines whether the user is interested, confused, or in a hurry.

[1075] Step 6:

[1076] Server: Provides the emotion information analyzed by the emotion engine to the analysis means.

[1077] Step 7:

[1078] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[1079] Step 8:

[1080] Server: Adjust the answer appropriately based on the perceived emotion information. For example, if the user is confused, add an additional explanation: "Would you like more options?"

[1081] Step 9:

[1082] Server: Sends the generated answer text to the speech synthesis means.

[1083] Step 10:

[1084] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[1085] Step 11:

[1086] Server: Sends the generated audio data to the device.

[1087] Step 12:

[1088] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[1089] This specific process allows the user to ask a question by voice and not only receive an appropriate answer quickly by voice, but also obtain a more friendly and understandable answer based on emotion recognition by the emotion engine.

[1090] Example 2

[1091] 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."

[1092] Conventional voice response systems can automatically generate and provide voice responses to user questions, but they have the problem of being unable to recognize the user's emotions and provide appropriate responses based on those emotions. This can lead to a poor user experience, especially when the user is confused or expecting something, as the system is unable to respond in accordance with that emotion. Furthermore, the responses often do not match the user's intentions, making it necessary to improve user satisfaction.

[1093] 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.

[1094] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into speech, an output means for playing back the generated speech, and an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the recognized emotion. This makes it possible to provide an appropriate answer according to the user's emotion, thereby improving the user experience and increasing user satisfaction. Furthermore, by using an intent recognition means, the server can accurately understand the user's intent and generate a more accurate answer.

[1095] "Audio input means" is a device or software for capturing speech produced by a user and transmitting that data to a processing system.

[1096] "Speech recognition means" means technology or equipment for converting voice data into text data, including algorithms and services used for that purpose.

[1097] "Generative AI" is an AI model used to analyze given input data and generate answers in a natural language format.

[1098] "Analysis means" refers to technology or devices that use generative artificial intelligence to analyze input data, understand the intent of the question, and generate an appropriate answer.

[1099] "Speech synthesis means" refers to a technique or device for converting generated character data into voice data.

[1100] "Output means" refers to hardware or software for reproducing audio data and providing information to the user by audio.

[1101] An "emotion recognition means" is a technology or device that identifies emotions from user input data and appropriately adjusts the system's response based on those emotions.

[1102] "Intention recognition means" refers to technology or devices that infer a user's intention from input data and generate appropriate analysis and responses.

[1103] A "product knowledge database" is a database that stores detailed information about products and serves as the basis for the system to search for information and generate answers.

[1104] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, an output unit, and an emotion recognition unit.

[1105] Specifically, when a user speaks a question through the microphone of a device on which the software is installed, the voice input means receives the voice and sends it to the server. For example, consider a situation where a user asks, "What is the price plan for my new smartphone?" The device's microphone captures the voice and transfers the data to the server. The server then uses the voice recognition means to convert the voice data into text data. This process can be performed using voice recognition services such as Google Cloud Speech-to-Text or Amazon Transcribe.

[1106] The emotion recognition means then analyzes the text data and recognizes the user's emotions. For example, it can recognize whether the user is excited or confused from the text "New smartphone pricing plan." This process can be performed using IBM Watson's Tone Analyzer, among other tools. The recognized emotion information is provided to the analysis means.

[1107] Next, an analysis means using generative AI (e.g., GPT-4) analyzes the text data and emotional information to generate an appropriate answer to the user's question. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month." This process is performed using a generative AI model.

[1108] The generated answer is sent to a speech synthesis means and converted into voice data. This voice synthesis can be performed using speech synthesis services such as Amazon Polly or Google Text-to-Speech. Finally, the generated voice data is sent to the device and played back to the user through the speaker.

[1109] The answer is also adjusted based on the emotion information recognized by the emotion recognition means. For example, if the user is confused, an explanation such as "Would you like more options?" is added to the answer.

[1110] Specific examples

[1111] User asks: "What plan is available for my new smartphone?"

[1112] Device audio input: Captures audio through the microphone and sends it to the server.

[1113] Server speech recognition: Converts speech into text data such as "Please tell me the pricing plan for my new smartphone."

[1114] Emotion recognition: Analyzes text data to recognize when a user is confused.

[1115] Analysis and Answer Generation: Generate the answer "New smartphone plans start at 5,000 yen per month."

[1116] Tailor your response: For confused users, add the explanation "Would you like more options?"

[1117] Speech synthesis: Converting the generated text into audio data.

[1118] Speech output: Provides answers to the user through a speaker.

[1119] In this way, the system based on the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is useful in situations where detailed information about products is provided, realizing efficient customer service.

[1120] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1121] Step 1:

[1122] Voice input

[1123] User: The user asks the device, "What is the pricing plan for my new smartphone?"

[1124] On the device: The built-in microphone is used to capture the user's voice, and the voice data is generated and temporarily stored in the device's memory.

[1125] Step 2:

[1126] Sending audio data

[1127] Terminal: The captured audio data is compressed and sent to the server via the network using a transfer protocol such as HTTP or HTTPS.

[1128] Input: User's voice data

[1129] Output: Audio data sent to the server

[1130] Step 3:

[1131] Voice Recognition

[1132] Server: Passes the received voice data to the voice recognition means.

[1133] Server: Converts voice data into text data using a speech recognition method. Calls a speech recognition API (e.g., Google Cloud Speech-to-Text) and converts voice data into text data.

[1134] Input: Audio data

[1135] Output: The result of converting speech to text data, e.g. "What is the price plan for my new smartphone?"

[1136] Step 4:

[1137] emotion recognition

[1138] Server: Passes the text data to the emotion recognition means.

[1139] Server: Uses emotion recognition means to recognize the user's emotion from the text data. Calls an emotion recognition API (e.g., IBM Watson Tone Analyzer) to obtain emotion information from the text data.

[1140] Input: Character data

[1141] Output: Emotional information such as user confusion or expectation

[1142] Step 5:

[1143] Question analysis and answer generation

[1144] Server: Passes text data and emotion information to the analysis means.

[1145] Server: Uses generative AI (e.g., GPT-4) to analyze text data and emotional information and generate appropriate answers. Enter prompt sentences into the generative AI model to generate answers.

[1146] Input: Text data, emotion information

[1147] Output: Answer text "New smartphone plans start at 5,000 yen per month."

[1148] Step 6:

[1149] Adjusting your answers

[1150] Server: Adjust the generated answer based on sentiment information, for example, adding additional explanation to the answer if the user is confused.

[1151] Input: Answer text, emotion information

[1152] Output: Tailored response text, e.g. "New smartphone plans start at ¥5,000 / month. Would you like more options?"

