system

Abstract

Provide a system. A device for receiving voice input; means for converting voice data into text data; A means for generating minutes based on text data; A method for analyzing the text data of minutes to infer the personalities and thoughts of participants, a means for generating advice to facilitate consensus building; a means for displaying the generated advice; A system including:

Description

[Technical Field]

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

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

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

[0004] In today's world, creating minutes from audio recordings of meetings and discussions and then analyzing the participants' personalities and ways of thinking from those minutes requires a great deal of time and effort. Conventional methods often delay the process of reconciling conflicts of opinion and building consensus, making efficient decision-making difficult. Therefore, there is a need for a system that can automatically generate minutes from meeting audio and support consensus building to reconcile the opinions of participants. [Means for solving the problem]

[0005] In order to solve the above-mentioned problems, the present invention provides the following means: a system including means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, and means for displaying the generated advice. This system makes it possible to automatically record minutes of the audio of a meeting and support efficient consensus building based on the minutes.

[0006] The "means for receiving voice input" refers to a device for collecting voice data from a meeting, discussion, etc., and a system having such a function.

[0007] "Means for converting voice data into text data" refers to technology and devices for analyzing voice signals and converting their contents into text information.

[0008] The "means for generating minutes based on text data" is a system for aggregating converted text data and creating an organized document as a record of meetings and discussions.

[0009] "Means for analyzing text data in minutes and inferring participants' personalities and ways of thinking" is a system that uses natural language processing technology to analyze the text in minutes and infer the personality and thought patterns of the speaker from their choice of words and content.

[0010] The "means for generating advice to promote consensus building" is a system for coordinating opinions and presenting specific guidelines for action based on the results of an analysis of the participants' personalities and ways of thinking.

[0011] The "means for displaying the generated advice" refers to a display device or a system having a screen display function for displaying the advice generated to promote consensus building in an easy-to-understand manner to the user. [Brief explanation of the drawings]

[0012] [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

[0013] 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.

[0014] First, the terms used in the following description will be explained.

[0015] 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).

[0016] 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.

[0017] 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.

[0018] 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.

[0019] 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."

[0020] [First embodiment]

[0021] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0022] 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.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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."

[0033] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0034] First, the user applies a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives the audio input in real time and saves it locally as an audio file. The device also transmits the received audio data to a server via the Internet.

[0035] The server processes the received voice data and converts it into text using a speech recognition engine. At this time, it identifies the content of each speaker's speech, formats the converted text, and generates meeting minutes. For example, minutes could be created in the format "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0036] The server then analyzes the text data of the generated minutes using natural language processing technology. This analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of the participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0037] The server then generates advice based on the analysis results to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0038] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0039] This system automatically generates meeting minutes based on audio data from meetings and analyzes the personalities and ways of thinking of the participants, enabling efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, the system can present effective methods for consolidating differing opinions, increasing the sense of satisfaction of the entire team.

[0040] The above is a specific embodiment of the present invention.

[0041] The processing flow will be explained below.

[0042] Step 1:

[0043] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0044] Step 2:

[0045] The server receives the voice data sent from the terminal and converts the received voice data into text data using a voice recognition engine (for example, Google (registered trademark) Cloud Speech-to-Text API). At this time, the content of each speaker's speech is identified and the converted text data is classified.

[0046] Step 3:

[0047] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0048] Step 4:

[0049] The server analyzes the generated minutes' text data using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0050] Step 5:

[0051] Based on the analysis results, the server generates specific advice to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0052] Step 6:

[0053] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice to guide the progress of the meeting and make decisions.

[0054] In this way, the system performs a series of processes from voice input to generating and analyzing minutes and providing advice on consensus building.

[0055] Example 1

[0056] 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."

[0057] Accurately and efficiently recording the contents of meetings and discussions and automatically generating meeting minutes based on that information is important, especially for large organizations and projects. However, current systems do not adequately process audio data in real time, analyze participants' personalities and ways of thinking, or generate advice to promote consensus building. This results in time-consuming meetings and decision-making, reducing productivity.

[0058] 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.

[0059] In this invention, the server includes means for locally saving voice data, means for transmitting the saved voice data to the server, means for converting the voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, and means for displaying the generated advice. This makes it possible to convert the contents of the meeting into text in real time and provide analysis and advice based on the contents.

