System

Abstract

A system is provided.SOLUTION: The system includes a voice acquiring means for acquiring a surrounding voice in real time, a voice recognizing means for converting the acquired voice into text data, an analyzing means for analyzing the text data by using a generation AI model and identifying a part including a specific term, a comment generating means for generating a comment relating to the specific term, and a comment providing means for providing the generated comment to a user in real time.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In today's business environment and technological development, there are increasing opportunities for people with knowledge in different fields to cooperate. However, the frequent use of technical terms and complex concepts can sometimes cause communication barriers due to differences in expertise. This problem is particularly noticeable in meetings and discussions, hindering smooth communication. Therefore, there is a need for a means to explain technical terms and complex concepts in a simple, easy-to-understand manner in real time. [Means for solving the problem]

[0005] The present invention provides a system including a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating explanatory texts related to the specific terms, and an explanation provision means for providing the generated explanatory texts to users in real time. This system helps people with knowledge in different fields to communicate smoothly by providing users with conversations containing technical terms and complex concepts in an easy-to-understand format in real time.

[0006] "Audio acquisition means" is a general term for devices and software for acquiring surrounding audio in real time.

[0007] "Speech recognition means" is a general term for algorithms and programs for converting acquired speech into text data.

[0008] "Analysis means" is a general term for technologies and functions that use generative AI models to analyze text data and identify parts that contain specific terms.

[0009] "Explanation generation means" is a general term for algorithms and programs for generating explanatory text for specific terms.

[0010] "Explanation providing means" is a general term for devices and software for providing generated explanatory text to users in real time.

[0011] "Generative AI model" is a general term for artificial intelligence models and systems that analyze text data and generate explanatory text based on specific terms and concepts.

[0012] "User" refers to a person who uses the system to receive explanations of technical terms and complex concepts. [Brief explanation of the drawings]

[0013] [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

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0035] First, the device uses an audio device to capture surrounding sounds in real time, collects the audio data, and then transmits the captured audio data to a server via the Internet.

[0036] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0037] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0038] The generated commentary is then sent to the terminal via the Internet. The commentary providing means uses the terminal's audio device to whisper (play low-volume audio) the commentary to the user, allowing the user to understand the content of the technical conversation in real time.

[0039] Specific examples

[0040] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0041] The generated explanation is sent to the device and whispered to the user by the explanation providing means, allowing the user to understand the technical term "deep learning" in real time.

[0042] In this way, the present invention can be a powerful support tool for smooth communication between people with knowledge in different fields. Furthermore, by explaining technical terms in real time, it can improve learning efficiency and comprehension.

[0043] The processing flow will be explained below.

[0044] Step 1:

[0045] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures the audio data.

[0046] Step 2:

[0047] The device sends the acquired voice data to the server. Specifically, the device uploads the voice data to the server via the Internet.

[0048] Step 3:

[0049] The server converts the transmitted voice data into text data using a voice recognition means, which utilizes a highly accurate voice recognition algorithm to convert the voice into text.

[0050] Step 4:

[0051] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts in the text data.

[0052] Step 5:

[0053] The server generates an explanation for the identified technical term using an explanation generation means. For example, if "deep learning" is identified, the generative AI model creates an explanation such as "deep learning is a type of machine learning that uses neural networks."

[0054] Step 6:

[0055] The server transmits the generated commentary to the terminal. Specifically, the server transmits the generated commentary to the terminal via the Internet.

[0056] Step 7:

[0057] The device uses the explanation providing means to whisper the explanation to the user, and the built-in speaker of the audio glasses plays the explanation in a whisper to the user.

[0058] Step 8:

[0059] Users can listen to the commentary in real time and understand the content of the conversation, and can participate in professional conversations by referring to the commentary provided.

[0060] By following these steps in sequence, users can understand the surrounding jargon-laden conversation in real time.

[0061] Example 1

[0062] 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."

[0063] In technical conversations and discussions, there are many technical terms and complex concepts that are difficult for general users to understand. This makes it difficult for people with knowledge in different fields to communicate smoothly, causing problems such as delays in information sharing and decision-making. Furthermore, if understanding of technical terms is not provided in real time during meetings and discussions, users' learning efficiency and level of understanding may decrease.

[0064] 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.

[0065] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative artificial intelligence model and identifying portions containing technical terms, an explanation generation means for generating explanations of the technical terms, and an explanation provision means for providing the generated explanations to users in real time. This allows users to receive explanations of technical terms in real time, enabling information sharing across different fields of expertise to be carried out quickly and effectively.

[0066] "Ambient audio" is audio data including speech and background sounds obtained from the user's surrounding environment.

[0067] "Audio acquisition means" refers to a device or method for acquiring ambient audio in real time, and includes a microphone and an audio device.

[0068] "Speech recognition means" refers to a technology or system that converts acquired voice data into text data, and includes speech recognition software and algorithms.

[0069] A "generative artificial intelligence model" is an artificial intelligence model that learns from given data and automatically generates explanatory text and analysis, and includes machine learning models that use neural networks.

[0070] An "analysis means" is a process or system that analyzes text data and identifies specific technical terms and important concepts, and includes natural language processing techniques.

[0071] The "explanation generation means" is a system or method that generates easy-to-understand explanations for the technical terms and concepts identified by the analysis means.

[0072] The "explanation providing means" is a means for providing the generated explanation to the user, and includes text display and audio output.

[0073] An "audio device" is a device for reproducing sound, including speakers and earphones.

[0074] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0075] First, the device uses an audio device (such as audio glasses) to capture surrounding sounds in real time and collects that audio data. For example, if an engineer is talking about "deep learning" during a meeting, the device will record the conversation.

[0076] The device then transmits the captured audio data over the Internet to a server, where the data is compressed and transmitted using a secure protocol (e.g., HTTPS).

[0077] The server converts the received voice data into text data using a voice recognition means (for example, a general voice recognition service). For example, it converts "Voice during a meeting: 'About deep learning...'" into text data such as "About deep learning...".

[0078] The server then analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis means identifies parts of the text data that contain technical terms and important concepts. For example, it recognizes "deep learning" as a technical term.

[0079] The server's explanation generation means then generates simple explanations for the identified technical terms. For example, for the analyzed term "deep learning," it generates the explanation "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0080] The generated commentary is then sent to the device via the Internet, with care taken to minimize data size and ensure speedy transmission.

[0081] Finally, the terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, thereby enabling the user to understand the content of the specialized conversation in real time.

[0082] For example:

[0083] For example, if an engineer starts talking about "deep learning" during a meeting, the device (audio glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data." The generated explanation will be sent to the device, and the explanation provision means will whisper it to the user.

[0084] Here are some specific prompts that can help users use the system to understand meeting terminology in real time:

[0085] What is Deep Learning?

[0086] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0087] Step 1:

[0088] The terminal uses an audio device to capture surrounding sounds in real time. The audio capture means continuously collects audio data, such as speech during a meeting, using a microphone. The input is the surrounding sounds, and the output is the captured audio data.

[0089] Step 2:

[0090] The device transmits the acquired audio data to a server via the Internet. The transmitted audio data is compressed as a pre-processing step and securely transmitted using a secure protocol (e.g., HTTPS). The input is the acquired audio data, and the output is the audio data transmitted to the server.

[0091] Step 3:

[0092] The server converts the received voice data into text data using a voice recognition method (for example, a general voice recognition service). A voice recognition algorithm analyzes the voice waveform and generates a corresponding string of characters. The input is the voice data received by the server, and the output is the converted text data. For example, voice data such as "Voice during meeting: 'About deep learning...'" is converted into text data such as "About deep learning...".

[0093] Step 4:

[0094] The server analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis method identifies technical terms and important concepts from the text data. The input is the converted text data, and the output is the identified technical terms and concepts. For example, "deep learning" is recognized as a technical term.

[0095] Step 5:

[0096] The server's explanation generation means generates simple explanations for the identified technical terms. A generative artificial intelligence model is used to generate explanations related to the identified technical terms. The input is the identified technical terms, and the output is the generated explanation. For example, for "deep learning," the explanation generated is "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0097] Step 6:

[0098] The server then sends the generated commentary to the terminal via the Internet. The data is kept lightweight and can be sent quickly. The input is the generated commentary, and the output is the commentary sent to the terminal.

[0099] Step 7:

[0100] The terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, allowing the user to understand the content of the specialized conversation in real time. The input is the explanation sent to the terminal, and the output is the explanation provided to the user.

[0101] (Application example 1)

[0102] 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."

[0103] The technical terms and work instructions used in factories can be difficult for unskilled workers to understand. In such cases, a lack of understanding or misunderstanding can have a negative impact on safety and efficiency, so a system is needed that can instantly explain the technical terms and work instructions and help workers understand them.

[0104] 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.

[0105] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating explanations for the specific terms, and an explanation provision means for providing the generated explanations to users in real time. This enables a robot to understand work instructions and technical terms in a factory and provide explanations for them.

[0106] The "audio acquisition means" refers to a device or sensor for acquiring surrounding audio in real time.

[0107] "Speech recognition means" refers to a technology or device that converts acquired speech into text data.

[0108] "Analysis means" refers to a technology or device that uses a generative AI model to analyze text data and identify specific terms or important parts.

[0109] The term "explanation generating means" refers to a technique or device that generates an explanation about a specific term.

[0110] The "explanation providing means" refers to a technique or device that provides the generated explanation to the user.

[0111] A "generative AI model" is an artificial intelligence model that analyzes text data and generates explanatory text.

[0112] A "robot" is an artificially created device that can perform specific tasks automatically.

[0113] This invention is a system mainly composed of a voice acquisition means, a voice recognition means, an analysis means, an explanation generation means, and an explanation provision means. The system aims to help workers understand work instructions and technical terms in a factory by explaining them in real time, thereby improving safety and efficiency. Specific embodiments are as follows.

[0114] First, the robot uses an audio device to capture surrounding sounds in real time and collect the audio data.The robot then transmits the captured audio data to a server via the Internet.In this process, a microphone serves as the audio capture means, and the SpeechRecognition library is used for voice recognition.

[0115] The server converts the received voice data into text data using a speech recognition means. The speech recognition means uses a highly accurate speech recognition algorithm, allowing it to accurately convert conversational voice into text data. The converted text data is analyzed by a generative AI model, which functions as an analysis means. The generative AI model uses OpenAI's API, and the analysis means identifies specific terms and important concepts from the text data.

[0116] Next, the server uses the generative AI model to generate explanations for specific terms from the analyzed text data. The explanation generation means uses advanced natural language processing technology to generate easy-to-understand explanations for technical terms and important concepts. For example, for the term "PLC," the following explanation is generated: "PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[0117] The generated explanatory text is again sent to the robot terminal via the Internet. To provide this explanatory text to the user, the explanation providing means whispers (plays audio at a low volume) using the robot's audio device, allowing the worker to understand immediately. Furthermore, the explanatory text can also be displayed on a display. For example, if a worker in a factory is talking about how to operate a new machine and the term "PLC" is mentioned, the robot will capture the voice, send it to the server, analyze it, generate an explanatory text, and provide the following explanation:

[0118] "A PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[0119] In this way, users can easily understand specific technical terms and work instructions in real time. Implementing this invention provides a powerful support tool for smooth communication between people with different fields of expertise, which is expected to result in improved learning efficiency and comprehension.

[0120] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0121] Step 1:

[0122] The device uses an audio device to capture surrounding sounds in real time.

[0123] Input: Ambient audio

[0124] Data processing: Collected as audio data using an audio device

[0125] Output: Audio data

[0126] Step 2:

[0127] The terminal transmits the acquired voice data to a server via the Internet.

[0128] Input: Audio data

[0129] Data calculation: Converts voice data into packet data for transmission

[0130] Output: Packet data to the server

[0131] Step 3:

[0132] The server converts the received voice data into text data using a voice recognition means.

[0133] Input: Audio data

[0134] Data calculation: Convert to text data using a speech recognition algorithm (SpeechRecognition library)

[0135] Output: Text data

[0136] Step 4:

[0137] The server uses a generative AI model to analyze the text data and identify specific terms and key concepts.

[0138] Input: Text data

[0139] Data Computation: Analysis and term identification using generative AI models (OpenAI API)

[0140] Output: Identified terms or key concepts

[0141] Step 5:

[0142] The server generates a description for the identified term.

[0143] Input: Identified terms or key concepts

[0144] Data calculation: Generating explanatory text based on natural language processing (generative AI model)

[0145] Output:Explanation

[0146] Example: Example prompt: "Please explain the following technical term: PLC"

[0147] Step 6:

[0148] The server transmits the generated commentary to the terminal via the Internet.

[0149] Input:Explanation

[0150] Data calculation: Converts into packet data for sending explanatory text

[0151] Output: Packet data to the terminal

[0152] Step 7:

[0153] The explanation providing means of the terminal provides the explanation text to the user in real time.

[0154] Input:Explanation

[0155] Data calculation: Low volume voice playback or text display on the display

[0156] Output: Providing a description as audio or text

[0157] This allows the device to provide users with real-time explanations of technical terms and work instructions to aid understanding.

[0158] 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.

[0159] The present invention is a system that captures ambient audio, converts it into text data, analyzes technical terms and complex concepts using a generative AI model, generates explanatory text, and provides it to users. The system also incorporates an emotion engine that recognizes the user's emotions and adjusts the explanatory text based on the user's emotional state.

[0160] First, the device uses an audio device to capture surrounding sounds in real time and collect the audio data, which is then transmitted to a server via the Internet.

[0161] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0162] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0163] The generated commentary is then sent to the terminal via the Internet. The commentary providing means whispers the commentary to the user using an audio device, specifically audio glasses. The user can listen to the commentary in real time and understand the technical conversation content.

[0164] Furthermore, the system incorporates an emotion engine that can adjust the commentary according to the user's emotional state. The emotion engine recognizes emotions by analyzing the user's tone of voice and facial expressions. Specifically, if the user is feeling unsure or confused, the system can simplify the commentary and add more detailed explanations. If the system determines that the user understands, it can omit or simplify the commentary.

[0165] Specific examples

[0166] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0167] The generated commentary is sent to the device and whispered to the user by the commentary provider. The emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information is provided, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network."

[0168] Conversely, if the emotion engine determines that the user understands, it will provide only a concise explanation, such as "Deep learning is a type of machine learning."