[1153] Step 7:

[1154] Speech synthesis

[1155] Server: Pass the adjusted answer text to the speech synthesis means.

[1156] Server: Convert the response text into audio data using a speech synthesis tool. Call a speech synthesis API (e.g., Amazon Polly).

[1157] Input: Adjusted answer text

[1158] Output: Voice data, for example, "New smartphone plans start at 5,000 yen per month. Would you like more options?"

[1159] Step 8:

[1160] Audio Output

[1161] Server: Sends the generated audio data to the device.

[1162] Terminal: The generated audio data is played back through a speaker and provided to the user.

[1163] Input: Audio data

[1164] Output: The answer provided to the user verbally

[1165] (Application example 2)

[1166] 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."

[1167] In factories, it is important for workers to quickly and accurately obtain the information they need. However, conventional methods require workers to manually search for information, which is time-consuming and labor-intensive and inefficient. In addition, it is difficult to respond appropriately while taking into account the emotions and fatigue levels of workers, which can affect the quality of work. This poses the problem of reduced work efficiency and an increased risk of errors.

[1168] 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 a voice input means, a voice recognition means for converting voice input data into character data, an analysis means for analyzing the character data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into voice, an output means for playing back the voice generated by the voice synthesis means, an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the emotion, and a means for providing voice instructions in cooperation with the device to support work in the factory. As a result, when a worker asks a question by voice, appropriate information can be provided in real time, and a response can be made that takes into account the worker's emotion and fatigue level.

[1169] "Voice input means" refers to a device or function that captures the user's voice.

[1170] "Speech recognition means" refers to software or a device that converts voice data captured by a voice input means into text data.

[1171] "Generative AI" is a type of AI that analyzes text data and generates appropriate answers.

[1172] The "analysis means" is a function or device that uses generative artificial intelligence to analyze character data and generate answers to user questions.

[1173] The "voice synthesis means" is software or a device that converts the answer generated by the analysis means into voice.

[1174] The "output means" is a device or function for reproducing the voice generated by the voice synthesis means to the user.

[1175] "Emotion recognition means" refers to a device or function that recognizes the user's emotions and adjusts the content of the response based on those emotions.

[1176] "Devices for supporting work within a factory" refers to systems and devices that are intended to support work within a factory.

[1177] The present invention is a system that responds to factory workers' voice questions by providing appropriate information in real time and taking into consideration the worker's emotions and fatigue level. The system includes a voice input means, a voice recognition means, an analysis means using generative artificial intelligence, a voice synthesis means, an output means, an emotion recognition means, and a means for linking with devices for supporting factory work.

[1178] Program processing explanation

[1179] 1. Voice input

[1180] User: A worker in a factory asks the robot, "What should I do next?"

[1181] Robot: Captures the worker's voice through a built-in microphone.

[1182] 2. Voice Recognition

[1183] Robot: Sends captured audio data to the server.

[1184] Server: Uses a speech recognition method (for example, Google Speech-to-Text API) to convert the received voice data into text data.

[1185] 3. Emotion recognition

[1186] Server: The emotion recognition means receives the text data and analyzes the worker's emotions. For example, it recognizes that the worker is tired.

[1187] Server: Provides the recognized emotion information to the analysis means.

[1188] 4. Question Analysis and Answer Generation

[1189] Server: Using generative artificial intelligence (e.g., GPT-4), it analyzes the received text data and emotional information, understands the intent of the question, and then generates an appropriate response.

[1190] Example: Generate the answer "The next thing to do is check the tool kit."

[1191] 5. Adjusting your answers

[1192] Server: Based on the recognized emotional information, the server adjusts the response appropriately, for example adding an instruction to a tired worker saying, "Take it easy and take a short break."

[1193] 6. Speech Synthesis

[1194] Server: The generated answer text is sent to a speech synthesis means (e.g., Google Text-to-Speech API) and converted into voice data.

[1195] 7. Audio Output

[1196] Robot: The generated voice data is played back through the robot's speaker to provide a response to the worker.

[1197] Specific examples

[1198] Scenario 1: Dealing with fatigued workers

[1199] User Question: "What do I do next?"

[1200] Voice input capture: The robot captures the voice and sends it to the server.

[1201] Server speech recognition processing: Converts the speech into text data such as "What should I do next?"

[1202] Emotion recognition using an emotion engine: Recognizing that a worker is tired from text data.

[1203] Analysis and answer generation using generative artificial intelligence: The answer generated is, "The next thing you should do is check your tool kit."

[1204] Response adjustment: For tired workers, add the instruction, "Take it easy and take a short break."

[1205] Speech synthesis: Converting the generated text into speech.

[1206] Voice output: Provides answers to the worker through the robot's speaker.

[1207] Prompt Sentence Examples

[1208] When asked, "What's next?"

[1209] Recognize when a worker is tired and take that into account when generating text that answers the question "What task needs to be done next?"

[1210] example:

[1211] Question: "What should I do next?"

[1212] (Recognize fatigue): The next thing to do is check your tool kit. Take a short break and don't push yourself too hard.

[1213] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1214] Step 1:

[1215] Voice Input Processing

[1216] Subject: User

[1217] Specific operation: A worker in a factory asks the robot, "What should I do next?"

[1218] Input: Worker's voice

[1219] Data processing or data calculation: Audio is captured by a microphone.

[1220] Output: Captured audio data

[1221] Step 2:

[1222] Sending audio data

[1223] Subject: Terminal (robot)

[1224] Specific operation: The robot sends the captured voice data to the server.

[1225] Input: Audio data

[1226] Data processing or data calculation: The voice data is packetized and sent to the server via the network.

[1227] Output: Audio data sent to the server

[1228] Step 3:

[1229] Speech Recognition Processing

[1230] Subject: Server

[1231] Specific operation: The server converts the voice data into text data using a voice recognition method (for example, Google Speech-to-Text API).

[1232] Input: Audio data

[1233] Data processing or data calculation: Analyzing voice data and converting it into corresponding text data.

[1234] Output: Converted character data

[1235] Step 4:

[1236] Emotion Recognition Processing

[1237] Subject: Server

[1238] Specific operation: The server uses emotion recognition means to analyze the worker's emotions from the text data.

[1239] Input: Character data

[1240] Data processing or data calculation: Analyze emotions using natural language processing techniques and generate emotional information.

[1241] Output: Emotional information

[1242] Step 5:

[1243] Question analysis and answer generation processing

[1244] Subject: Server

[1245] Specific operation: The server uses generative artificial intelligence (e.g., GPT-4) to analyze text data and emotional information and generate an appropriate response.

[1246] Input: Text data and emotion information

[1247] Data processing or data computation: Generate prompts for a generative artificial intelligence model, which then analyzes and generates an answer.