[0060] A "means for receiving audio input" is a device or method for collecting and transmitting audio from a meeting or discussion to the system.

[0061] The "means for locally storing voice data" refers to a device or method for temporarily recording collected voice data in a storage device of the terminal.

[0062] "Means for transmitting stored audio data to a server" refers to a device or method for uploading locally stored audio data to a server via the Internet.

[0063] "Means for converting voice data into text data" refers to a device or method for converting voice data into text data using voice recognition technology.

[0064] The "means for generating minutes from text data" refers to a device or method for organizing and formatting the converted text data and compiling it into minutes of the meeting.

[0065] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to devices and methods that use natural language processing technology to analyze and infer the personalities and thought patterns of speakers from the contents of minutes.

[0066] The "means for generating advice to promote consensus building" refers to a device or method for generating specific advice to help reach consensus in a meeting based on the analysis results.

[0067] The "means for displaying the generated advice" refers to a device or method for visually presenting the generated advice to the user.

[0068] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0069] First, the user sends audio input to the system using a microphone to collect the conference audio. The device receives this audio input in real time and saves it locally as an audio file. It is desirable for the device to have a high-quality microphone and audio recording software built in. For example, a microphone built into a PC or tablet and a dedicated audio recording application can be used.

[0070] The device then transmits the saved audio data over the Internet to a server, preferably using the HTTP protocol and a secure TLS connection, where it is temporarily stored in cloud storage and then transferred to a server for analysis.

[0071] The server converts the received voice data into text data using a speech recognition service. Specifically, it uses a speech recognition engine such as Google Cloud Speech-to-Text API. This engine analyzes the voice data and generates corresponding text data. For example, minutes are created in the format "Speaker A: We should adopt an agile methodology for progressing the project" and "Speaker B: I agree, but we should also consider a waterfall methodology for some phases."

[0072] The server then analyzes the text data in the minutes using natural language processing techniques. This includes sentiment analysis, keyword extraction, and linguistic style analysis, and is performed using Python libraries such as NLTK and spaCy. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0073] Furthermore, the server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI (registered trademark) GPT-4 (registered trademark)) to generate advice such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0074] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0075] As a concrete example, consider a situation where a user is discussing how to proceed with a project. If a user says, "We should adopt an agile methodology," during a meeting, the system recognizes the statement and records it in the minutes as "Speaker A: We should adopt an agile methodology." The system then analyzes the entire statement and determines that "Speaker A tends to value flexibility," generating advice such as, "We recommend adopting an agile methodology in the early stages of the project, and combining it with a waterfall methodology as the project progresses." These processes are implemented using Python scripts, speech recognition services, natural language processing libraries, and generative AI models.

[0076] Example prompt sentence:

[0077] 1. Please explain in detail how you convert meeting audio data into text and how you perform analytics based on that data.

[0078] 2. "Please explain in detail how you use a generative AI model to generate advice based on analysis of meeting minutes."

[0079] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0080] Step 1:

[0081] The user sends audio input to the system using a microphone to collect the conference audio. Specifically, when the user presses the record button at the start of the conference, the microphone starts recording the audio. The input is the conference audio, and the output is audio data.

[0082] Step 2:

[0083] The device receives the audio and stores it locally as an audio file in real time. Specifically, the audio recording software receives the audio stream and stores it as a .wav file. The input is an audio stream, and the output is a local audio file.

[0084] Step 3:

[0085] The device sends the stored audio data to the server over the Internet. Specifically, it uploads the audio file to the server using the HTTP protocol. A secure TLS connection is ensured during this process. The input is the local audio file, and the output is the audio data on the server.

[0086] Step 4:

[0087] The server converts the received voice data into text data using a voice recognition engine. Here, for example, the Google Cloud Speech-to-Text API is used. Specifically, the voice data is sent to the API and the returned text data is received. The input is voice data and the output is text data.

[0088] Step 5:

[0089] The server analyzes the converted text data and generates minutes. Specifically, it analyzes the text data from the speech recognition engine, identifies and organizes the statements made by each speaker. For example, it formats the minutes in a format such as "Speaker A: This is an important point" and "Speaker B: I agree." The input is text data, and the output is the minutes.