[0169] In this way, the present invention can provide explanations flexibly according to the user's level of understanding and emotional state, which is very useful for understanding technical conversation content in real time.

[0170] The processing flow will be explained below.

[0171] Step 1:

[0172] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures audio data.

[0173] Step 2:

[0174] The device sends the acquired voice data to the server, which then uploads the voice data to the server via the Internet.

[0175] Step 3:

[0176] The server uses a speech recognition means to convert the transmitted voice data into text data, and applies a speech recognition algorithm to analyze the voice data as text information.

[0177] Step 4:

[0178] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts within the text data.

[0179] Step 5:

[0180] The server generates an explanation for the identified technical term using an explanation generation means. Specifically, the generative AI model creates a detailed explanation for the term "deep learning" identified.

[0181] Step 6:

[0182] The server sends the generated commentary to the terminal, and the server sends the commentary to the terminal via the Internet.

[0183] Step 7:

[0184] The terminal uses the explanation providing means to whisper the generated explanation to the user, and the built-in speaker of the audio glasses plays the explanation and whispers it to the user.

[0185] Step 8:

[0186] The device uses a built-in emotion engine to recognize the user's emotional state. Specifically, the emotion engine analyzes the user's tone of voice and facial expressions to identify the user's emotional state.

[0187] Step 9:

[0188] The server receives feedback from the emotion engine and adjusts the explanation based on the user's emotional state, for example, making the explanation more detailed if the user is confused, and more concise if the user understands.

[0189] Step 10:

[0190] The device will then provide the adjusted commentary to the user again, and will whisper the updated commentary to the user using the device's audio device.

[0191] By performing these steps consecutively, users can not only more easily understand the technical terms used in conversations, but also respond flexibly according to their individual levels of understanding and emotional state.

[0192] Example 2

[0193] 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."

[0194] In an environment where technical conversations and terminology are commonplace, users often lack the necessary expertise to fully understand the content of the conversation. While there is a demand for flexible explanations that adapt to the user's emotional state, no such systems currently exist. A method is needed to provide explanations in real time and adjust the explanations according to the user's level of understanding and emotions.

[0195] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating an explanation for the specific term, an explanation provision means for providing the generated explanation to the user in real time, and an emotion recognition means for recognizing the user's emotional state and adjusting the explanation. This makes it possible to understand the content of specialized conversations in real time and provide flexible explanations according to the user's emotional state.

[0196] The "audio acquisition means" is a device or system for acquiring ambient audio in real time.

[0197] "Speech recognition means" refers to an algorithm or system for converting captured speech into text data.

[0198] A "generative AI model" is a model that uses artificial intelligence to process and analyze text data.

[0199] "Analysis means" refers to a means for analyzing text data using a generative AI model and identifying portions containing specific terms.

[0200] The "explanation generating means" is a means for generating an explanation about a specific term.

[0201] The "explanation providing means" is a means for providing the user with an explanation in real time.

[0202] The "emotion recognition means" is a means for recognizing the user's emotional state and adjusting the commentary accordingly.

[0203] The present invention is a system for understanding specialized conversation content in real time, and specifically includes a voice acquisition means, a voice recognition means, a generative AI model, an analysis means, an explanation generation means, an explanation provision means, and an emotion recognition means.

[0204] Audio acquisition means

[0205] The device uses an audio device (such as audio glasses) to capture surrounding sounds in real time. A dedicated microphone built into the device is equipped with noise-canceling technology to remove ambient noise.

[0206] Voice recognition means

[0207] The server receives the voice data sent from the device and converts it into text data using a highly accurate voice recognition algorithm (for example, Google Cloud Speech-to-Text API).

[0208] Generative AI models and analytical methods

[0209] The server analyzes the text data using a generative AI model (e.g., GPT-4). The analysis method identifies sections containing technical terms and important concepts and generates plain-language explanations for those sections.

[0210] Explanation generation method

[0211] The server generates explanatory text based on the analysis. The generated explanatory text explains technical terms in an easy-to-understand format and is provided to users.

[0212] Means of providing explanations

[0213] The generated commentary is sent to the device via the Internet, and the device whispers the commentary to the user using a commentary providing means (e.g., audio glasses). During this process, the text is converted into speech using speech synthesis technology (e.g., Amazon Polly).

[0214] emotion recognition means

[0215] The emotion engine analyzes the user's tone of voice and facial expressions to recognize their emotional state, allowing the commentary to be tailored to the user's level of understanding and emotion. The emotion recognition method uses the Facial Emotion Recognition (FER) API and voice analysis tools.

[0216] Specific examples

[0217] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0218] The generated explanation is sent to the device and whispered to the user by the explanation provision means. At this time, the emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information such as "Deep learning uses specific algorithms, one of which is a convolutional neural network" is provided. Conversely, if the emotion engine determines that the user understands, only a concise explanation is provided. A simplified explanation such as "Deep learning is a type of machine learning" is provided.

[0219] Prompt Sentence Examples

[0220] "Analyze the conversational content obtained from the audio data and generate concise explanations for technical terms. For example, if we talk about 'deep learning,' please provide a definition and a brief explanation."

[0221] "Try to adapt your explanation to take into account the user's emotional state. If the user is confused, provide additional details. If they understand, provide a concise explanation."

[0222] In this way, the present invention is a system that flexibly provides explanatory text according to the user's level of understanding and emotional state, and is extremely useful for understanding specialized conversation content in real time.

[0223] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0224] Step 1:

[0225] Acquiring audio data

[0226] Description: The device uses the audio device to capture ambient sounds in real time.

[0227] Input: An audio capture method receives ambient sound in real time.

[0228] Output: The captured audio data.

[0229] How it works: The device's built-in microphone collects audio from within the conference room and uses noise-canceling technology to remove it.

[0230] Step 2:

[0231] Sending audio data

[0232] Description: The audio data acquired by the device is sent to a server via the Internet.

[0233] Input: Audio data stored in device storage.

[0234] Output: The audio data sent to the server.

[0235] Specific operation: The device encodes the voice data using a security protocol (SSL / TLS) and transmits it over the network to the server.

[0236] Step 3:

[0237] Converting audio data to text

[0238] Description: The server converts the received voice data into text data using a highly accurate voice recognition algorithm.

[0239] Input: The audio data received by the server.

[0240] Output: The converted text data.

[0241] What it does: The server runs a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API) and generates text data called "deep learning."

[0242] Step 4:

[0243] Text data analysis

[0244] Description: The server analyzes text data using a generative AI model.

[0245] Input: The converted text data.

[0246] Output: Parsed data, terminology identified.

[0247] What happens: The server uses a generative AI model (e.g., GPT-4) to analyze the text and identify the term "deep learning."

[0248] Step 5:

[0249] Generate explanatory text

[0250] Description: The server generates a description based on the analysis.

[0251] Input: Parsed data, identified terminology.

[0252] Output: The generated description.

[0253] Specific operation: The server generates an explanatory text such as, "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0254] Step 6:

[0255] Submit commentary

[0256] Description: Sends a server-generated explanatory text to the device.

[0257] Input: The generated description.

[0258] Output: A description sent to the terminal.

[0259] Specific operation: The server encodes the generated commentary and sends it to the terminal via the Internet.

[0260] Step 7:

[0261] Providing explanatory text

[0262] Description: Provides the user with a voice commentary of the information received by the device.

[0263] Input: The explanatory text received on the terminal.

[0264] Output: The audio description provided to the user.

[0265] Specific operation: The device uses speech synthesis technology (e.g., Amazon Polly) to convert the explanatory text into speech and whisper it to the user through the audio glasses.

[0266] Step 8:

[0267] Recognizing the user's emotional state

[0268] Description: Emotion engine recognizes the user's emotional state in real time.

[0269] Input: User's voice tone and facial expression data.

[0270] Output: Perceived emotional state of the user.

[0271] How it works: The emotion engine uses the Facial Emotion Recognition API and voice analysis tools to analyze the user's facial expressions and voice tone to determine their emotional state, such as confusion or understanding.

[0272] Step 9:

[0273] Adjusted explanatory text

[0274] Description: Adjusts explanatory text based on perceived emotional state.

[0275] Input: User's emotional state, initial generated commentary.

[0276] Output: Adjusted explanatory text.

[0277] What it does: If the emotion engine determines that the user is confused, it will provide additional information, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network." Conversely, if the user understands, it will provide a concise explanation, "Deep learning is a type of machine learning."

[0278] (Application example 2)

[0279] 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."

[0280] In traditional brick-and-mortar stores, it can be difficult for store clerks to provide accurate explanations to customers when technical terms or complex concepts are discussed. This can be particularly difficult when the clerk lacks specialized knowledge or the customer does not understand the technical terms. Furthermore, no system has existed to date that can provide explanations tailored to the customer's level of understanding and emotional state.

[0281] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0282] In this invention, the server

[0283] an audio acquisition means for acquiring surrounding audio in real time;

[0284] A speech recognition means for converting the acquired speech into text data;

[0285] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[0286] Explanation generation means for generating an explanation for a specific term;

[0287] an emotion recognition means for recognizing the emotional state of a user and adjusting the commentary;

[0288] An explanation providing means for providing the generated explanation to the user in real time by voice through smart glasses;

[0289] This will enable store staff and customers in physical stores to explain technical terms and complex concepts in real time, in line with the user's emotional state.

[0290] The "audio acquisition means" is a device or system for collecting ambient audio in real time.

[0291] "Speech recognition means" refers to a technique or algorithm for converting acquired voice data into text data.

[0292] A "generative AI model" is a model that uses artificial intelligence technology to analyze text data and identify specific patterns and terms.

[0293] An "analysis means" is a device or system that analyzes text data and extracts specific terms or important parts from it.

[0294] An "explanation generation means" is a device or system for generating an explanation about a specific term or concept.

[0295] An "emotion recognition means" is a device or system for recognizing a user's emotional state and adjusting commentary based on that information.

[0296] The "explanation providing means" is a device or system for providing the generated commentary to the user in real time.

[0297] "Smart glasses" are wearable devices that provide users with visual or audio information when worn by the user.

[0298] An "audio device" is an electronic device for inputting and outputting sound.

[0299] A "brick and mortar store" is a store where goods or services are sold at a physical location.

[0300] The present invention is a system that supports customer service in brick-and-mortar stores by capturing surrounding voices in real time, generating explanatory text for technical terms and complex concepts, and providing it to users. This system includes a voice capture means, a voice recognition means, an analysis means using a generative AI model, an explanation generation means, an emotion recognition means, and an explanation provision means. Furthermore, by using smart glasses, the explanatory text can be provided in real time by voice.

[0301] 1. Audio acquisition method

[0302] The device captures ambient sounds in real time using smart glasses with built-in microphones.

[0303] 2. Voice Recognition Method

[0304] The captured voice is converted into text data by a speech recognition means, which uses a highly accurate speech recognition algorithm, such as Google's speech recognition service.

[0305] 3. Analysis method

[0306] The converted text data is then analyzed by an analysis means using a generative AI model, which identifies specific terms from the text data and generates corresponding explanatory text.

[0307] 4. Explanation Generation Method

[0308] Based on the analyzed data, explanations for specific terms and concepts are generated and presented to users in a format that is easy to understand.

[0309] 5. Emotion recognition means

[0310] The system recognizes the user's emotional state when providing commentary. The emotion recognition method analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. Specifically, the EmotionRecognizer API is used.

[0311] 6. Means of providing explanations

[0312] The generated explanations are provided to users in real time as audio through the smart glasses, allowing store staff to provide appropriate explanations to customers about technical terms and complex concepts.

[0313] Examples:

[0314] For example, if a store clerk explains, "This product is equipped with a new biometric sensor," the smart glasses capture this speech and send it to a server. The server converts the speech into text and identifies the term "biometric sensor." Using a generative AI model, an explanation is generated, such as, "A biometric sensor is a technology that measures and recognizes physical characteristics." If the emotion recognition method determines that the user is confused, a more detailed explanation is added. For example, specific examples such as "This includes fingerprint and facial recognition" are provided.

[0315] Example prompt sentence:

[0316] My customer seems confused. Can you explain more about biometric sensors?

[0317] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0318] Step 1:

[0319] The device captures the surrounding audio in real time. Here, the microphone built into the smart glasses is used. The input is the surrounding audio, and the output is audio data.

[0320] Step 2:

[0321] The voice data acquired by the device is sent to a server via the Internet. The input is the voice data, and the output is the voice data sent to the server.

[0322] Step 3:

[0323] The server converts the received voice data into text data using a voice recognition means. Here, a voice recognition algorithm (e.g., Google's voice recognition service) is used. The input is voice data, and the output is text data.

[0324] Step 4:

[0325] The server analyzes the text data using a generative AI model. The analysis means identifies specific terms from the text data. The input is the text data, and the output is the analysis results that include the specific terms.

[0326] Step 5:

[0327] The server generates an explanatory text using an explanatory text generation means based on the analysis results. The input is the analysis results, and the output is the explanatory text.

[0328] Step 6:

[0329] The server analyzes the user's emotional state using an emotion recognition means. The emotion recognition means analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. The input is the user's tone of voice and facial expressions, and the output is the analysis result of the user's emotional state.

[0330] Step 7:

[0331] The server adjusts the commentary based on the analysis of the emotional state. If the user is confused, it adds a detailed explanation, and if the user understands, it simplifies it. The input is the analysis of the emotional state and the commentary, and the output is the adjusted commentary.

[0332] Step 8:

[0333] The server sends the generated commentary to the terminal. The input is the adjusted commentary, and the output is the commentary sent to the smart glasses.

[0334] Step 9:

[0335] The terminal uses the commentary providing means to provide the commentary text to the user in real time via an audio device of the smart glasses, the input being the adjusted commentary text and the output being the audio commentary provided to the user.

[0336] 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.

[0337] 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.

[0338] 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.

[0339] [Second embodiment]

[0340] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0341] 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.

[0342] 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).

[0343] 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.

[0344] 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.

[0345] 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).

[0346] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0347] 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.

[0348] 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.

[0349] 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.

[0350] 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.

[0351] 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."

[0352] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0353] First, the device uses an audio device to capture surrounding sounds in real time, collects the audio data, and then transmits the captured audio data to a server via the Internet.