[1248] Output: Generated answer text

[1249] Step 6:

[1250] Response adjustment process

[1251] Subject: Server

[1252] Specific operation: The server adjusts the generated answer appropriately based on the recognized emotion information.

[1253] Input: Generated answer text and sentiment information

[1254] Data processing or data calculation: Refer to the emotional information and add supplements or corrections to the answer text.

[1255] Output: Adjusted answer text

[1256] Step 7:

[1257] Speech synthesis processing

[1258] Subject: Server

[1259] Specific operation: The server converts the adjusted answer text into audio data using a speech synthesis means (e.g., Google Text-to-Speech API).

[1260] Input: Adjusted answer text

[1261] Data processing or data calculation: Text data is input into a speech synthesis engine to generate corresponding speech data.

[1262] Output: Generated audio data

[1263] Step 8:

[1264] Sending audio data

[1265] Subject: Server

[1266] Specific operation: The server sends the generated voice data to the robot.

[1267] Input: Generated audio data

[1268] Data processing or data calculation: The voice data is packetized and sent to the robot via the network.

[1269] Output: Audio data sent to the robot

[1270] Step 9:

[1271] Audio Output Processing

[1272] Subject: Terminal (robot)

[1273] Specific operation: The robot plays back the audio data through a speaker and provides a response to the worker.

[1274] Input: Transmitted audio data

[1275] Data processing or data calculation: Deserializing the audio data and converting it into a format that can be played by the speaker.

[1276] Output: Audio response that can be heard by the worker

[1277] 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.

[1278] 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.

[1279] 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.

[1280] [Fourth embodiment]

[1281] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1282] 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.

[1283] 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).

[1284] 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.

[1285] 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.

[1286] 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).

[1287] 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.

[1288] 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.

[1289] 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.

[1290] 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.

[1291] 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.

[1292] 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.

[1293] 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."

[1294] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[1295] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server converts the voice into text data using voice recognition means. Then, analysis means using generative artificial intelligence analyzes the text data and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by voice synthesis means, and finally provided to the user through output means (e.g., a speaker).

[1296] Below, the specific program processing will be explained in natural language.

[1297] Program processing explanation

[1298] 1. Voice input

[1299] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[1300] Device: Captures the user's voice through a microphone.

[1301] 2. Voice Recognition

[1302] Device: Sends captured audio data to the server.

[1303] Server: Uses speech recognition means to convert the received voice data into text data.

[1304] Server: For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[1305] 3. Question Analysis and Answer Generation

[1306] Server: Analyzes the received text data using generative AI. The analysis means understands the meaning of the question and searches for information from a related product knowledge database.

[1307] Server: For example, generate information such as "New smartphone plans start at 5,000 yen per month" as an answer to a question.

[1308] 4. Speech Synthesis

[1309] Server: Sends the generated answer text to the speech synthesis means.

[1310] Server: The speech synthesis means converts the text data into speech data, generating speech data such as "The new smartphone plan starts at 5,000 yen per month."

[1311] 5. Audio Output

[1312] Server: Sends the generated audio data to the device.

[1313] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to questions.

[1314] Specific examples

[1315] User Question: What is the pricing plan for my new smartphone?

[1316] Capture audio input: The device captures audio and sends it to the server.

[1317] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[1318] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[1319] Speech synthesis: Converting the generated text into speech.

[1320] Audio output: Provides answers to the user through the device's speaker.

[1321] In this way, the system based on the present invention can provide a prompt and accurate answer to a user's voice question. Furthermore, by linking with an intent recognition means and a product knowledge database, it can generate even more appropriate answers, improving the user experience. This system is particularly useful in situations where detailed information about a product is to be provided, enabling efficient customer service.

[1322] The processing flow will be explained below.

[1323] Step 1:

[1324] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[1325] Step 2:

[1326] Device: Captures the user's voice through a microphone.

[1327] Step 3:

[1328] Device: Sends captured audio data to the server.

[1329] Step 4:

[1330] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[1331] Step 5:

[1332] Server: Analyzes the received text data using generative AI. The analysis means analyzes the text data and understands the intent of the question.

[1333] Step 6:

[1334] Server: Searches for information corresponding to the question from a product knowledge database. For example, retrieves information on "new smartphone pricing plans" from the database.

[1335] Step 7:

[1336] Server: Generates an answer to the question based on the search results. For example, it generates the text "New smartphone plans start at 5,000 yen per month" as an answer to the user's question.

[1337] Step 8:

[1338] Server: Sends the generated answer text to the speech synthesis means.

[1339] Step 9:

[1340] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[1341] Step 10:

[1342] Server: Sends the generated audio data to the device.

[1343] Step 11:

[1344] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[1345] This series of processes allows the user to ask a question by voice and quickly receive an appropriate answer by voice.

[1346] Example 1

[1347] 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."

[1348] Modern information provision systems require a smooth process for users to ask questions by voice and receive appropriate answers. However, existing systems often have problems with speech recognition accuracy and answer generation and reproduction. In particular, they can be difficult to provide accurate and appropriate answers in real time. They also lack the ability to accurately understand user intent and generate the optimal answer. This can lead to a poor user experience.

[1349] 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.

[1350] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence to generate an answer, a voice synthesis means for converting the generated answer into speech, a means for transmitting the generated speech data to an output means, and an output means for playing back the speech generated by the voice synthesis means. This allows a user to obtain a highly accurate answer in real time when asking a question by voice. Furthermore, by linking with an intent recognition means and a product knowledge database, it is possible to generate more appropriate answers and improve the user experience.

[1351] "Voice input means" refers to a device or software that captures a user's speech or voice data and records it as digital data.

[1352] "Speech recognition means" refers to technology that analyzes voice data and converts the voice into corresponding text data.

[1353] "Generative AI" refers to an AI technology that analyzes given text or data and creates optimal answers or products based on that content.

[1354] "Analysis means" refers to a process or device that analyzes received text data, understands the user's intent, and generates an appropriate response.

[1355] "Speech synthesis means" refers to technology that converts text data into voice data.

[1356] "Output means" refers to a device or system for playing back the generated audio data.

[1357] "Intention recognition means" refers to the technology or process for inferring a user's intentions or desires from the data entered by the user.

[1358] A "product knowledge database" refers to a database that stores detailed information such as the features, prices, and specifications of specific products.

[1359] "Search methods" refer to the techniques and processes used to search for and retrieve data or information based on specific criteria.

[1360] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit.

[1361] When a user asks a question, their voice is captured through a voice input means (a microphone or the built-in microphone of a smart device). This voice data is sent to a server by the device. For example, when using a smartphone or tablet, the voice data is sent via the Internet.

[1362] The server uses a speech recognition tool to convert the received voice data into text data. For example, Google Cloud Speech-to-Text API can be used. The voice data is converted into text data such as "Please tell me the price plan for my new smartphone."