[0354] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0355] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0356] The generated commentary is then sent to the terminal via the Internet. The commentary providing means uses the terminal's audio device to whisper (play low-volume audio) the commentary to the user, allowing the user to understand the content of the technical conversation in real time.

[0357] Specific examples

[0358] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0359] The generated explanation is sent to the device and whispered to the user by the explanation providing means, allowing the user to understand the technical term "deep learning" in real time.

[0360] In this way, the present invention can be a powerful support tool for smooth communication between people with knowledge in different fields. Furthermore, by explaining technical terms in real time, it can improve learning efficiency and comprehension.

[0361] The processing flow will be explained below.

[0362] Step 1:

[0363] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures the audio data.

[0364] Step 2:

[0365] The device sends the acquired voice data to the server. Specifically, the device uploads the voice data to the server via the Internet.

[0366] Step 3:

[0367] The server converts the transmitted voice data into text data using a voice recognition means, which utilizes a highly accurate voice recognition algorithm to convert the voice into text.

[0368] Step 4:

[0369] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts in the text data.

[0370] Step 5:

[0371] The server generates an explanation for the identified technical term using an explanation generation means. For example, if "deep learning" is identified, the generative AI model creates an explanation such as "deep learning is a type of machine learning that uses neural networks."

[0372] Step 6:

[0373] The server transmits the generated commentary to the terminal. Specifically, the server transmits the generated commentary to the terminal via the Internet.

[0374] Step 7:

[0375] The device uses the explanation providing means to whisper the explanation to the user, and the built-in speaker of the audio glasses plays the explanation in a whisper to the user.

[0376] Step 8:

[0377] Users can listen to the commentary in real time and understand the content of the conversation, and can participate in professional conversations by referring to the commentary provided.

[0378] By following these steps in sequence, users can understand the surrounding jargon-laden conversation in real time.

[0379] Example 1

[0380] 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."

[0381] In technical conversations and discussions, there are many technical terms and complex concepts that are difficult for general users to understand. This makes it difficult for people with knowledge in different fields to communicate smoothly, causing problems such as delays in information sharing and decision-making. Furthermore, if understanding of technical terms is not provided in real time during meetings and discussions, users' learning efficiency and level of understanding may decrease.

[0382] 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.

[0383] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative artificial intelligence model and identifying portions containing technical terms, an explanation generation means for generating explanations of the technical terms, and an explanation provision means for providing the generated explanations to users in real time. This allows users to receive explanations of technical terms in real time, enabling information sharing across different fields of expertise to be carried out quickly and effectively.

[0384] "Ambient audio" is audio data including speech and background sounds obtained from the user's surrounding environment.

[0385] "Audio acquisition means" refers to a device or method for acquiring ambient audio in real time, and includes a microphone and an audio device.

[0386] "Speech recognition means" refers to a technology or system that converts acquired voice data into text data, and includes speech recognition software and algorithms.

[0387] A "generative artificial intelligence model" is an artificial intelligence model that learns from given data and automatically generates explanatory text and analysis, and includes machine learning models that use neural networks.

[0388] An "analysis means" is a process or system that analyzes text data and identifies specific technical terms and important concepts, and includes natural language processing techniques.

[0389] The "explanation generation means" is a system or method that generates easy-to-understand explanations for the technical terms and concepts identified by the analysis means.

[0390] The "explanation providing means" is a means for providing the generated explanation to the user, and includes text display and audio output.

[0391] An "audio device" is a device for reproducing sound, including speakers and earphones.

[0392] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0393] First, the device uses an audio device (such as audio glasses) to capture surrounding sounds in real time and collects that audio data. For example, if an engineer is talking about "deep learning" during a meeting, the device will record the conversation.

[0394] The device then transmits the captured audio data over the Internet to a server, where the data is compressed and transmitted using a secure protocol (e.g., HTTPS).

[0395] The server converts the received voice data into text data using a voice recognition means (for example, a general voice recognition service). For example, it converts "Voice during a meeting: 'About deep learning...'" into text data such as "About deep learning...".

[0396] The server then analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis means identifies parts of the text data that contain technical terms and important concepts. For example, it recognizes "deep learning" as a technical term.

[0397] The server's explanation generation means then generates simple explanations for the identified technical terms. For example, for the analyzed term "deep learning," it generates the explanation "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0398] The generated commentary is then sent to the device via the Internet, with care taken to minimize data size and ensure speedy transmission.

[0399] Finally, the terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, thereby enabling the user to understand the content of the specialized conversation in real time.

[0400] For example:

[0401] For example, if an engineer starts talking about "deep learning" during a meeting, the device (audio glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data." The generated explanation will be sent to the device, and the explanation provision means will whisper it to the user.

[0402] Here are some specific prompts that can help users use the system to understand meeting terminology in real time:

[0403] What is Deep Learning?

[0404] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0405] Step 1:

[0406] The terminal uses an audio device to capture surrounding sounds in real time. The audio capture means continuously collects audio data, such as speech during a meeting, using a microphone. The input is the surrounding sounds, and the output is the captured audio data.

[0407] Step 2:

[0408] The device transmits the acquired audio data to a server via the Internet. The transmitted audio data is compressed as a pre-processing step and securely transmitted using a secure protocol (e.g., HTTPS). The input is the acquired audio data, and the output is the audio data transmitted to the server.

[0409] Step 3:

[0410] The server converts the received voice data into text data using a voice recognition method (for example, a general voice recognition service). A voice recognition algorithm analyzes the voice waveform and generates a corresponding string of characters. The input is the voice data received by the server, and the output is the converted text data. For example, voice data such as "Voice during meeting: 'About deep learning...'" is converted into text data such as "About deep learning...".

[0411] Step 4:

[0412] The server analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis method identifies technical terms and important concepts from the text data. The input is the converted text data, and the output is the identified technical terms and concepts. For example, "deep learning" is recognized as a technical term.

[0413] Step 5:

[0414] The server's explanation generation means generates simple explanations for the identified technical terms. A generative artificial intelligence model is used to generate explanations related to the identified technical terms. The input is the identified technical terms, and the output is the generated explanation. For example, for "deep learning," the explanation generated is "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0415] Step 6:

[0416] The server then sends the generated commentary to the terminal via the Internet. The data is kept lightweight and can be sent quickly. The input is the generated commentary, and the output is the commentary sent to the terminal.

[0417] Step 7:

[0418] The terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, allowing the user to understand the content of the specialized conversation in real time. The input is the explanation sent to the terminal, and the output is the explanation provided to the user.

[0419] (Application example 1)

[0420] 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."

[0421] The technical terms and work instructions used in factories can be difficult for unskilled workers to understand. In such cases, a lack of understanding or misunderstanding can have a negative impact on safety and efficiency, so a system is needed that can instantly explain the technical terms and work instructions and help workers understand them.

[0422] 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.

[0423] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating explanations for the specific terms, and an explanation provision means for providing the generated explanations to users in real time. This enables a robot to understand work instructions and technical terms in a factory and provide explanations for them.

[0424] The "audio acquisition means" refers to a device or sensor for acquiring surrounding audio in real time.

[0425] "Speech recognition means" refers to a technology or device that converts acquired speech into text data.

[0426] "Analysis means" refers to a technology or device that uses a generative AI model to analyze text data and identify specific terms or important parts.

[0427] The term "explanation generating means" refers to a technique or device that generates an explanation about a specific term.

[0428] The "explanation providing means" refers to a technique or device that provides the generated explanation to the user.

[0429] A "generative AI model" is an artificial intelligence model that analyzes text data and generates explanatory text.

[0430] A "robot" is an artificially created device that can perform specific tasks automatically.

[0431] This invention is a system mainly composed of a voice acquisition means, a voice recognition means, an analysis means, an explanation generation means, and an explanation provision means. The system aims to help workers understand work instructions and technical terms in a factory by explaining them in real time, thereby improving safety and efficiency. Specific embodiments are as follows.

[0432] First, the robot uses an audio device to capture surrounding sounds in real time and collect the audio data.The robot then transmits the captured audio data to a server via the Internet.In this process, a microphone serves as the audio capture means, and the SpeechRecognition library is used for voice recognition.

[0433] The server converts the received voice data into text data using a speech recognition means. The speech recognition means uses a highly accurate speech recognition algorithm, allowing it to accurately convert conversational voice into text data. The converted text data is analyzed by a generative AI model, which functions as an analysis means. The generative AI model uses OpenAI's API, and the analysis means identifies specific terms and important concepts from the text data.

[0434] Next, the server uses the generative AI model to generate explanations for specific terms from the analyzed text data. The explanation generation means uses advanced natural language processing technology to generate easy-to-understand explanations for technical terms and important concepts. For example, for the term "PLC," the following explanation is generated: "PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[0435] The generated explanatory text is again sent to the robot terminal via the Internet. To provide this explanatory text to the user, the explanation providing means whispers (plays audio at a low volume) using the robot's audio device, allowing the worker to understand immediately. Furthermore, the explanatory text can also be displayed on a display. For example, if a worker in a factory is talking about how to operate a new machine and the term "PLC" is mentioned, the robot will capture the voice, send it to the server, analyze it, generate an explanatory text, and provide the following explanation:

[0436] "A PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[0437] In this way, users can easily understand specific technical terms and work instructions in real time. Implementing this invention provides a powerful support tool for smooth communication between people with different fields of expertise, which is expected to result in improved learning efficiency and comprehension.

[0438] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0439] Step 1:

[0440] The device uses an audio device to capture surrounding sounds in real time.

[0441] Input: Ambient audio

[0442] Data processing: Collected as audio data using an audio device

[0443] Output: Audio data

[0444] Step 2:

[0445] The terminal transmits the acquired voice data to a server via the Internet.

[0446] Input: Audio data

[0447] Data calculation: Converts voice data into packet data for transmission

[0448] Output: Packet data to the server

[0449] Step 3:

[0450] The server converts the received voice data into text data using a voice recognition means.

[0451] Input: Audio data

[0452] Data calculation: Convert to text data using a speech recognition algorithm (SpeechRecognition library)

[0453] Output: Text data

[0454] Step 4:

[0455] The server uses a generative AI model to analyze the text data and identify specific terms and key concepts.

[0456] Input: Text data

[0457] Data Computation: Analysis and term identification using generative AI models (OpenAI API)

[0458] Output: Identified terms or key concepts

[0459] Step 5:

[0460] The server generates a description for the identified term.

[0461] Input: Identified terms or key concepts

[0462] Data calculation: Generating explanatory text based on natural language processing (generative AI model)

[0463] Output:Explanation

[0464] Example: Example prompt: "Please explain the following technical term: PLC"

[0465] Step 6:

[0466] The server transmits the generated commentary to the terminal via the Internet.

[0467] Input:Explanation

[0468] Data calculation: Converts into packet data for sending explanatory text

[0469] Output: Packet data to the terminal

[0470] Step 7:

[0471] The explanation providing means of the terminal provides the explanation text to the user in real time.

[0472] Input:Explanation

[0473] Data calculation: Low volume voice playback or text display on the display

[0474] Output: Providing a description as audio or text

[0475] This allows the device to provide users with real-time explanations of technical terms and work instructions to aid understanding.

[0476] 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.

[0477] The present invention is a system that captures ambient audio, converts it into text data, analyzes technical terms and complex concepts using a generative AI model, generates explanatory text, and provides it to users. The system also incorporates an emotion engine that recognizes the user's emotions and adjusts the explanatory text based on the user's emotional state.

[0478] First, the device uses an audio device to capture surrounding sounds in real time and collect the audio data, which is then transmitted to a server via the Internet.

[0479] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0480] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0481] The generated commentary is then sent to the terminal via the Internet. The commentary providing means whispers the commentary to the user using an audio device, specifically audio glasses. The user can listen to the commentary in real time and understand the technical conversation content.

[0482] Furthermore, the system incorporates an emotion engine that can adjust the commentary according to the user's emotional state. The emotion engine recognizes emotions by analyzing the user's tone of voice and facial expressions. Specifically, if the user is feeling unsure or confused, the system can simplify the commentary and add more detailed explanations. If the system determines that the user understands, it can omit or simplify the commentary.

[0483] Specific examples

[0484] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0485] The generated commentary is sent to the device and whispered to the user by the commentary provider. The emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information is provided, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network."

[0486] Conversely, if the emotion engine determines that the user understands, it will provide only a concise explanation, such as "Deep learning is a type of machine learning."

[0487] In this way, the present invention can provide explanations flexibly according to the user's level of understanding and emotional state, which is very useful for understanding technical conversation content in real time.

[0488] The processing flow will be explained below.

[0489] Step 1:

[0490] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures audio data.

[0491] Step 2:

[0492] The device sends the acquired voice data to the server, which then uploads the voice data to the server via the Internet.

[0493] Step 3:

[0494] The server uses a speech recognition means to convert the transmitted voice data into text data, and applies a speech recognition algorithm to analyze the voice data as text information.

[0495] Step 4:

[0496] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts within the text data.

[0497] Step 5:

[0498] The server generates an explanation for the identified technical term using an explanation generation means. Specifically, the generative AI model creates a detailed explanation for the term "deep learning" identified.

[0499] Step 6:

[0500] The server sends the generated commentary to the terminal, and the server sends the commentary to the terminal via the Internet.

[0501] Step 7:

[0502] The terminal uses the explanation providing means to whisper the generated explanation to the user, and the built-in speaker of the audio glasses plays the explanation and whispers it to the user.

[0503] Step 8:

[0504] The device uses a built-in emotion engine to recognize the user's emotional state. Specifically, the emotion engine analyzes the user's tone of voice and facial expressions to identify the user's emotional state.

[0505] Step 9:

[0506] The server receives feedback from the emotion engine and adjusts the explanation based on the user's emotional state, for example, making the explanation more detailed if the user is confused, and more concise if the user understands.

[0507] Step 10:

[0508] The device will then provide the adjusted commentary to the user again, and will whisper the updated commentary to the user using the device's audio device.

[0509] By performing these steps consecutively, users can not only more easily understand the technical terms used in conversations, but also respond flexibly according to their individual levels of understanding and emotional state.

[0510] Example 2

[0511] 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."

[0512] In an environment where technical conversations and terminology are commonplace, users often lack the necessary expertise to fully understand the content of the conversation. While there is a demand for flexible explanations that adapt to the user's emotional state, no such systems currently exist. A method is needed to provide explanations in real time and adjust the explanations according to the user's level of understanding and emotions.