[1363] Next, the text data is analyzed using generative artificial intelligence (e.g., OpenAI GPT-4). This analysis method understands the meaning of the user's question and generates an answer based on related information. Specifically, based on the context and keywords of the question, an answer appropriate to the question is searched from a product knowledge database and an appropriate answer text is generated.

[1364] The server sends the generated response text to a speech synthesis means (e.g., Amazon Polly), which converts the text data into speech data. The speech data generated is, "The new smartphone plan starts at 5,000 yen per month."

[1365] Finally, the generated voice data is sent to the terminal and played back through the terminal's output means (speaker or headset), allowing the user to receive the answer to their question by voice.

[1366] For example, if a user asks, "What is the pricing plan for my new smartphone?", the following prompt is input to the generative AI:

[1367] Example prompt sentence:

[1368] User Question: What is the pricing plan for my new smartphone?

[1369] Generative AI answer: New smartphone plans start at 5,000 yen per month.

[1370] Based on this prompt, the generative artificial intelligence generates an appropriate response, which is then converted into voice data by a speech synthesis means and finally provided to the user. This system allows users to receive fast and accurate information by voice. It is particularly useful in situations where detailed information about products is to be provided, and realizes efficient customer service.

[1371] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1372] Step 1: Voice Input

[1373] User: Asks the device, "What is the price plan for my new smartphone?" The user's input is voice data.

[1374] Device: Uses a built-in microphone to capture the user's voice, which is then stored digitally.

[1375] Output: The captured audio data is generated.

[1376] Step 2: Sending audio data

[1377] On your device: The captured audio data is sent to the server using the HTTPS protocol, which sends the data securely over your internet connection.

[1378] Input: Audio data

[1379] Output: Audio data sent to the server

[1380] Step 3: Voice Recognition

[1381] Server: Calls the API of the speech recognition method (for example, Google Cloud Speech-to-Text API) and converts the received voice data into text data.

[1382] Input: Audio data sent to the server

[1383] Data processing: Converting audio data into text using speech recognition tools. This process involves analyzing the audio data and generating a corresponding string of characters.

[1384] Output: Converted text data (e.g., "What is the price plan for my new smartphone?")

[1385] Step 4: Parsing the question and generating an answer

[1386] Server: Input the text data into a generative artificial intelligence (e.g., OpenAI GPT-4) and analyze it.

[1387] Input: Converted character data

[1388] Data calculation: The generative AI model generates appropriate answers to questions based on prompts, searching for information from a product knowledge database based on the context and keywords of the question.

[1389] Output: Generated answer text (e.g., "New smartphone plans start at 5,000 yen per month.")

[1390] Step 5: Text-to-speech synthesis of the answer text

[1391] Server: The generated answer text is input into a speech synthesis means (e.g., Amazon Polly) and converted into voice data.

[1392] Input: Generated answer text

[1393] Data processing: Converting text into audio data using speech synthesis tools.

[1394] Output: Generated speech data (e.g., "New smartphone plans start at 5,000 yen per month.")

[1395] Step 6: Sending audio data

[1396] Server: Sends the generated audio data to the device using the HTTP / HTTPS protocol to ensure secure communication.

[1397] Input: Generated audio data

[1398] Output: Audio data sent to the device

[1399] Step 7: Audio Output

[1400] Terminal: Passes the audio data received by the terminal to the output means (speaker or headset).

[1401] Input: Audio data sent to the device

[1402] What it does: Plays audio data through a speaker and provides answers to the user.

[1403] Output: A spoken response to the user (e.g., "New smartphone plans start at ¥5,000 per month.")

[1404] By going through the above processing steps, when a user asks a question by voice, the user can quickly and accurately receive a response by voice.

[1405] (Application example 1)

[1406] 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."

[1407] Conventional voice recognition systems and automated response systems focus only on simple voice input and conversion to text data, and lack the ability to respond to detailed user needs and complex questions. Furthermore, especially in new digital markets such as virtual stores, there is a growing need for systems that allow users to quickly and accurately obtain product information and engage in real-time voice interaction. Therefore, it is necessary to provide a new system that meets these needs and improves the user experience.

[1408] 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.

[1409] In this invention, the server includes a speech recognition unit that converts voice input data into text data, an analysis unit that uses generative artificial intelligence to analyze the text data and generate a response, and a speech synthesis unit that converts the generated response into speech. This allows the server to capture voice data when a user asks a question by voice in a virtual store, execute processing using cloud computing technology, and provide the user with a quick and accurate response. Furthermore, by obtaining information from a product knowledge database and accurately recognizing the user's intent, the server can generate a more appropriate response.

[1410] "Audio input means" refers to a device and method for capturing user-uttered speech as a digital signal.

[1411] "Speech recognition means" refers to the technology and devices for converting voice input data into text data.

[1412] "Analysis means" refers to technology and devices that have the function of analyzing received text data using generative artificial intelligence and generating answers to users' questions.

[1413] "Speech synthesis means" refers to the technology and device that converts the generated response text into voice data.

[1414] "Output means" refers to a device that reproduces the voice generated by the voice synthesis means.

[1415] "Cloud computing technology" refers to technology that performs distributed processing over the Internet and efficiently executes large-scale data analysis and calculations.

[1416] A "virtual store" refers to a virtual shopping space built on the Internet where users can browse and purchase products.

[1417] "Apparatus" means a machine or mechanism designed for a specific purpose.

[1418] "Generative AI" refers to artificial intelligence technology that has the ability to learn large amounts of data and generate new information and answers based on human language and knowledge.

[1419] A "product knowledge database" refers to a database that stores detailed information related to products.

[1420] The present invention relates to a system in which a user can ask a question by voice and a response to the question is provided by voice, and the system uses a voice input means, a voice recognition means, an analysis means, a voice synthesis means, an output means, cloud computing technology, and a voice data capture device in a virtual store.

[1421] When a user asks a question through a voice input device in the virtual store, the voice is captured as a digital signal through a microphone. This digital signal is voice input data, and the terminal transmits the voice input data to a cloud server.

[1422] The cloud server uses the Google Cloud Speech-to-Text API to convert this voice input data into text data. This text data represents the user's question. For example, if a user asks, "Tell me about this product," the cloud server converts this into text data: "Tell me about this product."

[1423] Next, this text data is analyzed using generative artificial intelligence. OpenAI's GPT-3.5 is used as the analysis method. The cloud server uses the analysis method to analyze the text data and generate the optimal answer to the user's question. The analysis method understands the meaning of the question and extracts the necessary information from the product knowledge database.

[1424] The answer text generated by the analysis means contains specific information such as "This product is the latest model, priced at 50,000 yen, and in stock." This answer text is converted into audio data using the Google Cloud Text-to-Speech API. This audio data is provided in an audio format that is easy for the user to understand.

[1425] Finally, the generated voice data is played back by a voice synthesis means, and the speaker of the smart glasses or head-mounted display is used as an output means, allowing the user to receive a voice response to their question.

[1426] Specific examples are shown below.

[1427] When a user asks, "Tell me about this product," the smart glasses' microphone captures the voice and sends it to a cloud server. The cloud server converts the text data using the Google Cloud Speech-to-Text API, analyzes it using GPT-3.5, and generates an answer. The text data is then converted back into audio using the Google Cloud Text-to-Speech API, and the audio is played through the smart glasses' speaker.

[1428] An example of a prompt is as follows:

[1429] User Asks: "Tell me about this product"

[1430] Data to send to Google Cloud Speech-to-Text API: Audio data

[1431] Prompt GPT-3.5: "The user says, 'Tell me about this product.' Extract information from the corresponding product database and generate an appropriate answer."

[1432] Data to send to Google Cloud Text-to-Speech API: Generated response text

[1433] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1434] Step 1:

[1435] Voice input

[1436] A user asks a question to a terminal in a virtual store, such as "Tell me about this product." The terminal uses a voice input means to capture the user's voice as a digital signal. This captured voice data is the input. This input data is ready to be sent to the subsequent processing steps.

[1437] Step 2:

[1438] Voice Recognition

[1439] The device sends the captured voice data to a cloud server, which uses the Google Cloud Speech-to-Text API to convert the voice data into text data. During this conversion process, the voice waveform is analyzed and converted into the corresponding text, "Tell me about this product." This text data is the output. The phonemes of the voice are analyzed, and the most appropriate text is generated using a language model.

[1440] Step 3:

[1441] Question analysis and answer generation

[1442] The cloud server analyzes the text data using generative artificial intelligence (OpenAI's GPT-3.5). This analysis method understands the user's question and extracts appropriate information from a product knowledge database. The input is the converted text data "Tell me about this product," and the output is the answer text generated based on the analysis: "This product is the latest model, priced at 50,000 yen. In stock." The generative AI model analyzes the question and generates relevant information by filtering it from the database.

[1443] Step 4:

[1444] Speech synthesis

[1445] The cloud server sends the generated answer text to the Google Cloud Text-to-Speech API, which converts it into audio data. The input is the answer text, and the output is the corresponding audio data. In this process, the answer text is synthesized into natural-sounding audio and provided in a format that is easy for the user to understand. The generated audio data is prepared here.

[1446] Step 5:

[1447] Audio Output

[1448] The cloud server sends the generated voice data to the device. The device uses an output means (speakers in smart glasses or a head-mounted display) to play this voice to the user. The input is the generated voice data, and the output is a voice response played to the user. The device outputs the voice data through a designated speaker and provides it to the user.

[1449] These steps enable the system to quickly and accurately provide a spoken answer to a user's spoken question.

[1450] 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.

[1451] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. This system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, and an output unit. It also includes a configuration that further combines an emotion engine that recognizes the user's emotions and provides appropriate answers based on those emotions.

[1452] Specifically, when a user speaks a question through the microphone of a terminal on which the software is installed, the voice input means receives the voice and sends it to the server. The server then converts the voice into text data using a voice recognition means. The server then uses an emotion engine to recognize the user's emotion from the text data and provides the emotion information to the analysis means. Next, the analysis means using generative artificial intelligence analyzes the text data and emotion information and generates an appropriate answer to the user's question. This generated answer text is converted into voice data by a voice synthesis means and finally provided to the user through an output means (e.g., a speaker).

[1453] Below, the specific program processing will be explained in natural language.

[1454] Program processing explanation

[1455] 1. Voice input

[1456] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[1457] Device: Captures the user's voice through a microphone.

[1458] 2. Voice Recognition

[1459] Device: Sends captured audio data to the server.

[1460] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[1461] 3. Emotion recognition

[1462] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it recognizes whether the user is excited or confused from a text such as "New smartphone pricing plan."

[1463] Server: Provides the recognized emotion information to the analysis means.

[1464] 4. Question Analysis and Answer Generation

[1465] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[1466] 5. Adjusting your answers

[1467] Server: Adjust the answer appropriately based on the perceived emotion information, for example adding more detailed explanation if the user is confused.

[1468] 6. Speech Synthesis

[1469] Server: Sends the generated answer text to the speech synthesis means.

[1470] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[1471] 7. Audio Output

[1472] Server: Sends the generated audio data to the device.

[1473] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[1474] Specific examples

[1475] User Question: What is the pricing plan for my new smartphone?

[1476] Capture audio input: The device captures audio and sends it to the server.

[1477] Server speech recognition processing: Converts the speech into text data such as "Please tell me the pricing plan for my new smartphone."

[1478] Emotion recognition using an emotion engine: Recognizes when a user is confused from text data.

[1479] Analysis and answer generation using generative artificial intelligence: Generate the answer, "The new smartphone plan starts at 5,000 yen per month."

[1480] Tailor your response: For confused users, add an additional "Would you like more options?"

[1481] Speech synthesis: Converting the generated text into speech.

[1482] Audio output: Provides answers to the user through the device's speaker.

[1483] In this way, the system according to the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is particularly useful in situations where detailed information about products is provided, realizing efficient customer service.

[1484] The processing flow will be explained below.

[1485] Step 1:

[1486] User: A user visits a SoftBank shop and asks the terminal, "Please tell me the pricing plan for my new smartphone."

[1487] Step 2:

[1488] Device: Captures the user's voice through a microphone.

[1489] Step 3:

[1490] Device: Sends captured audio data to the server.

[1491] Step 4:

[1492] Server: Uses speech recognition to convert the received voice data into text data. For example, convert the voice data "Please tell me the price plan for my new smartphone" into text data "Please tell me the price plan for my new smartphone."

[1493] Step 5:

[1494] Server: The emotion engine receives the text data and analyzes the user's emotions. For example, it analyzes the text data and determines whether the user is interested, confused, or in a hurry.

[1495] Step 6:

[1496] Server: Provides the emotion information analyzed by the emotion engine to the analysis means.

[1497] Step 7:

[1498] Server: Analyzes the received text data and emotional information using generative AI. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month" as an answer to the question.

[1499] Step 8:

[1500] Server: Adjust the answer appropriately based on the perceived emotion information. For example, if the user is confused, add an additional explanation: "Would you like more options?"

[1501] Step 9:

[1502] Server: Sends the generated answer text to the speech synthesis means.

[1503] Step 10:

[1504] Server: The speech synthesis means converts the text data into speech data. For example, it generates speech data such as "The new smartphone plan starts at 5,000 yen per month."