[0513] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating an explanation for the specific term, an explanation provision means for providing the generated explanation to the user in real time, and an emotion recognition means for recognizing the user's emotional state and adjusting the explanation. This makes it possible to understand the content of specialized conversations in real time and provide flexible explanations according to the user's emotional state.

[0514] The "audio acquisition means" is a device or system for acquiring ambient audio in real time.

[0515] "Speech recognition means" refers to an algorithm or system for converting captured speech into text data.

[0516] A "generative AI model" is a model that uses artificial intelligence to process and analyze text data.

[0517] "Analysis means" refers to a means for analyzing text data using a generative AI model and identifying portions containing specific terms.

[0518] The "explanation generating means" is a means for generating an explanation about a specific term.

[0519] The "explanation providing means" is a means for providing the user with an explanation in real time.

[0520] The "emotion recognition means" is a means for recognizing the user's emotional state and adjusting the commentary accordingly.

[0521] The present invention is a system for understanding specialized conversation content in real time, and specifically includes a voice acquisition means, a voice recognition means, a generative AI model, an analysis means, an explanation generation means, an explanation provision means, and an emotion recognition means.

[0522] Audio acquisition means

[0523] The device uses an audio device (such as audio glasses) to capture surrounding sounds in real time. A dedicated microphone built into the device is equipped with noise-canceling technology to remove ambient noise.

[0524] Voice recognition means

[0525] The server receives the voice data sent from the device and converts it into text data using a highly accurate voice recognition algorithm (for example, Google Cloud Speech-to-Text API).

[0526] Generative AI models and analytical methods

[0527] The server analyzes the text data using a generative AI model (e.g., GPT-4). The analysis method identifies sections containing technical terms and important concepts and generates plain-language explanations for those sections.

[0528] Explanation generation method

[0529] The server generates explanatory text based on the analysis. The generated explanatory text explains technical terms in an easy-to-understand format and is provided to users.

[0530] Means of providing explanations

[0531] The generated commentary is sent to the device via the Internet, and the device whispers the commentary to the user using a commentary providing means (e.g., audio glasses). During this process, the text is converted into speech using speech synthesis technology (e.g., Amazon Polly).

[0532] emotion recognition means

[0533] The emotion engine analyzes the user's tone of voice and facial expressions to recognize their emotional state, allowing the commentary to be tailored to the user's level of understanding and emotion. The emotion recognition method uses the Facial Emotion Recognition (FER) API and voice analysis tools.

[0534] Specific examples

[0535] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0536] The generated explanation is sent to the device and whispered to the user by the explanation provision means. At this time, the emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information such as "Deep learning uses specific algorithms, one of which is a convolutional neural network" is provided. Conversely, if the emotion engine determines that the user understands, only a concise explanation is provided. A simplified explanation such as "Deep learning is a type of machine learning" is provided.

[0537] Prompt Sentence Examples

[0538] "Analyze the conversational content obtained from the audio data and generate concise explanations for technical terms. For example, if we talk about 'deep learning,' please provide a definition and a brief explanation."

[0539] "Try to adapt your explanation to take into account the user's emotional state. If the user is confused, provide additional details. If they understand, provide a concise explanation."

[0540] In this way, the present invention is a system that flexibly provides explanatory text according to the user's level of understanding and emotional state, and is extremely useful for understanding specialized conversation content in real time.

[0541] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0542] Step 1:

[0543] Acquiring audio data

[0544] Description: The device uses the audio device to capture ambient sounds in real time.

[0545] Input: An audio capture method receives ambient sound in real time.

[0546] Output: The captured audio data.

[0547] How it works: The device's built-in microphone collects audio from within the conference room and uses noise-canceling technology to remove it.

[0548] Step 2:

[0549] Sending audio data

[0550] Description: The audio data acquired by the device is sent to a server via the Internet.

[0551] Input: Audio data stored in device storage.

[0552] Output: The audio data sent to the server.

[0553] Specific operation: The device encodes the voice data using a security protocol (SSL / TLS) and transmits it over the network to the server.

[0554] Step 3:

[0555] Converting audio data to text

[0556] Description: The server converts the received voice data into text data using a highly accurate voice recognition algorithm.

[0557] Input: The audio data received by the server.

[0558] Output: The converted text data.

[0559] What it does: The server runs a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API) and generates text data called "deep learning."

[0560] Step 4:

[0561] Text data analysis

[0562] Description: The server analyzes text data using a generative AI model.

[0563] Input: The converted text data.

[0564] Output: Parsed data, terminology identified.

[0565] What happens: The server uses a generative AI model (e.g., GPT-4) to analyze the text and identify the term "deep learning."

[0566] Step 5:

[0567] Generate explanatory text

[0568] Description: The server generates a description based on the analysis.

[0569] Input: Parsed data, identified terminology.

[0570] Output: The generated description.

[0571] Specific operation: The server generates an explanatory text such as, "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0572] Step 6:

[0573] Submit commentary

[0574] Description: Sends a server-generated explanatory text to the device.

[0575] Input: The generated description.

[0576] Output: A description sent to the terminal.

[0577] Specific operation: The server encodes the generated commentary and sends it to the terminal via the Internet.

[0578] Step 7:

[0579] Providing explanatory text

[0580] Description: Provides the user with a voice commentary of the information received by the device.

[0581] Input: The explanatory text received on the terminal.

[0582] Output: The audio description provided to the user.

[0583] Specific operation: The device uses speech synthesis technology (e.g., Amazon Polly) to convert the explanatory text into speech and whisper it to the user through the audio glasses.

[0584] Step 8:

[0585] Recognizing the user's emotional state

[0586] Description: Emotion engine recognizes the user's emotional state in real time.

[0587] Input: User's voice tone and facial expression data.

[0588] Output: Perceived emotional state of the user.

[0589] How it works: The emotion engine uses the Facial Emotion Recognition API and voice analysis tools to analyze the user's facial expressions and voice tone to determine their emotional state, such as confusion or understanding.

[0590] Step 9:

[0591] Adjusted explanatory text

[0592] Description: Adjusts explanatory text based on perceived emotional state.

[0593] Input: User's emotional state, initial generated commentary.

[0594] Output: Adjusted explanatory text.

[0595] What it does: If the emotion engine determines that the user is confused, it will provide additional information, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network." Conversely, if the user understands, it will provide a concise explanation, "Deep learning is a type of machine learning."

[0596] (Application example 2)

[0597] 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."

[0598] In traditional brick-and-mortar stores, it can be difficult for store clerks to provide accurate explanations to customers when technical terms or complex concepts are discussed. This can be particularly difficult when the clerk lacks specialized knowledge or the customer does not understand the technical terms. Furthermore, no system has existed to date that can provide explanations tailored to the customer's level of understanding and emotional state.

[0599] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0600] In this invention, the server

[0601] an audio acquisition means for acquiring surrounding audio in real time;

[0602] A speech recognition means for converting the acquired speech into text data;

[0603] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[0604] Explanation generation means for generating an explanation for a specific term;

[0605] an emotion recognition means for recognizing the emotional state of a user and adjusting the commentary;

[0606] An explanation providing means for providing the generated explanation to the user in real time by voice through smart glasses;

[0607] This will enable store staff and customers in physical stores to explain technical terms and complex concepts in real time, in line with the user's emotional state.

[0608] The "audio acquisition means" is a device or system for collecting ambient audio in real time.

[0609] "Speech recognition means" refers to a technique or algorithm for converting acquired voice data into text data.

[0610] A "generative AI model" is a model that uses artificial intelligence technology to analyze text data and identify specific patterns and terms.

[0611] An "analysis means" is a device or system that analyzes text data and extracts specific terms or important parts from it.

[0612] An "explanation generation means" is a device or system for generating an explanation about a specific term or concept.

[0613] An "emotion recognition means" is a device or system for recognizing a user's emotional state and adjusting commentary based on that information.

[0614] The "explanation providing means" is a device or system for providing the generated commentary to the user in real time.

[0615] "Smart glasses" are wearable devices that provide users with visual or audio information when worn by the user.

[0616] An "audio device" is an electronic device for inputting and outputting sound.

[0617] A "brick and mortar store" is a store where goods or services are sold at a physical location.

[0618] The present invention is a system that supports customer service in brick-and-mortar stores by capturing surrounding voices in real time, generating explanatory text for technical terms and complex concepts, and providing it to users. This system includes a voice capture means, a voice recognition means, an analysis means using a generative AI model, an explanation generation means, an emotion recognition means, and an explanation provision means. Furthermore, by using smart glasses, the explanatory text can be provided in real time by voice.

[0619] 1. Audio acquisition method

[0620] The device captures ambient sounds in real time using smart glasses with built-in microphones.

[0621] 2. Voice Recognition Method

[0622] The captured voice is converted into text data by a speech recognition means, which uses a highly accurate speech recognition algorithm, such as Google's speech recognition service.

[0623] 3. Analysis method

[0624] The converted text data is then analyzed by an analysis means using a generative AI model, which identifies specific terms from the text data and generates corresponding explanatory text.

[0625] 4. Explanation Generation Method

[0626] Based on the analyzed data, explanations for specific terms and concepts are generated and presented to users in a format that is easy to understand.

[0627] 5. Emotion recognition means

[0628] The system recognizes the user's emotional state when providing commentary. The emotion recognition method analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. Specifically, the EmotionRecognizer API is used.

[0629] 6. Means of providing explanations

[0630] The generated explanations are provided to users in real time as audio through the smart glasses, allowing store staff to provide appropriate explanations to customers about technical terms and complex concepts.

[0631] Examples:

[0632] For example, if a store clerk explains, "This product is equipped with a new biometric sensor," the smart glasses capture this speech and send it to a server. The server converts the speech into text and identifies the term "biometric sensor." Using a generative AI model, an explanation is generated, such as, "A biometric sensor is a technology that measures and recognizes physical characteristics." If the emotion recognition method determines that the user is confused, a more detailed explanation is added. For example, specific examples such as "This includes fingerprint and facial recognition" are provided.

[0633] Example prompt sentence:

[0634] My customer seems confused. Can you explain more about biometric sensors?

[0635] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0636] Step 1:

[0637] The device captures the surrounding audio in real time. Here, the microphone built into the smart glasses is used. The input is the surrounding audio, and the output is audio data.

[0638] Step 2:

[0639] The voice data acquired by the device is sent to a server via the Internet. The input is the voice data, and the output is the voice data sent to the server.

[0640] Step 3:

[0641] The server converts the received voice data into text data using a voice recognition means. Here, a voice recognition algorithm (e.g., Google's voice recognition service) is used. The input is voice data, and the output is text data.

[0642] Step 4:

[0643] The server analyzes the text data using a generative AI model. The analysis means identifies specific terms from the text data. The input is the text data, and the output is the analysis results that include the specific terms.

[0644] Step 5:

[0645] The server generates an explanatory text using an explanatory text generation means based on the analysis results. The input is the analysis results, and the output is the explanatory text.

[0646] Step 6:

[0647] The server analyzes the user's emotional state using an emotion recognition means. The emotion recognition means analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. The input is the user's tone of voice and facial expressions, and the output is the analysis result of the user's emotional state.

[0648] Step 7:

[0649] The server adjusts the commentary based on the analysis of the emotional state. If the user is confused, it adds a detailed explanation, and if the user understands, it simplifies it. The input is the analysis of the emotional state and the commentary, and the output is the adjusted commentary.

[0650] Step 8:

[0651] The server sends the generated commentary to the terminal. The input is the adjusted commentary, and the output is the commentary sent to the smart glasses.

[0652] Step 9:

[0653] The terminal uses the commentary providing means to provide the commentary text to the user in real time via an audio device of the smart glasses, the input being the adjusted commentary text and the output being the audio commentary provided to the user.

[0654] 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.

[0655] 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.

[0656] 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.

[0657] [Third embodiment]

[0658] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0659] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0660] 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).

[0661] 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.

[0662] 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.

[0663] 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).

[0664] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0665] 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.

[0666] 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.

[0667] 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.

[0668] 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.

[0669] 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."

[0670] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0671] First, the device uses an audio device to capture surrounding sounds in real time, collects the audio data, and then transmits the captured audio data to a server via the Internet.

[0672] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0673] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0674] The generated commentary is then sent to the terminal via the Internet. The commentary providing means uses the terminal's audio device to whisper (play low-volume audio) the commentary to the user, allowing the user to understand the content of the technical conversation in real time.

[0675] Specific examples

[0676] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0677] The generated explanation is sent to the device and whispered to the user by the explanation providing means, allowing the user to understand the technical term "deep learning" in real time.

[0678] In this way, the present invention can be a powerful support tool for smooth communication between people with knowledge in different fields. Furthermore, by explaining technical terms in real time, it can improve learning efficiency and comprehension.

[0679] The processing flow will be explained below.

[0680] Step 1:

[0681] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures the audio data.

[0682] Step 2:

[0683] The device sends the acquired voice data to the server. Specifically, the device uploads the voice data to the server via the Internet.

[0684] Step 3:

[0685] The server converts the transmitted voice data into text data using a voice recognition means, which utilizes a highly accurate voice recognition algorithm to convert the voice into text.

[0686] Step 4:

[0687] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts in the text data.

[0688] Step 5:

[0689] The server generates an explanation for the identified technical term using an explanation generation means. For example, if "deep learning" is identified, the generative AI model creates an explanation such as "deep learning is a type of machine learning that uses neural networks."

[0690] Step 6:

[0691] The server transmits the generated commentary to the terminal. Specifically, the server transmits the generated commentary to the terminal via the Internet.

[0692] Step 7:

[0693] The device uses the explanation providing means to whisper the explanation to the user, and the built-in speaker of the audio glasses plays the explanation in a whisper to the user.

[0694] Step 8:

[0695] Users can listen to the commentary in real time and understand the content of the conversation, and can participate in professional conversations by referring to the commentary provided.

[0696] By following these steps in sequence, users can understand the surrounding jargon-laden conversation in real time.

[0697] Example 1

[0698] 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."

[0699] In technical conversations and discussions, there are many technical terms and complex concepts that are difficult for general users to understand. This makes it difficult for people with knowledge in different fields to communicate smoothly, causing problems such as delays in information sharing and decision-making. Furthermore, if understanding of technical terms is not provided in real time during meetings and discussions, users' learning efficiency and level of understanding may decrease.