[1505] Step 11:

[1506] Server: Sends the generated audio data to the device.

[1507] Step 12:

[1508] Terminal: Plays back audio data to the user through a speaker, which is the output means, and provides answers to the user's questions.

[1509] This specific process allows the user to ask a question by voice and not only receive an appropriate answer quickly by voice, but also obtain a more friendly and understandable answer based on emotion recognition by the emotion engine.

[1510] Example 2

[1511] 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."

[1512] Conventional voice response systems can automatically generate and provide voice responses to user questions, but they have the problem of being unable to recognize the user's emotions and provide appropriate responses based on those emotions. This can lead to a poor user experience, especially when the user is confused or expecting something, as the system is unable to respond in accordance with that emotion. Furthermore, the responses often do not match the user's intentions, making it necessary to improve user satisfaction.

[1513] 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.

[1514] In this invention, the server includes a voice input means, a voice recognition means for converting the voice input data into text data, an analysis means for analyzing the text data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into speech, an output means for playing back the generated speech, and an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the recognized emotion. This makes it possible to provide an appropriate answer according to the user's emotion, thereby improving the user experience and increasing user satisfaction. Furthermore, by using an intent recognition means, the server can accurately understand the user's intent and generate a more accurate answer.

[1515] "Audio input means" is a device or software for capturing speech produced by a user and transmitting that data to a processing system.

[1516] "Speech recognition means" means technology or equipment for converting voice data into text data, including algorithms and services used for that purpose.

[1517] "Generative AI" is an AI model used to analyze given input data and generate answers in a natural language format.

[1518] "Analysis means" refers to technology or devices that use generative artificial intelligence to analyze input data, understand the intent of the question, and generate an appropriate answer.

[1519] "Speech synthesis means" refers to a technique or device for converting generated character data into voice data.

[1520] "Output means" refers to hardware or software for reproducing audio data and providing information to the user by audio.

[1521] An "emotion recognition means" is a technology or device that identifies emotions from user input data and appropriately adjusts the system's response based on those emotions.

[1522] "Intention recognition means" refers to technology or devices that infer a user's intention from input data and generate appropriate analysis and responses.

[1523] A "product knowledge database" is a database that stores detailed information about products and serves as the basis for the system to search for information and generate answers.

[1524] The present invention relates to a system that automatically generates and provides answers to questions posed by a user through speech. The system includes a speech input unit, a speech recognition unit, an analysis unit using generative artificial intelligence, a speech synthesis unit, an output unit, and an emotion recognition unit.

[1525] Specifically, when a user speaks a question through the microphone of a device on which the software is installed, the voice input means receives the voice and sends it to the server. For example, consider a situation where a user asks, "What is the price plan for my new smartphone?" The device's microphone captures the voice and transfers the data to the server. The server then uses the voice recognition means to convert the voice data into text data. This process can be performed using voice recognition services such as Google Cloud Speech-to-Text or Amazon Transcribe.

[1526] The emotion recognition means then analyzes the text data and recognizes the user's emotions. For example, it can recognize whether the user is excited or confused from the text "New smartphone pricing plan." This process can be performed using IBM Watson's Tone Analyzer, among other tools. The recognized emotion information is provided to the analysis means.

[1527] Next, an analysis means using generative AI (e.g., GPT-4) analyzes the text data and emotional information to generate an appropriate answer to the user's question. The analysis means understands the intent of the question and searches for information from a product knowledge database. For example, it generates information such as "New smartphone pricing plans start at 5,000 yen per month." This process is performed using a generative AI model.

[1528] The generated answer is sent to a speech synthesis means and converted into voice data. This voice synthesis can be performed using speech synthesis services such as Amazon Polly or Google Text-to-Speech. Finally, the generated voice data is sent to the device and played back to the user through the speaker.

[1529] The answer is also adjusted based on the emotion information recognized by the emotion recognition means. For example, if the user is confused, an explanation such as "Would you like more options?" is added to the answer.

[1530] Specific examples

[1531] User asks: "What plan is available for my new smartphone?"

[1532] Device audio input: Captures audio through the microphone and sends it to the server.

[1533] Server speech recognition: Converts speech into text data such as "Please tell me the pricing plan for my new smartphone."

[1534] Emotion recognition: Analyzes text data to recognize when a user is confused.

[1535] Analysis and Answer Generation: Generate the answer "New smartphone plans start at 5,000 yen per month."

[1536] Tailor your response: For confused users, add the explanation "Would you like more options?"

[1537] Speech synthesis: Converting the generated text into audio data.

[1538] Speech output: Provides answers to the user through a speaker.

[1539] In this way, the system based on the present invention can not only provide a quick and accurate answer to a user's voice question, but also recognize the user's emotions and adjust the answer accordingly to provide an optimal user experience. This system is useful in situations where detailed information about products is provided, realizing efficient customer service.

[1540] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1541] Step 1:

[1542] Voice input

[1543] User: The user asks the device, "What is the pricing plan for my new smartphone?"

[1544] On the device: The built-in microphone is used to capture the user's voice, and the voice data is generated and temporarily stored in the device's memory.

[1545] Step 2:

[1546] Sending audio data

[1547] Terminal: The captured audio data is compressed and sent to the server via the network using a transfer protocol such as HTTP or HTTPS.

[1548] Input: User's voice data

[1549] Output: Audio data sent to the server

[1550] Step 3:

[1551] Voice Recognition

[1552] Server: Passes the received voice data to the voice recognition means.

[1553] Server: Converts voice data into text data using a speech recognition method. Calls a speech recognition API (e.g., Google Cloud Speech-to-Text) and converts voice data into text data.

[1554] Input: Audio data

[1555] Output: The result of converting speech to text data, e.g. "What is the price plan for my new smartphone?"

[1556] Step 4:

[1557] emotion recognition

[1558] Server: Passes the text data to the emotion recognition means.

[1559] Server: Uses emotion recognition means to recognize the user's emotion from the text data. Calls an emotion recognition API (e.g., IBM Watson Tone Analyzer) to obtain emotion information from the text data.

[1560] Input: Character data

[1561] Output: Emotional information such as user confusion or expectation

[1562] Step 5:

[1563] Question analysis and answer generation

[1564] Server: Passes text data and emotion information to the analysis means.

[1565] Server: Uses generative AI (e.g., GPT-4) to analyze text data and emotional information and generate appropriate answers. Enter prompt sentences into the generative AI model to generate answers.

[1566] Input: Text data, emotion information

[1567] Output: Answer text "New smartphone plans start at 5,000 yen per month."

[1568] Step 6:

[1569] Adjusting your answers

[1570] Server: Adjust the generated answer based on sentiment information, for example, adding additional explanation to the answer if the user is confused.

[1571] Input: Answer text, emotion information

[1572] Output: Tailored response text, e.g. "New smartphone plans start at ¥5,000 / month. Would you like more options?"

[1573] Step 7:

[1574] Speech synthesis

[1575] Server: Pass the adjusted answer text to the speech synthesis means.

[1576] Server: Convert the response text into audio data using a speech synthesis tool. Call a speech synthesis API (e.g., Amazon Polly).

[1577] Input: Adjusted answer text

[1578] Output: Voice data, for example, "New smartphone plans start at 5,000 yen per month. Would you like more options?"

[1579] Step 8:

[1580] Audio Output

[1581] Server: Sends the generated audio data to the device.

[1582] Terminal: The generated audio data is played back through a speaker and provided to the user.

[1583] Input: Audio data

[1584] Output: The answer provided to the user verbally

[1585] (Application example 2)

[1586] 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."

[1587] In factories, it is important for workers to quickly and accurately obtain the information they need. However, conventional methods require workers to manually search for information, which is time-consuming and labor-intensive and inefficient. In addition, it is difficult to respond appropriately while taking into account the emotions and fatigue levels of workers, which can affect the quality of work. This poses the problem of reduced work efficiency and an increased risk of errors.

[1588] 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 a voice input means, a voice recognition means for converting voice input data into character data, an analysis means for analyzing the character data using generative artificial intelligence and generating an answer, a voice synthesis means for converting the generated answer into voice, an output means for playing back the voice generated by the voice synthesis means, an emotion recognition means for recognizing the user's emotion and adjusting the answer based on the emotion, and a means for providing voice instructions in cooperation with the device to support work in the factory. As a result, when a worker asks a question by voice, appropriate information can be provided in real time, and a response can be made that takes into account the worker's emotion and fatigue level.

[1589] "Voice input means" refers to a device or function that captures the user's voice.

[1590] "Speech recognition means" refers to software or a device that converts voice data captured by a voice input means into text data.

[1591] "Generative AI" is a type of AI that analyzes text data and generates appropriate answers.

[1592] The "analysis means" is a function or device that uses generative artificial intelligence to analyze character data and generate answers to user questions.

[1593] The "voice synthesis means" is software or a device that converts the answer generated by the analysis means into voice.

[1594] The "output means" is a device or function for reproducing the voice generated by the voice synthesis means to the user.

[1595] "Emotion recognition means" refers to a device or function that recognizes the user's emotions and adjusts the content of the response based on those emotions.

[1596] "Devices for supporting work within a factory" refers to systems and devices that are intended to support work within a factory.

[1597] The present invention is a system that responds to factory workers' voice questions by providing appropriate information in real time and taking into consideration the worker's emotions and fatigue level. The system includes a voice input means, a voice recognition means, an analysis means using generative artificial intelligence, a voice synthesis means, an output means, an emotion recognition means, and a means for linking with devices for supporting factory work.

[1598] Program processing explanation

[1599] 1. Voice input

[1600] User: A worker in a factory asks the robot, "What should I do next?"

[1601] Robot: Captures the worker's voice through a built-in microphone.

[1602] 2. Voice Recognition

[1603] Robot: Sends captured audio data to the server.

[1604] Server: Uses a speech recognition method (for example, Google Speech-to-Text API) to convert the received voice data into text data.

[1605] 3. Emotion recognition

[1606] Server: The emotion recognition means receives the text data and analyzes the worker's emotions. For example, it recognizes that the worker is tired.

[1607] Server: Provides the recognized emotion information to the analysis means.

[1608] 4. Question Analysis and Answer Generation

[1609] Server: Using generative artificial intelligence (e.g., GPT-4), it analyzes the received text data and emotional information, understands the intent of the question, and then generates an appropriate response.

[1610] Example: Generate the answer "The next thing to do is check the tool kit."

[1611] 5. Adjusting your answers

[1612] Server: Based on the recognized emotional information, the server adjusts the response appropriately, for example adding an instruction to a tired worker saying, "Take it easy and take a short break."

[1613] 6. Speech Synthesis

[1614] Server: The generated answer text is sent to a speech synthesis means (e.g., Google Text-to-Speech API) and converted into voice data.

[1615] 7. Audio Output

[1616] Robot: The generated voice data is played back through the robot's speaker to provide a response to the worker.

[1617] Specific examples

[1618] Scenario 1: Dealing with fatigued workers

[1619] User Question: "What do I do next?"

[1620] Voice input capture: The robot captures the voice and sends it to the server.

[1621] Server speech recognition processing: Converts the speech into text data such as "What should I do next?"

[1622] Emotion recognition using an emotion engine: Recognizing that a worker is tired from text data.

[1623] Analysis and answer generation using generative artificial intelligence: The answer generated is, "The next thing you should do is check your tool kit."

[1624] Response adjustment: For tired workers, add the instruction, "Take it easy and take a short break."

[1625] Speech synthesis: Converting the generated text into speech.

[1626] Voice output: Provides answers to the worker through the robot's speaker.

[1627] Prompt Sentence Examples

[1628] When asked, "What's next?"

[1629] Recognize when a worker is tired and take that into account when generating text that answers the question "What task needs to be done next?"

[1630] example:

[1631] Question: "What should I do next?"

[1632] (Recognize fatigue): The next thing to do is check your tool kit. Take a short break and don't push yourself too hard.

[1633] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1634] Step 1:

[1635] Voice Input Processing

[1636] Subject: User

[1637] Specific operation: A worker in a factory asks the robot, "What should I do next?"

[1638] Input: Worker's voice

[1639] Data processing or data calculation: Audio is captured by a microphone.

[1640] Output: Captured audio data

[1641] Step 2:

[1642] Sending audio data

[1643] Subject: Terminal (robot)

[1644] Specific operation: The robot sends the captured voice data to the server.

[1645] Input: Audio data

[1646] Data processing or data calculation: The voice data is packetized and sent to the server via the network.

[1647] Output: Audio data sent to the server

[1648] Step 3:

[1649] Speech Recognition Processing

[1650] Subject: Server

[1651] Specific operation: The server converts the voice data into text data using a voice recognition method (for example, Google Speech-to-Text API).

[1652] Input: Audio data

[1653] Data processing or data calculation: Analyzing voice data and converting it into corresponding text data.

[1654] Output: Converted character data

[1655] Step 4:

[1656] Emotion Recognition Processing

[1657] Subject: Server

[1658] Specific operation: The server uses emotion recognition means to analyze the worker's emotions from the text data.

[1659] Input: Character data

[1660] Data processing or data calculation: Analyze emotions using natural language processing techniques and generate emotional information.