[0700] 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.

[0701] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative artificial intelligence model and identifying portions containing technical terms, an explanation generation means for generating explanations of the technical terms, and an explanation provision means for providing the generated explanations to users in real time. This allows users to receive explanations of technical terms in real time, enabling information sharing across different fields of expertise to be carried out quickly and effectively.

[0702] "Ambient audio" is audio data including speech and background sounds obtained from the user's surrounding environment.

[0703] "Audio acquisition means" refers to a device or method for acquiring ambient audio in real time, and includes a microphone and an audio device.

[0704] "Speech recognition means" refers to a technology or system that converts acquired voice data into text data, and includes speech recognition software and algorithms.

[0705] A "generative artificial intelligence model" is an artificial intelligence model that learns from given data and automatically generates explanatory text and analysis, and includes machine learning models that use neural networks.

[0706] An "analysis means" is a process or system that analyzes text data and identifies specific technical terms and important concepts, and includes natural language processing techniques.

[0707] The "explanation generation means" is a system or method that generates easy-to-understand explanations for the technical terms and concepts identified by the analysis means.

[0708] The "explanation providing means" is a means for providing the generated explanation to the user, and includes text display and audio output.

[0709] An "audio device" is a device for reproducing sound, including speakers and earphones.

[0710] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0711] First, the device uses an audio device (such as audio glasses) to capture surrounding sounds in real time and collects that audio data. For example, if an engineer is talking about "deep learning" during a meeting, the device will record the conversation.

[0712] The device then transmits the captured audio data over the Internet to a server, where the data is compressed and transmitted using a secure protocol (e.g., HTTPS).

[0713] The server converts the received voice data into text data using a voice recognition means (for example, a general voice recognition service). For example, it converts "Voice during a meeting: 'About deep learning...'" into text data such as "About deep learning...".

[0714] The server then analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis means identifies parts of the text data that contain technical terms and important concepts. For example, it recognizes "deep learning" as a technical term.

[0715] The server's explanation generation means then generates simple explanations for the identified technical terms. For example, for the analyzed term "deep learning," it generates the explanation "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0716] The generated commentary is then sent to the device via the Internet, with care taken to minimize data size and ensure speedy transmission.

[0717] Finally, the terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, thereby enabling the user to understand the content of the specialized conversation in real time.

[0718] For example:

[0719] For example, if an engineer starts talking about "deep learning" during a meeting, the device (audio glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data." The generated explanation will be sent to the device, and the explanation provision means will whisper it to the user.

[0720] Here are some specific prompts that can help users use the system to understand meeting terminology in real time:

[0721] What is Deep Learning?

[0722] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0723] Step 1:

[0724] The terminal uses an audio device to capture surrounding sounds in real time. The audio capture means continuously collects audio data, such as speech during a meeting, using a microphone. The input is the surrounding sounds, and the output is the captured audio data.

[0725] Step 2:

[0726] The device transmits the acquired audio data to a server via the Internet. The transmitted audio data is compressed as a pre-processing step and securely transmitted using a secure protocol (e.g., HTTPS). The input is the acquired audio data, and the output is the audio data transmitted to the server.

[0727] Step 3:

[0728] The server converts the received voice data into text data using a voice recognition method (for example, a general voice recognition service). A voice recognition algorithm analyzes the voice waveform and generates a corresponding string of characters. The input is the voice data received by the server, and the output is the converted text data. For example, voice data such as "Voice during meeting: 'About deep learning...'" is converted into text data such as "About deep learning...".

[0729] Step 4:

[0730] The server analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis method identifies technical terms and important concepts from the text data. The input is the converted text data, and the output is the identified technical terms and concepts. For example, "deep learning" is recognized as a technical term.

[0731] Step 5:

[0732] The server's explanation generation means generates simple explanations for the identified technical terms. A generative artificial intelligence model is used to generate explanations related to the identified technical terms. The input is the identified technical terms, and the output is the generated explanation. For example, for "deep learning," the explanation generated is "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0733] Step 6:

[0734] The server then sends the generated commentary to the terminal via the Internet. The data is kept lightweight and can be sent quickly. The input is the generated commentary, and the output is the commentary sent to the terminal.

[0735] Step 7:

[0736] The terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, allowing the user to understand the content of the specialized conversation in real time. The input is the explanation sent to the terminal, and the output is the explanation provided to the user.

[0737] (Application example 1)

[0738] 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."

[0739] The technical terms and work instructions used in factories can be difficult for unskilled workers to understand. In such cases, a lack of understanding or misunderstanding can have a negative impact on safety and efficiency, so a system is needed that can instantly explain the technical terms and work instructions and help workers understand them.

[0740] 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.

[0741] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating explanations for the specific terms, and an explanation provision means for providing the generated explanations to users in real time. This enables a robot to understand work instructions and technical terms in a factory and provide explanations for them.

[0742] The "audio acquisition means" refers to a device or sensor for acquiring surrounding audio in real time.

[0743] "Speech recognition means" refers to a technology or device that converts acquired speech into text data.

[0744] "Analysis means" refers to a technology or device that uses a generative AI model to analyze text data and identify specific terms or important parts.

[0745] The term "explanation generating means" refers to a technique or device that generates an explanation about a specific term.

[0746] The "explanation providing means" refers to a technique or device that provides the generated explanation to the user.

[0747] A "generative AI model" is an artificial intelligence model that analyzes text data and generates explanatory text.

[0748] A "robot" is an artificially created device that can perform specific tasks automatically.

[0749] This invention is a system mainly composed of a voice acquisition means, a voice recognition means, an analysis means, an explanation generation means, and an explanation provision means. The system aims to help workers understand work instructions and technical terms in a factory by explaining them in real time, thereby improving safety and efficiency. Specific embodiments are as follows.

[0750] First, the robot uses an audio device to capture surrounding sounds in real time and collect the audio data.The robot then transmits the captured audio data to a server via the Internet.In this process, a microphone serves as the audio capture means, and the SpeechRecognition library is used for voice recognition.

[0751] The server converts the received voice data into text data using a speech recognition means. The speech recognition means uses a highly accurate speech recognition algorithm, allowing it to accurately convert conversational voice into text data. The converted text data is analyzed by a generative AI model, which functions as an analysis means. The generative AI model uses OpenAI's API, and the analysis means identifies specific terms and important concepts from the text data.

[0752] Next, the server uses the generative AI model to generate explanations for specific terms from the analyzed text data. The explanation generation means uses advanced natural language processing technology to generate easy-to-understand explanations for technical terms and important concepts. For example, for the term "PLC," the following explanation is generated: "PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[0753] The generated explanatory text is again sent to the robot terminal via the Internet. To provide this explanatory text to the user, the explanation providing means whispers (plays audio at a low volume) using the robot's audio device, allowing the worker to understand immediately. Furthermore, the explanatory text can also be displayed on a display. For example, if a worker in a factory is talking about how to operate a new machine and the term "PLC" is mentioned, the robot will capture the voice, send it to the server, analyze it, generate an explanatory text, and provide the following explanation:

[0754] "A PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[0755] In this way, users can easily understand specific technical terms and work instructions in real time. Implementing this invention provides a powerful support tool for smooth communication between people with different fields of expertise, which is expected to result in improved learning efficiency and comprehension.

[0756] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0757] Step 1:

[0758] The device uses an audio device to capture surrounding sounds in real time.

[0759] Input: Ambient audio

[0760] Data processing: Collected as audio data using an audio device

[0761] Output: Audio data

[0762] Step 2:

[0763] The terminal transmits the acquired voice data to a server via the Internet.

[0764] Input: Audio data

[0765] Data calculation: Converts voice data into packet data for transmission

[0766] Output: Packet data to the server

[0767] Step 3:

[0768] The server converts the received voice data into text data using a voice recognition means.

[0769] Input: Audio data

[0770] Data calculation: Convert to text data using a speech recognition algorithm (SpeechRecognition library)

[0771] Output: Text data

[0772] Step 4:

[0773] The server uses a generative AI model to analyze the text data and identify specific terms and key concepts.

[0774] Input: Text data

[0775] Data Computation: Analysis and term identification using generative AI models (OpenAI API)

[0776] Output: Identified terms or key concepts

[0777] Step 5:

[0778] The server generates a description for the identified term.

[0779] Input: Identified terms or key concepts

[0780] Data calculation: Generating explanatory text based on natural language processing (generative AI model)

[0781] Output:Explanation

[0782] Example: Example prompt: "Please explain the following technical term: PLC"

[0783] Step 6:

[0784] The server transmits the generated commentary to the terminal via the Internet.

[0785] Input:Explanation

[0786] Data calculation: Converts into packet data for sending explanatory text

[0787] Output: Packet data to the terminal

[0788] Step 7:

[0789] The explanation providing means of the terminal provides the explanation text to the user in real time.

[0790] Input:Explanation

[0791] Data calculation: Low volume voice playback or text display on the display

[0792] Output: Providing a description as audio or text

[0793] This allows the device to provide users with real-time explanations of technical terms and work instructions to aid understanding.

[0794] 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.

[0795] The present invention is a system that captures ambient audio, converts it into text data, analyzes technical terms and complex concepts using a generative AI model, generates explanatory text, and provides it to users. The system also incorporates an emotion engine that recognizes the user's emotions and adjusts the explanatory text based on the user's emotional state.

[0796] First, the device uses an audio device to capture surrounding sounds in real time and collect the audio data, which is then transmitted to a server via the Internet.

[0797] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0798] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0799] The generated commentary is then sent to the terminal via the Internet. The commentary providing means whispers the commentary to the user using an audio device, specifically audio glasses. The user can listen to the commentary in real time and understand the technical conversation content.

[0800] Furthermore, the system incorporates an emotion engine that can adjust the commentary according to the user's emotional state. The emotion engine recognizes emotions by analyzing the user's tone of voice and facial expressions. Specifically, if the user is feeling unsure or confused, the system can simplify the commentary and add more detailed explanations. If the system determines that the user understands, it can omit or simplify the commentary.

[0801] Specific examples

[0802] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0803] The generated commentary is sent to the device and whispered to the user by the commentary provider. The emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information is provided, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network."

[0804] Conversely, if the emotion engine determines that the user understands, it will provide only a concise explanation, such as "Deep learning is a type of machine learning."

[0805] In this way, the present invention can provide explanations flexibly according to the user's level of understanding and emotional state, which is very useful for understanding technical conversation content in real time.

[0806] The processing flow will be explained below.

[0807] Step 1:

[0808] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures audio data.

[0809] Step 2:

[0810] The device sends the acquired voice data to the server, which then uploads the voice data to the server via the Internet.

[0811] Step 3:

[0812] The server uses a speech recognition means to convert the transmitted voice data into text data, and applies a speech recognition algorithm to analyze the voice data as text information.

[0813] Step 4:

[0814] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts within the text data.

[0815] Step 5:

[0816] The server generates an explanation for the identified technical term using an explanation generation means. Specifically, the generative AI model creates a detailed explanation for the term "deep learning" identified.

[0817] Step 6:

[0818] The server sends the generated commentary to the terminal, and the server sends the commentary to the terminal via the Internet.

[0819] Step 7:

[0820] The terminal uses the explanation providing means to whisper the generated explanation to the user, and the built-in speaker of the audio glasses plays the explanation and whispers it to the user.

[0821] Step 8:

[0822] The device uses a built-in emotion engine to recognize the user's emotional state. Specifically, the emotion engine analyzes the user's tone of voice and facial expressions to identify the user's emotional state.

[0823] Step 9:

[0824] The server receives feedback from the emotion engine and adjusts the explanation based on the user's emotional state, for example, making the explanation more detailed if the user is confused, and more concise if the user understands.

[0825] Step 10:

[0826] The device will then provide the adjusted commentary to the user again, and will whisper the updated commentary to the user using the device's audio device.

[0827] By performing these steps consecutively, users can not only more easily understand the technical terms used in conversations, but also respond flexibly according to their individual levels of understanding and emotional state.

[0828] Example 2

[0829] 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."

[0830] In an environment where technical conversations and terminology are commonplace, users often lack the necessary expertise to fully understand the content of the conversation. While there is a demand for flexible explanations that adapt to the user's emotional state, no such systems currently exist. A method is needed to provide explanations in real time and adjust the explanations according to the user's level of understanding and emotions.

[0831] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating an explanation for the specific term, an explanation provision means for providing the generated explanation to the user in real time, and an emotion recognition means for recognizing the user's emotional state and adjusting the explanation. This makes it possible to understand the content of specialized conversations in real time and provide flexible explanations according to the user's emotional state.

[0832] The "audio acquisition means" is a device or system for acquiring ambient audio in real time.

[0833] "Speech recognition means" refers to an algorithm or system for converting captured speech into text data.

[0834] A "generative AI model" is a model that uses artificial intelligence to process and analyze text data.

[0835] "Analysis means" refers to a means for analyzing text data using a generative AI model and identifying portions containing specific terms.

[0836] The "explanation generating means" is a means for generating an explanation about a specific term.

[0837] The "explanation providing means" is a means for providing the user with an explanation in real time.

[0838] The "emotion recognition means" is a means for recognizing the user's emotional state and adjusting the commentary accordingly.

[0839] The present invention is a system for understanding specialized conversation content in real time, and specifically includes a voice acquisition means, a voice recognition means, a generative AI model, an analysis means, an explanation generation means, an explanation provision means, and an emotion recognition means.

[0840] Audio acquisition means

[0841] The device uses an audio device (such as audio glasses) to capture surrounding sounds in real time. A dedicated microphone built into the device is equipped with noise-canceling technology to remove ambient noise.

[0842] Voice recognition means

[0843] The server receives the voice data sent from the device and converts it into text data using a highly accurate voice recognition algorithm (for example, Google Cloud Speech-to-Text API).

[0844] Generative AI models and analytical methods

[0845] The server analyzes the text data using a generative AI model (e.g., GPT-4). The analysis method identifies sections containing technical terms and important concepts and generates plain-language explanations for those sections.

[0846] Explanation generation method

[0847] The server generates explanatory text based on the analysis. The generated explanatory text explains technical terms in an easy-to-understand format and is provided to users.