[1661] Output: Emotional information

[1662] Step 5:

[1663] Question analysis and answer generation processing

[1664] Subject: Server

[1665] Specific operation: The server uses generative artificial intelligence (e.g., GPT-4) to analyze text data and emotional information and generate an appropriate response.

[1666] Input: Text data and emotion information

[1667] Data processing or data computation: Generate prompts for a generative artificial intelligence model, which then analyzes and generates an answer.

[1668] Output: Generated answer text

[1669] Step 6:

[1670] Response adjustment process

[1671] Subject: Server

[1672] Specific operation: The server adjusts the generated answer appropriately based on the recognized emotion information.

[1673] Input: Generated answer text and sentiment information

[1674] Data processing or data calculation: Refer to the emotional information and add supplements or corrections to the answer text.

[1675] Output: Adjusted answer text

[1676] Step 7:

[1677] Speech synthesis processing

[1678] Subject: Server

[1679] Specific operation: The server converts the adjusted answer text into audio data using a speech synthesis means (e.g., Google Text-to-Speech API).

[1680] Input: Adjusted answer text

[1681] Data processing or data calculation: Text data is input into a speech synthesis engine to generate corresponding speech data.

[1682] Output: Generated audio data

[1683] Step 8:

[1684] Sending audio data

[1685] Subject: Server

[1686] Specific operation: The server sends the generated voice data to the robot.

[1687] Input: Generated audio data

[1688] Data processing or data calculation: The voice data is packetized and sent to the robot via the network.

[1689] Output: Audio data sent to the robot

[1690] Step 9:

[1691] Audio Output Processing

[1692] Subject: Terminal (robot)

[1693] Specific operation: The robot plays back the audio data through a speaker and provides a response to the worker.

[1694] Input: Transmitted audio data

[1695] Data processing or data calculation: Deserializing the audio data and converting it into a format that can be played by the speaker.

[1696] Output: Audio response that can be heard by the worker

[1697] 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.

[1698] 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.

[1699] 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.

[1700] 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.

[1701] 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.

[1702] 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.

[1703] 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).

[1704] 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.

[1705] 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."

[1706] 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.

[1707] 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).

[1708] 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.

[1709] 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.

[1710] 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.

[1711] 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.

[1712] 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.

[1713] 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.

[1714] 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.

[1715] 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.

[1716] 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.

[1717] 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.

[1718] The following is further disclosed regarding the above embodiment.

[1719] (Claim 1)

[1720] A voice input means;

[1721] a speech recognition means for converting speech input data into character data;

[1722] an analysis means for analyzing character data using generative artificial intelligence to generate an answer;

[1723] a speech synthesis means for converting the generated answer into speech;

[1724] an output means for reproducing the voice generated by the voice synthesis means;

[1725] A system including:

[1726] (Claim 2)

[1727] 10. The system of claim 1, further comprising an intent recognition means for inferring a user's intent before transmitting the speech input data to the generative artificial intelligence.

[1728] (Claim 3)

[1729] 2. The system of claim 1, further comprising a search means for retrieving the generated answer from a product knowledge database.

[1730] "Example 1"

[1731] (Claim 1)

[1732] A voice input means;

[1733] a speech recognition means for converting speech input data into character data;

[1734] an analysis means for analyzing character data using generative artificial intelligence to generate an answer;

[1735] a speech synthesis means for converting the generated answer into speech;

[1736] means for transmitting the generated voice data to an output means;

[1737] an output means for reproducing the voice generated by the voice synthesis means;

[1738] A system including:

[1739] (Claim 2)

[1740] 10. The system of claim 1, further comprising an intent recognition means for inferring a user's intent before transmitting the speech input data to the generative artificial intelligence.

[1741] (Claim 3)

[1742] 2. The system of claim 1, further comprising a search means for retrieving the generated answer from a product knowledge database.

[1743] "Application Example 1"

[1744] (Claim 1)

[1745] A voice input means;

[1746] a speech recognition means for converting speech input data into character data;

[1747] an analysis means for analyzing character data using generative artificial intelligence to generate an answer;

[1748] a speech synthesis means for converting the generated answer into speech;

[1749] an output means for reproducing the voice generated by the voice synthesis means;

[1750] means for performing processing using cloud computing technology;

[1751] a means for using the device to capture voice data when a user asks a question by voice in the virtual store;

[1752] A system including:

[1753] (Claim 2)

[1754] 10. The system of claim 1, further comprising an intent recognition means for inferring a user's intent before transmitting the speech input data to the generative artificial intelligence.

[1755] (Claim 3)

[1756] 2. The system of claim 1, further comprising a search means for retrieving the generated answer from a product knowledge database.

[1757] "Example 2: Combining Emotion Engines"

[1758] (Claim 1)

[1759] A voice input means;

[1760] a speech recognition means for converting speech input data into character data;

[1761] an analysis means for analyzing character data using generative artificial intelligence to generate an answer;

[1762] a speech synthesis means for converting the generated answer into speech;

[1763] an output means for playing back the generated audio;

[1764] emotion recognition means for recognizing an emotion of a user and adjusting a response based on the recognized emotion;

[1765] A system including:

[1766] (Claim 2)

[1767] 10. The system of claim 1, further comprising an intent recognition means for inferring a user's intent before transmitting the speech input data to the generative artificial intelligence.

[1768] (Claim 3)

[1769] 2. The system of claim 1, further comprising a search means for retrieving the generated answer from a product knowledge database.

[1770] "Application example 2 when combining emotion engines"

[1771] (Claim 1)

[1772] A voice input means;

[1773] a speech recognition means for converting speech input data into character data;

[1774] an analysis means for analyzing character data using generative artificial intelligence to generate an answer;

[1775] a speech synthesis means for converting the generated answer into speech;

[1776] an output means for reproducing the voice generated by the voice synthesis means;

[1777] emotion recognition means for recognizing a user's emotion and adjusting responses accordingly;

[1778] means for providing voice instructions in cooperation with the device to assist in factory operations;

[1779] A system including:

[1780] (Claim 2)

[1781] 10. The system of claim 1, further comprising an intent recognition means for inferring a user's intent before transmitting the speech input data to the generative artificial intelligence.

[1782] (Claim 3)

[1783] 2. The system of claim 1, further comprising a search means for retrieving the generated answer from a product knowledge database. [Explanation of symbols]

[1784] 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. A voice input means; a speech recognition means for converting speech input data into character data; an analysis means for analyzing character data using generative artificial intelligence to generate an answer; a speech synthesis means for converting the generated answer into speech; an output means for reproducing the voice generated by the voice synthesis means; A system including:

2. 10. The system of claim 1, further comprising an intent recognition means for inferring a user's intent before transmitting the speech input data to the generative artificial intelligence.

3. 2. The system of claim 1, further comprising a search means for retrieving the generated answer from a product knowledge database.

Images