[0848] Means of providing explanations

[0849] The generated commentary is sent to the device via the Internet, and the device whispers the commentary to the user using a commentary providing means (e.g., audio glasses). During this process, the text is converted into speech using speech synthesis technology (e.g., Amazon Polly).

[0850] emotion recognition means

[0851] The emotion engine analyzes the user's tone of voice and facial expressions to recognize their emotional state, allowing the commentary to be tailored to the user's level of understanding and emotion. The emotion recognition method uses the Facial Emotion Recognition (FER) API and voice analysis tools.

[0852] Specific examples

[0853] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0854] The generated explanation is sent to the device and whispered to the user by the explanation provision means. At this time, the emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information such as "Deep learning uses specific algorithms, one of which is a convolutional neural network" is provided. Conversely, if the emotion engine determines that the user understands, only a concise explanation is provided. A simplified explanation such as "Deep learning is a type of machine learning" is provided.

[0855] Prompt Sentence Examples

[0856] "Analyze the conversational content obtained from the audio data and generate concise explanations for technical terms. For example, if we talk about 'deep learning,' please provide a definition and a brief explanation."

[0857] "Try to adapt your explanation to take into account the user's emotional state. If the user is confused, provide additional details. If they understand, provide a concise explanation."

[0858] In this way, the present invention is a system that flexibly provides explanatory text according to the user's level of understanding and emotional state, and is extremely useful for understanding specialized conversation content in real time.

[0859] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0860] Step 1:

[0861] Acquiring audio data

[0862] Description: The device uses the audio device to capture ambient sounds in real time.

[0863] Input: An audio capture method receives ambient sound in real time.

[0864] Output: The captured audio data.

[0865] How it works: The device's built-in microphone collects audio from within the conference room and uses noise-canceling technology to remove it.

[0866] Step 2:

[0867] Sending audio data

[0868] Description: The audio data acquired by the device is sent to a server via the Internet.

[0869] Input: Audio data stored in device storage.

[0870] Output: The audio data sent to the server.

[0871] Specific operation: The device encodes the voice data using a security protocol (SSL / TLS) and transmits it over the network to the server.

[0872] Step 3:

[0873] Converting audio data to text

[0874] Description: The server converts the received voice data into text data using a highly accurate voice recognition algorithm.

[0875] Input: The audio data received by the server.

[0876] Output: The converted text data.

[0877] What it does: The server runs a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API) and generates text data called "deep learning."

[0878] Step 4:

[0879] Text data analysis

[0880] Description: The server analyzes text data using a generative AI model.

[0881] Input: The converted text data.

[0882] Output: Parsed data, terminology identified.

[0883] What happens: The server uses a generative AI model (e.g., GPT-4) to analyze the text and identify the term "deep learning."

[0884] Step 5:

[0885] Generate explanatory text

[0886] Description: The server generates a description based on the analysis.

[0887] Input: Parsed data, identified terminology.

[0888] Output: The generated description.

[0889] Specific operation: The server generates an explanatory text such as, "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0890] Step 6:

[0891] Submit commentary

[0892] Description: Sends a server-generated explanatory text to the device.

[0893] Input: The generated description.

[0894] Output: A description sent to the terminal.

[0895] Specific operation: The server encodes the generated commentary and sends it to the terminal via the Internet.

[0896] Step 7:

[0897] Providing explanatory text

[0898] Description: Provides the user with a voice commentary of the information received by the device.

[0899] Input: The explanatory text received on the terminal.

[0900] Output: The audio description provided to the user.

[0901] Specific operation: The device uses speech synthesis technology (e.g., Amazon Polly) to convert the explanatory text into speech and whisper it to the user through the audio glasses.

[0902] Step 8:

[0903] Recognizing the user's emotional state

[0904] Description: Emotion engine recognizes the user's emotional state in real time.

[0905] Input: User's voice tone and facial expression data.

[0906] Output: Perceived emotional state of the user.

[0907] How it works: The emotion engine uses the Facial Emotion Recognition API and voice analysis tools to analyze the user's facial expressions and voice tone to determine their emotional state, such as confusion or understanding.

[0908] Step 9:

[0909] Adjusted explanatory text

[0910] Description: Adjusts explanatory text based on perceived emotional state.

[0911] Input: User's emotional state, initial generated commentary.

[0912] Output: Adjusted explanatory text.

[0913] What it does: If the emotion engine determines that the user is confused, it will provide additional information, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network." Conversely, if the user understands, it will provide a concise explanation, "Deep learning is a type of machine learning."

[0914] (Application example 2)

[0915] 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."

[0916] In traditional brick-and-mortar stores, it can be difficult for store clerks to provide accurate explanations to customers when technical terms or complex concepts are discussed. This can be particularly difficult when the clerk lacks specialized knowledge or the customer does not understand the technical terms. Furthermore, no system has existed to date that can provide explanations tailored to the customer's level of understanding and emotional state.

[0917] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0918] In this invention, the server

[0919] an audio acquisition means for acquiring surrounding audio in real time;

[0920] A speech recognition means for converting the acquired speech into text data;

[0921] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[0922] Explanation generation means for generating an explanation for a specific term;

[0923] an emotion recognition means for recognizing the emotional state of a user and adjusting the commentary;

[0924] An explanation providing means for providing the generated explanation to the user in real time by voice through smart glasses;

[0925] This will enable store staff and customers in physical stores to explain technical terms and complex concepts in real time, in line with the user's emotional state.

[0926] The "audio acquisition means" is a device or system for collecting ambient audio in real time.

[0927] "Speech recognition means" refers to a technique or algorithm for converting acquired voice data into text data.

[0928] A "generative AI model" is a model that uses artificial intelligence technology to analyze text data and identify specific patterns and terms.

[0929] An "analysis means" is a device or system that analyzes text data and extracts specific terms or important parts from it.

[0930] An "explanation generation means" is a device or system for generating an explanation about a specific term or concept.

[0931] An "emotion recognition means" is a device or system for recognizing a user's emotional state and adjusting commentary based on that information.

[0932] The "explanation providing means" is a device or system for providing the generated commentary to the user in real time.

[0933] "Smart glasses" are wearable devices that provide users with visual or audio information when worn by the user.

[0934] An "audio device" is an electronic device for inputting and outputting sound.

[0935] A "brick and mortar store" is a store where goods or services are sold at a physical location.

[0936] The present invention is a system that supports customer service in brick-and-mortar stores by capturing surrounding voices in real time, generating explanatory text for technical terms and complex concepts, and providing it to users. This system includes a voice capture means, a voice recognition means, an analysis means using a generative AI model, an explanation generation means, an emotion recognition means, and an explanation provision means. Furthermore, by using smart glasses, the explanatory text can be provided in real time by voice.

[0937] 1. Audio acquisition method

[0938] The device captures ambient sounds in real time using smart glasses with built-in microphones.

[0939] 2. Voice Recognition Method

[0940] The captured voice is converted into text data by a speech recognition means, which uses a highly accurate speech recognition algorithm, such as Google's speech recognition service.

[0941] 3. Analysis method

[0942] The converted text data is then analyzed by an analysis means using a generative AI model, which identifies specific terms from the text data and generates corresponding explanatory text.

[0943] 4. Explanation Generation Method

[0944] Based on the analyzed data, explanations for specific terms and concepts are generated and presented to users in a format that is easy to understand.

[0945] 5. Emotion recognition means

[0946] The system recognizes the user's emotional state when providing commentary. The emotion recognition method analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. Specifically, the EmotionRecognizer API is used.

[0947] 6. Means of providing explanations

[0948] The generated explanations are provided to users in real time as audio through the smart glasses, allowing store staff to provide appropriate explanations to customers about technical terms and complex concepts.

[0949] Examples:

[0950] For example, if a store clerk explains, "This product is equipped with a new biometric sensor," the smart glasses capture this speech and send it to a server. The server converts the speech into text and identifies the term "biometric sensor." Using a generative AI model, an explanation is generated, such as, "A biometric sensor is a technology that measures and recognizes physical characteristics." If the emotion recognition method determines that the user is confused, a more detailed explanation is added. For example, specific examples such as "This includes fingerprint and facial recognition" are provided.

[0951] Example prompt sentence:

[0952] My customer seems confused. Can you explain more about biometric sensors?

[0953] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0954] Step 1:

[0955] The device captures the surrounding audio in real time. Here, the microphone built into the smart glasses is used. The input is the surrounding audio, and the output is audio data.

[0956] Step 2:

[0957] The voice data acquired by the device is sent to a server via the Internet. The input is the voice data, and the output is the voice data sent to the server.

[0958] Step 3:

[0959] The server converts the received voice data into text data using a voice recognition means. Here, a voice recognition algorithm (e.g., Google's voice recognition service) is used. The input is voice data, and the output is text data.

[0960] Step 4:

[0961] The server analyzes the text data using a generative AI model. The analysis means identifies specific terms from the text data. The input is the text data, and the output is the analysis results that include the specific terms.

[0962] Step 5:

[0963] The server generates an explanatory text using an explanatory text generation means based on the analysis results. The input is the analysis results, and the output is the explanatory text.

[0964] Step 6:

[0965] The server analyzes the user's emotional state using an emotion recognition means. The emotion recognition means analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. The input is the user's tone of voice and facial expressions, and the output is the analysis result of the user's emotional state.

[0966] Step 7:

[0967] The server adjusts the commentary based on the analysis of the emotional state. If the user is confused, it adds a detailed explanation, and if the user understands, it simplifies it. The input is the analysis of the emotional state and the commentary, and the output is the adjusted commentary.

[0968] Step 8:

[0969] The server sends the generated commentary to the terminal. The input is the adjusted commentary, and the output is the commentary sent to the smart glasses.

[0970] Step 9:

[0971] The terminal uses the commentary providing means to provide the commentary text to the user in real time via an audio device of the smart glasses, the input being the adjusted commentary text and the output being the audio commentary provided to the user.

[0972] 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.

[0973] 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.

[0974] 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.

[0975] [Fourth embodiment]

[0976] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0977] 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.

[0978] 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).

[0979] 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.

[0980] 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.

[0981] 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).

[0982] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0983] 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.

[0984] 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.

[0985] 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.

[0986] 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.

[0987] 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.

[0988] 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."

[0989] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[0990] First, the device uses an audio device to capture surrounding sounds in real time, collects the audio data, and then transmits the captured audio data to a server via the Internet.

[0991] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[0992] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[0993] The generated commentary is then sent to the terminal via the Internet. The commentary providing means uses the terminal's audio device to whisper (play low-volume audio) the commentary to the user, allowing the user to understand the content of the technical conversation in real time.

[0994] Specific examples

[0995] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[0996] The generated explanation is sent to the device and whispered to the user by the explanation providing means, allowing the user to understand the technical term "deep learning" in real time.

[0997] In this way, the present invention can be a powerful support tool for smooth communication between people with knowledge in different fields. Furthermore, by explaining technical terms in real time, it can improve learning efficiency and comprehension.

[0998] The processing flow will be explained below.

[0999] Step 1:

[1000] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures the audio data.

[1001] Step 2:

[1002] The device sends the acquired voice data to the server. Specifically, the device uploads the voice data to the server via the Internet.

[1003] Step 3:

[1004] The server converts the transmitted voice data into text data using a voice recognition means, which utilizes a highly accurate voice recognition algorithm to convert the voice into text.

[1005] Step 4:

[1006] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts in the text data.

[1007] Step 5:

[1008] The server generates an explanation for the identified technical term using an explanation generation means. For example, if "deep learning" is identified, the generative AI model creates an explanation such as "deep learning is a type of machine learning that uses neural networks."

[1009] Step 6:

[1010] The server transmits the generated commentary to the terminal. Specifically, the server transmits the generated commentary to the terminal via the Internet.

[1011] Step 7:

[1012] The device uses the explanation providing means to whisper the explanation to the user, and the built-in speaker of the audio glasses plays the explanation in a whisper to the user.

[1013] Step 8:

[1014] Users can listen to the commentary in real time and understand the content of the conversation, and can participate in professional conversations by referring to the commentary provided.

[1015] By following these steps in sequence, users can understand the surrounding jargon-laden conversation in real time.

[1016] Example 1

[1017] 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."

[1018] In technical conversations and discussions, there are many technical terms and complex concepts that are difficult for general users to understand. This makes it difficult for people with knowledge in different fields to communicate smoothly, causing problems such as delays in information sharing and decision-making. Furthermore, if understanding of technical terms is not provided in real time during meetings and discussions, users' learning efficiency and level of understanding may decrease.

[1019] 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.

[1020] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative artificial intelligence model and identifying portions containing technical terms, an explanation generation means for generating explanations of the technical terms, and an explanation provision means for providing the generated explanations to users in real time. This allows users to receive explanations of technical terms in real time, enabling information sharing across different fields of expertise to be carried out quickly and effectively.

[1021] "Ambient audio" is audio data including speech and background sounds obtained from the user's surrounding environment.

[1022] "Audio acquisition means" refers to a device or method for acquiring ambient audio in real time, and includes a microphone and an audio device.

[1023] "Speech recognition means" refers to a technology or system that converts acquired voice data into text data, and includes speech recognition software and algorithms.

[1024] A "generative artificial intelligence model" is an artificial intelligence model that learns from given data and automatically generates explanatory text and analysis, and includes machine learning models that use neural networks.

[1025] An "analysis means" is a process or system that analyzes text data and identifies specific technical terms and important concepts, and includes natural language processing techniques.

[1026] The "explanation generation means" is a system or method that generates easy-to-understand explanations for the technical terms and concepts identified by the analysis means.

[1027] The "explanation providing means" is a means for providing the generated explanation to the user, and includes text display and audio output.

[1028] An "audio device" is a device for reproducing sound, including speakers and earphones.

[1029] This invention is a system that captures surrounding voices in real time, converts them into text data, analyzes technical terms and complex concepts using a generative AI model, and generates explanatory text to provide to users. This system is mainly composed of the following elements: voice capture means, voice recognition means, analysis means, explanation generation means, and explanation provision means.

[1030] First, the device uses an audio device (such as audio glasses) to capture surrounding sounds in real time and collects that audio data. For example, if an engineer is talking about "deep learning" during a meeting, the device will record the conversation.

[1031] The device then transmits the captured audio data over the Internet to a server, where the data is compressed and transmitted using a secure protocol (e.g., HTTPS).

[1032] The server converts the received voice data into text data using a voice recognition means (for example, a general voice recognition service). For example, it converts "Voice during a meeting: 'About deep learning...'" into text data such as "About deep learning...".

[1033] The server then analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis means identifies parts of the text data that contain technical terms and important concepts. For example, it recognizes "deep learning" as a technical term.

[1034] The server's explanation generation means then generates simple explanations for the identified technical terms. For example, for the analyzed term "deep learning," it generates the explanation "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[1035] The generated commentary is then sent to the device via the Internet, with care taken to minimize data size and ensure speedy transmission.

[1036] Finally, the terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, thereby enabling the user to understand the content of the specialized conversation in real time.

[1037] For example:

[1038] For example, if an engineer starts talking about "deep learning" during a meeting, the device (audio glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data." The generated explanation will be sent to the device, and the explanation provision means will whisper it to the user.

[1039] Here are some specific prompts that can help users use the system to understand meeting terminology in real time:

[1040] What is Deep Learning?

[1041] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1042] Step 1:

[1043] The terminal uses an audio device to capture surrounding sounds in real time. The audio capture means continuously collects audio data, such as speech during a meeting, using a microphone. The input is the surrounding sounds, and the output is the captured audio data.

[1044] Step 2:

[1045] The device transmits the acquired audio data to a server via the Internet. The transmitted audio data is compressed as a pre-processing step and securely transmitted using a secure protocol (e.g., HTTPS). The input is the acquired audio data, and the output is the audio data transmitted to the server.

[1046] Step 3:

[1047] The server converts the received voice data into text data using a voice recognition method (for example, a general voice recognition service). A voice recognition algorithm analyzes the voice waveform and generates a corresponding string of characters. The input is the voice data received by the server, and the output is the converted text data. For example, voice data such as "Voice during meeting: 'About deep learning...'" is converted into text data such as "About deep learning...".

[1048] Step 4:

[1049] The server analyzes the converted text data using a generative AI model (e.g., a general natural language processing model). The analysis method identifies technical terms and important concepts from the text data. The input is the converted text data, and the output is the identified technical terms and concepts. For example, "deep learning" is recognized as a technical term.

[1050] Step 5:

[1051] The server's explanation generation means generates simple explanations for the identified technical terms. A generative artificial intelligence model is used to generate explanations related to the identified technical terms. The input is the identified technical terms, and the output is the generated explanation. For example, for "deep learning," the explanation generated is "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[1052] Step 6:

[1053] The server then sends the generated commentary to the terminal via the Internet. The data is kept lightweight and can be sent quickly. The input is the generated commentary, and the output is the commentary sent to the terminal.

[1054] Step 7:

[1055] The terminal uses the explanation providing means to whisper (play back audio at a low volume) the generated explanation to the user, allowing the user to understand the content of the specialized conversation in real time. The input is the explanation sent to the terminal, and the output is the explanation provided to the user.

[1056] (Application example 1)

[1057] 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."

[1058] The technical terms and work instructions used in factories can be difficult for unskilled workers to understand. In such cases, a lack of understanding or misunderstanding can have a negative impact on safety and efficiency, so a system is needed that can instantly explain the technical terms and work instructions and help workers understand them.

[1059] 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.

[1060] In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating explanations for the specific terms, and an explanation provision means for providing the generated explanations to users in real time. This enables a robot to understand work instructions and technical terms in a factory and provide explanations for them.

[1061] The "audio acquisition means" refers to a device or sensor for acquiring surrounding audio in real time.

[1062] "Speech recognition means" refers to a technology or device that converts acquired speech into text data.

[1063] "Analysis means" refers to a technology or device that uses a generative AI model to analyze text data and identify specific terms or important parts.

[1064] The term "explanation generating means" refers to a technique or device that generates an explanation about a specific term.

[1065] The "explanation providing means" refers to a technique or device that provides the generated explanation to the user.

[1066] A "generative AI model" is an artificial intelligence model that analyzes text data and generates explanatory text.

[1067] A "robot" is an artificially created device that can perform specific tasks automatically.

[1068] This invention is a system mainly composed of a voice acquisition means, a voice recognition means, an analysis means, an explanation generation means, and an explanation provision means. The system aims to help workers understand work instructions and technical terms in a factory by explaining them in real time, thereby improving safety and efficiency. Specific embodiments are as follows.

[1069] First, the robot uses an audio device to capture surrounding sounds in real time and collect the audio data.The robot then transmits the captured audio data to a server via the Internet.In this process, a microphone serves as the audio capture means, and the SpeechRecognition library is used for voice recognition.

[1070] The server converts the received voice data into text data using a speech recognition means. The speech recognition means uses a highly accurate speech recognition algorithm, allowing it to accurately convert conversational voice into text data. The converted text data is analyzed by a generative AI model, which functions as an analysis means. The generative AI model uses OpenAI's API, and the analysis means identifies specific terms and important concepts from the text data.

[1071] Next, the server uses the generative AI model to generate explanations for specific terms from the analyzed text data. The explanation generation means uses advanced natural language processing technology to generate easy-to-understand explanations for technical terms and important concepts. For example, for the term "PLC," the following explanation is generated: "PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[1072] The generated explanatory text is again sent to the robot terminal via the Internet. To provide this explanatory text to the user, the explanation providing means whispers (plays audio at a low volume) using the robot's audio device, allowing the worker to understand immediately. Furthermore, the explanatory text can also be displayed on a display. For example, if a worker in a factory is talking about how to operate a new machine and the term "PLC" is mentioned, the robot will capture the voice, send it to the server, analyze it, generate an explanatory text, and provide the following explanation:

[1073] "A PLC (Programmable Logic Controller) is a computer used to manage the automation of industrial machinery. It is a programmable logic device that efficiently controls machines and production processes."

[1074] In this way, users can easily understand specific technical terms and work instructions in real time. Implementing this invention provides a powerful support tool for smooth communication between people with different fields of expertise, which is expected to result in improved learning efficiency and comprehension.

[1075] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1076] Step 1:

[1077] The device uses an audio device to capture surrounding sounds in real time.

[1078] Input: Ambient audio

[1079] Data processing: Collected as audio data using an audio device

[1080] Output: Audio data

[1081] Step 2:

[1082] The terminal transmits the acquired voice data to a server via the Internet.

[1083] Input: Audio data

[1084] Data calculation: Converts voice data into packet data for transmission

[1085] Output: Packet data to the server

[1086] Step 3:

[1087] The server converts the received voice data into text data using a voice recognition means.

[1088] Input: Audio data

[1089] Data calculation: Convert to text data using a speech recognition algorithm (SpeechRecognition library)

[1090] Output: Text data

[1091] Step 4:

[1092] The server uses a generative AI model to analyze the text data and identify specific terms and key concepts.

[1093] Input: Text data

[1094] Data Computation: Analysis and term identification using generative AI models (OpenAI API)

[1095] Output: Identified terms or key concepts

[1096] Step 5:

[1097] The server generates a description for the identified term.

[1098] Input: Identified terms or key concepts

[1099] Data calculation: Generating explanatory text based on natural language processing (generative AI model)

[1100] Output:Explanation

[1101] Example: Example prompt: "Please explain the following technical term: PLC"

[1102] Step 6:

[1103] The server transmits the generated commentary to the terminal via the Internet.

[1104] Input:Explanation

[1105] Data calculation: Converts into packet data for sending explanatory text

[1106] Output: Packet data to the terminal

[1107] Step 7:

[1108] The explanation providing means of the terminal provides the explanation text to the user in real time.

[1109] Input:Explanation

[1110] Data calculation: Low volume voice playback or text display on the display

[1111] Output: Providing a description as audio or text

[1112] This allows the device to provide users with real-time explanations of technical terms and work instructions to aid understanding.

[1113] 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.

[1114] The present invention is a system that captures ambient audio, converts it into text data, analyzes technical terms and complex concepts using a generative AI model, generates explanatory text, and provides it to users. The system also incorporates an emotion engine that recognizes the user's emotions and adjusts the explanatory text based on the user's emotional state.

[1115] First, the device uses an audio device to capture surrounding sounds in real time and collect the audio data, which is then transmitted to a server via the Internet.

[1116] The server converts the received voice data into text data using a voice recognition means. The voice recognition means uses a highly accurate voice recognition algorithm, enabling accurate conversion of conversational voice into text data.

[1117] The server then analyzes the converted text data using a generative AI model, which identifies sections containing technical terms and key concepts and generates plain-language explanations of those terms and concepts.

[1118] The generated commentary is then sent to the terminal via the Internet. The commentary providing means whispers the commentary to the user using an audio device, specifically audio glasses. The user can listen to the commentary in real time and understand the technical conversation content.

[1119] Furthermore, the system incorporates an emotion engine that can adjust the commentary according to the user's emotional state. The emotion engine recognizes emotions by analyzing the user's tone of voice and facial expressions. Specifically, if the user is feeling unsure or confused, the system can simplify the commentary and add more detailed explanations. If the system determines that the user understands, it can omit or simplify the commentary.

[1120] Specific examples

[1121] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[1122] The generated commentary is sent to the device and whispered to the user by the commentary provider. The emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information is provided, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network."

[1123] Conversely, if the emotion engine determines that the user understands, it will provide only a concise explanation, such as "Deep learning is a type of machine learning."

[1124] In this way, the present invention can provide explanations flexibly according to the user's level of understanding and emotional state, which is very useful for understanding technical conversation content in real time.

[1125] The processing flow will be explained below.

[1126] Step 1:

[1127] The device uses an audio device to capture surrounding sounds in real time. Specifically, the built-in microphone of the audio glasses captures audio data.

[1128] Step 2:

[1129] The device sends the acquired voice data to the server, which then uploads the voice data to the server via the Internet.

[1130] Step 3:

[1131] The server uses a speech recognition means to convert the transmitted voice data into text data, and applies a speech recognition algorithm to analyze the voice data as text information.

[1132] Step 4:

[1133] The server uses the generative AI model to analyze the converted text data, which identifies technical terms and key concepts within the text data.

[1134] Step 5:

[1135] The server generates an explanation for the identified technical term using an explanation generation means. Specifically, the generative AI model creates a detailed explanation for the term "deep learning" identified.

[1136] Step 6:

[1137] The server sends the generated commentary to the terminal, and the server sends the commentary to the terminal via the Internet.

[1138] Step 7:

[1139] The terminal uses the explanation providing means to whisper the generated explanation to the user, and the built-in speaker of the audio glasses plays the explanation and whispers it to the user.

[1140] Step 8:

[1141] The device uses a built-in emotion engine to recognize the user's emotional state. Specifically, the emotion engine analyzes the user's tone of voice and facial expressions to identify the user's emotional state.

[1142] Step 9:

[1143] The server receives feedback from the emotion engine and adjusts the explanation based on the user's emotional state, for example, making the explanation more detailed if the user is confused, and more concise if the user understands.

[1144] Step 10:

[1145] The device will then provide the adjusted commentary to the user again, and will whisper the updated commentary to the user using the device's audio device.

[1146] By performing these steps consecutively, users can not only more easily understand the technical terms used in conversations, but also respond flexibly according to their individual levels of understanding and emotional state.

[1147] Example 2

[1148] 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."

[1149] In an environment where technical conversations and terminology are commonplace, users often lack the necessary expertise to fully understand the content of the conversation. While there is a demand for flexible explanations that adapt to the user's emotional state, no such systems currently exist. A method is needed to provide explanations in real time and adjust the explanations according to the user's level of understanding and emotions.

[1150] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a voice acquisition means for acquiring surrounding voices in real time, a voice recognition means for converting the acquired voices into text data, an analysis means for analyzing the text data using a generative AI model and identifying portions containing specific terms, an explanation generation means for generating an explanation for the specific term, an explanation provision means for providing the generated explanation to the user in real time, and an emotion recognition means for recognizing the user's emotional state and adjusting the explanation. This makes it possible to understand the content of specialized conversations in real time and provide flexible explanations according to the user's emotional state.

[1151] The "audio acquisition means" is a device or system for acquiring ambient audio in real time.

[1152] "Speech recognition means" refers to an algorithm or system for converting captured speech into text data.

[1153] A "generative AI model" is a model that uses artificial intelligence to process and analyze text data.

[1154] "Analysis means" refers to a means for analyzing text data using a generative AI model and identifying portions containing specific terms.

[1155] The "explanation generating means" is a means for generating an explanation about a specific term.

[1156] The "explanation providing means" is a means for providing the user with an explanation in real time.

[1157] The "emotion recognition means" is a means for recognizing the user's emotional state and adjusting the commentary accordingly.

[1158] The present invention is a system for understanding specialized conversation content in real time, and specifically includes a voice acquisition means, a voice recognition means, a generative AI model, an analysis means, an explanation generation means, an explanation provision means, and an emotion recognition means.

[1159] Audio acquisition means

[1160] The device uses an audio device (such as audio glasses) to capture surrounding sounds in real time. A dedicated microphone built into the device is equipped with noise-canceling technology to remove ambient noise.

[1161] Voice recognition means

[1162] The server receives the voice data sent from the device and converts it into text data using a highly accurate voice recognition algorithm (for example, Google Cloud Speech-to-Text API).

[1163] Generative AI models and analytical methods

[1164] The server analyzes the text data using a generative AI model (e.g., GPT-4). The analysis method identifies sections containing technical terms and important concepts and generates plain-language explanations for those sections.

[1165] Explanation generation method

[1166] The server generates explanatory text based on the analysis. The generated explanatory text explains technical terms in an easy-to-understand format and is provided to users.

[1167] Means of providing explanations

[1168] The generated commentary is sent to the device via the Internet, and the device whispers the commentary to the user using a commentary providing means (e.g., audio glasses). During this process, the text is converted into speech using speech synthesis technology (e.g., Amazon Polly).

[1169] emotion recognition means

[1170] The emotion engine analyzes the user's tone of voice and facial expressions to recognize their emotional state, allowing the commentary to be tailored to the user's level of understanding and emotion. The emotion recognition method uses the Facial Emotion Recognition (FER) API and voice analysis tools.

[1171] Specific examples

[1172] For example, if an engineer starts talking about "deep learning" during a meeting, the device (Audio Glasses) will record the conversation. The recorded audio data will be sent to a server, where it will be converted into the text "deep learning" by the speech recognition means. The analysis means will then identify "deep learning" as a technical term from this text, and the explanation generation means will generate an explanation such as "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[1173] The generated explanation is sent to the device and whispered to the user by the explanation provision means. At this time, the emotion engine analyzes the user's facial expressions and tone of voice, and if it determines that the user is confused, it adds a more detailed explanation. Additional information such as "Deep learning uses specific algorithms, one of which is a convolutional neural network" is provided. Conversely, if the emotion engine determines that the user understands, only a concise explanation is provided. A simplified explanation such as "Deep learning is a type of machine learning" is provided.

[1174] Prompt Sentence Examples

[1175] "Analyze the conversational content obtained from the audio data and generate concise explanations for technical terms. For example, if we talk about 'deep learning,' please provide a definition and a brief explanation."

[1176] "Try to adapt your explanation to take into account the user's emotional state. If the user is confused, provide additional details. If they understand, provide a concise explanation."

[1177] In this way, the present invention is a system that flexibly provides explanatory text according to the user's level of understanding and emotional state, and is extremely useful for understanding specialized conversation content in real time.

[1178] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1179] Step 1:

[1180] Acquiring audio data

[1181] Description: The device uses the audio device to capture ambient sounds in real time.

[1182] Input: An audio capture method receives ambient sound in real time.

[1183] Output: The captured audio data.

[1184] How it works: The device's built-in microphone collects audio from within the conference room and uses noise-canceling technology to remove it.

[1185] Step 2:

[1186] Sending audio data

[1187] Description: The audio data acquired by the device is sent to a server via the Internet.

[1188] Input: Audio data stored in device storage.

[1189] Output: The audio data sent to the server.

[1190] Specific operation: The device encodes the voice data using a security protocol (SSL / TLS) and transmits it over the network to the server.

[1191] Step 3:

[1192] Converting audio data to text

[1193] Description: The server converts the received voice data into text data using a highly accurate voice recognition algorithm.

[1194] Input: The audio data received by the server.

[1195] Output: The converted text data.

[1196] What it does: The server runs a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API) and generates text data called "deep learning."

[1197] Step 4:

[1198] Text data analysis

[1199] Description: The server analyzes text data using a generative AI model.

[1200] Input: The converted text data.

[1201] Output: Parsed data, terminology identified.

[1202] What happens: The server uses a generative AI model (e.g., GPT-4) to analyze the text and identify the term "deep learning."

[1203] Step 5:

[1204] Generate explanatory text

[1205] Description: The server generates a description based on the analysis.

[1206] Input: Parsed data, identified terminology.

[1207] Output: The generated description.

[1208] Specific operation: The server generates an explanatory text such as, "Deep learning is a type of machine learning that uses neural networks to learn complex patterns from large amounts of data."

[1209] Step 6:

[1210] Submit commentary

[1211] Description: Sends a server-generated explanatory text to the device.

[1212] Input: The generated description.

[1213] Output: A description sent to the terminal.

[1214] Specific operation: The server encodes the generated commentary and sends it to the terminal via the Internet.

[1215] Step 7:

[1216] Providing explanatory text

[1217] Description: Provides the user with a voice commentary of the information received by the device.

[1218] Input: The explanatory text received on the terminal.

[1219] Output: The audio description provided to the user.

[1220] Specific operation: The device uses speech synthesis technology (e.g., Amazon Polly) to convert the explanatory text into speech and whisper it to the user through the audio glasses.

[1221] Step 8:

[1222] Recognizing the user's emotional state

[1223] Description: Emotion engine recognizes the user's emotional state in real time.

[1224] Input: User's voice tone and facial expression data.

[1225] Output: Perceived emotional state of the user.

[1226] How it works: The emotion engine uses the Facial Emotion Recognition API and voice analysis tools to analyze the user's facial expressions and voice tone to determine their emotional state, such as confusion or understanding.

[1227] Step 9:

[1228] Adjusted explanatory text

[1229] Description: Adjusts explanatory text based on perceived emotional state.

[1230] Input: User's emotional state, initial generated commentary.

[1231] Output: Adjusted explanatory text.

[1232] What it does: If the emotion engine determines that the user is confused, it will provide additional information, such as "Deep learning uses specific algorithms, one of which is a convolutional neural network." Conversely, if the user understands, it will provide a concise explanation, "Deep learning is a type of machine learning."

[1233] (Application example 2)

[1234] 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."

[1235] In traditional brick-and-mortar stores, it can be difficult for store clerks to provide accurate explanations to customers when technical terms or complex concepts are discussed. This can be particularly difficult when the clerk lacks specialized knowledge or the customer does not understand the technical terms. Furthermore, no system has existed to date that can provide explanations tailored to the customer's level of understanding and emotional state.

[1236] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1237] In this invention, the server

[1238] an audio acquisition means for acquiring surrounding audio in real time;

[1239] A speech recognition means for converting the acquired speech into text data;

[1240] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[1241] Explanation generation means for generating an explanation for a specific term;

[1242] an emotion recognition means for recognizing the emotional state of a user and adjusting the commentary;

[1243] An explanation providing means for providing the generated explanation to the user in real time by voice through smart glasses;

[1244] This will enable store staff and customers in physical stores to explain technical terms and complex concepts in real time, in line with the user's emotional state.

[1245] The "audio acquisition means" is a device or system for collecting ambient audio in real time.

[1246] "Speech recognition means" refers to a technique or algorithm for converting acquired voice data into text data.

[1247] A "generative AI model" is a model that uses artificial intelligence technology to analyze text data and identify specific patterns and terms.

[1248] An "analysis means" is a device or system that analyzes text data and extracts specific terms or important parts from it.

[1249] An "explanation generation means" is a device or system for generating an explanation about a specific term or concept.

[1250] An "emotion recognition means" is a device or system for recognizing a user's emotional state and adjusting commentary based on that information.

[1251] The "explanation providing means" is a device or system for providing the generated commentary to the user in real time.

[1252] "Smart glasses" are wearable devices that provide users with visual or audio information when worn by the user.

[1253] An "audio device" is an electronic device for inputting and outputting sound.

[1254] A "brick and mortar store" is a store where goods or services are sold at a physical location.

[1255] The present invention is a system that supports customer service in brick-and-mortar stores by capturing surrounding voices in real time, generating explanatory text for technical terms and complex concepts, and providing it to users. This system includes a voice capture means, a voice recognition means, an analysis means using a generative AI model, an explanation generation means, an emotion recognition means, and an explanation provision means. Furthermore, by using smart glasses, the explanatory text can be provided in real time by voice.

[1256] 1. Audio acquisition method

[1257] The device captures ambient sounds in real time using smart glasses with built-in microphones.

[1258] 2. Voice Recognition Method

[1259] The captured voice is converted into text data by a speech recognition means, which uses a highly accurate speech recognition algorithm, such as Google's speech recognition service.

[1260] 3. Analysis method

[1261] The converted text data is then analyzed by an analysis means using a generative AI model, which identifies specific terms from the text data and generates corresponding explanatory text.

[1262] 4. Explanation Generation Method

[1263] Based on the analyzed data, explanations for specific terms and concepts are generated and presented to users in a format that is easy to understand.

[1264] 5. Emotion recognition means

[1265] The system recognizes the user's emotional state when providing commentary. The emotion recognition method analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. Specifically, the EmotionRecognizer API is used.

[1266] 6. Means of providing explanations

[1267] The generated explanations are provided to users in real time as audio through the smart glasses, allowing store staff to provide appropriate explanations to customers about technical terms and complex concepts.

[1268] Examples:

[1269] For example, if a store clerk explains, "This product is equipped with a new biometric sensor," the smart glasses capture this speech and send it to a server. The server converts the speech into text and identifies the term "biometric sensor." Using a generative AI model, an explanation is generated, such as, "A biometric sensor is a technology that measures and recognizes physical characteristics." If the emotion recognition method determines that the user is confused, a more detailed explanation is added. For example, specific examples such as "This includes fingerprint and facial recognition" are provided.

[1270] Example prompt sentence:

[1271] My customer seems confused. Can you explain more about biometric sensors?

[1272] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1273] Step 1:

[1274] The device captures the surrounding audio in real time. Here, the microphone built into the smart glasses is used. The input is the surrounding audio, and the output is audio data.

[1275] Step 2:

[1276] The voice data acquired by the device is sent to a server via the Internet. The input is the voice data, and the output is the voice data sent to the server.

[1277] Step 3:

[1278] The server converts the received voice data into text data using a voice recognition means. Here, a voice recognition algorithm (e.g., Google's voice recognition service) is used. The input is voice data, and the output is text data.

[1279] Step 4:

[1280] The server analyzes the text data using a generative AI model. The analysis means identifies specific terms from the text data. The input is the text data, and the output is the analysis results that include the specific terms.

[1281] Step 5:

[1282] The server generates an explanatory text using an explanatory text generation means based on the analysis results. The input is the analysis results, and the output is the explanatory text.

[1283] Step 6:

[1284] The server analyzes the user's emotional state using an emotion recognition means. The emotion recognition means analyzes the user's tone of voice and facial expressions to determine whether the user is confused or understanding. The input is the user's tone of voice and facial expressions, and the output is the analysis result of the user's emotional state.

[1285] Step 7:

[1286] The server adjusts the commentary based on the analysis of the emotional state. If the user is confused, it adds a detailed explanation, and if the user understands, it simplifies it. The input is the analysis of the emotional state and the commentary, and the output is the adjusted commentary.

[1287] Step 8:

[1288] The server sends the generated commentary to the terminal. The input is the adjusted commentary, and the output is the commentary sent to the smart glasses.

[1289] Step 9:

[1290] The terminal uses the commentary providing means to provide the commentary text to the user in real time via an audio device of the smart glasses, the input being the adjusted commentary text and the output being the audio commentary provided to the user.

[1291] 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.

[1292] 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.

[1293] 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.

[1294] 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.

[1295] 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.

[1296] 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.

[1297] 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).

[1298] 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.

[1299] 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."

[1300] 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.

[1301] 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).

[1302] 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.

[1303] 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.

[1304] 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.

[1305] 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.

[1306] 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.

[1307] 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.

[1308] 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.

[1309] 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.

[1310] 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.

[1311] 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.

[1312] The following is further disclosed regarding the above embodiment.

[1313] (Claim 1)

[1314] an audio acquisition means for acquiring surrounding audio in real time;

[1315] A speech recognition means for converting the acquired speech into text data;

[1316] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[1317] Explanation generation means for generating an explanation for a specific term;

[1318] an explanation providing means for providing the generated explanation to the user in real time;

[1319] A system including:

[1320] (Claim 2)

[1321] 2. The system of claim 1, wherein the explanation providing means provides the explanation to the user by voice via an audio device.

[1322] (Claim 3)

[1323] 2. The system according to claim 1, wherein the analysis means generates explanatory text by referring to background information related to a particular term.

[1324] "Example 1"

[1325] (Claim 1)

[1326] an audio acquisition means for acquiring surrounding audio in real time;

[1327] A speech recognition means for converting the acquired speech into text data;

[1328] an analysis means for analyzing text data using a generative artificial intelligence model and identifying portions containing technical terms;

[1329] Explanation generation means for generating explanations of technical terms;

[1330] an explanation providing means for providing the generated explanation to the user in real time;

[1331] A system including:

[1332] (Claim 2)

[1333] 2. The system according to claim 1, wherein the explanation providing means provides the explanation to the user by voice via an audio device.

[1334] (Claim 3)

[1335] 2. The system according to claim 1, wherein the analysis means generates the explanatory text by referring to background information related to the technical term.

[1336] "Application Example 1"

[1337] (Claim 1)

[1338] an audio acquisition means for acquiring surrounding audio in real time;

[1339] A speech recognition means for converting the acquired speech into text data;

[1340] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[1341] Explanation generation means for generating an explanation for a specific term;

[1342] an explanation providing means for providing the generated explanation to the user in real time;

[1343] A robot that can understand work instructions and technical terms in a factory and provide explanations.

[1344] A system including:

[1345] (Claim 2)

[1346] 2. The system of claim 1, wherein the explanation providing means provides the explanation to the user by voice via an audio device.

[1347] (Claim 3)

[1348] 2. The system according to claim 1, wherein the analysis means generates explanatory text by referring to background information related to a particular term.

[1349] "Example 2: Combining Emotion Engines"

[1350] (Claim 1)

[1351] an audio acquisition means for acquiring surrounding audio in real time;

[1352] A speech recognition means for converting the acquired speech into text data;

[1353] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[1354] Explanation generation means for generating an explanation for a specific term;

[1355] an explanation providing means for providing the generated explanation to the user in real time;

[1356] an emotion recognition means for recognizing an emotional state of a user and adjusting the commentary;

[1357] A system including:

[1358] (Claim 2)

[1359] 2. The system of claim 1, wherein the explanation providing means provides the explanation to the user by voice via an audio device.

[1360] (Claim 3)

[1361] 2. The system according to claim 1, wherein the analysis means generates explanatory text by referring to background information related to a particular term.

[1362] "Application example 2 when combining emotion engines"

[1363] (Claim 1)

[1364] an audio acquisition means for acquiring surrounding audio in real time;

[1365] A speech recognition means for converting the acquired speech into text data;

[1366] an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms;

[1367] Explanation generation means for generating an explanation for a specific term;

[1368] an emotion recognition means for recognizing the emotional state of a user and adjusting the commentary;

[1369] An explanation providing means for providing the generated explanation to the user in real time by voice through smart glasses;

[1370] A system including:

[1371] (Claim 2)

[1372] 10. The system of claim 1, wherein the generated commentary is provided to the user audibly via an audio device of the smart glasses.

[1373] (Claim 3)

[1374] 2. The system of claim 1, wherein the analysis means generates explanatory text by referring to background information related to a particular term, and adjusts the explanatory text based on the user's emotional state. [Explanation of symbols]

[1375] 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. an audio acquisition means for acquiring surrounding audio in real time; A speech recognition means for converting the acquired speech into text data; an analysis means for analyzing text data using a generative AI model and identifying portions containing specific terms; Explanation generation means for generating an explanation for a specific term; an explanation providing means for providing the generated explanation to the user in real time; A system including:

2. 2. The system according to claim 1, wherein the explanation providing means provides the explanation to the user by voice via an audio device.

3. 2. The system according to claim 1, wherein the analysis means generates the explanatory text by referring to background information related to a specific term.

Images