[0090] Step 6:

[0091] The server analyzes the generated text data of the minutes using natural language processing technology. Specifically, it uses Python's NLTK and spaCy to perform sentiment analysis, keyword extraction, and linguistic style analysis. For example, it extracts each speaker's speaking tendencies and keywords. The input is the minutes, and the output is the analysis results.

[0092] Step 7:

[0093] The server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-4) to create optimal advice from the analysis results. For example, it generates advice such as "We recommend taking a flexible approach." The input is the analysis results, and the output is advice.

[0094] Step 8:

[0095] The server sends the generated advice to the device, which then visually displays it to the user. Specifically, we build a UI that displays the advice on the device's display. The input is the advice, and the output is a visual confirmation for the user.

[0096] (Application example 1)

[0097] 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."

[0098] Conventional systems for automatically generating meeting minutes are designed for use in conference rooms and lack the functionality to record and analyze conversations with passengers while traveling and provide advice. Furthermore, when passengers exchange opinions or hold discussions in real time in an in-car environment, there are insufficient means to properly record and share the content. Furthermore, since there is no effective in-car display of advice for consensus building based on the content of conversations, there is a need for a system that supports efficient decision-making while traveling.

[0099] 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.

[0100] In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, means for collecting in-vehicle voice data, means for transmitting the collected voice data to a cloud server in real time, and means for displaying the minutes in the autonomously driving vehicle. This makes it possible to automatically generate minutes of discussions held by passengers while traveling, and to analyze the opinions and ways of thinking of each passenger and provide advice to promote consensus building in real time.

[0101] The "means for receiving audio input" refers to a device in which a microphone installed in the vehicle captures audio data and transmits it to the system.

[0102] The "means for converting voice data into text data" refers to a device or program that uses voice recognition technology to convert collected voice data into text information in real time.

[0103] The "means for generating minutes based on text data" refers to a device or program that analyzes the converted text data, systematically organizes the contents of the meeting or discussion, and compiles them in the form of minutes.

[0104] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to a device or program that uses natural language processing technology to analyze the contents of minutes and infer the personalities and ways of thinking of each participant based on their remarks and emotions.

[0105] The "means for generating advice to promote consensus building" is a device or program that generates specific proposals and advice for achieving better consensus building based on the results of meeting minutes and participant analysis.

[0106] The "means for displaying the generated advice" is a display device or a program for visually presenting the generated advice to the user.

[0107] "Means for collecting voice data inside the vehicle" refers to a device or system that uses microphones and sensors installed inside the vehicle to record conversations between passengers and surrounding sounds.

[0108] "Means for transmitting collected voice data to a cloud server in real time" refers to a device or program that transfers collected voice data from a device such as an edge computer to a cloud server and processes it in real time.

[0109] "Means for displaying minutes within an autonomous vehicle" refers to a device or program that provides the generated minutes and advice to the occupants through a display or various display devices within the autonomous vehicle.

[0110] The present invention provides a system for recording discussions and conversations among occupants in an autonomous vehicle, generating minutes in real time, and promoting consensus building. Specific embodiments of the system are described below.

[0111] System Configuration

[0112] The system has the following configuration:

[0113] 1. In-car audio collection means:

[0114] A microphone is installed inside the vehicle to collect passengers' comments and surrounding sounds in real time, allowing discussions to be recorded sequentially.

[0115] 2. Edge computers:

[0116] Using an edge computer (e.g., NVIDIA Jetson Nano) installed in the vehicle, the collected voice data is temporarily stored locally and transmitted to a cloud server via the Internet.

[0117] 3. Cloud Server:

[0118] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the received voice data into text data.

[0119] 4. Minutes generation method:

[0120] Based on the converted text data, the system identifies what each speaker said and generates minutes, systematically organizing the conversation and recording it in an easy-to-understand format.

[0121] 5. Natural Language Processing (NLP):

[0122] The text data in the minutes is analyzed using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. Specifically, sentiment analysis, keyword extraction, and analysis of speaking style are performed.

[0123] 6. Advice Generation Methods:

[0124] Based on the analysis results, advice is automatically generated to promote consensus building, for example by providing suggestions to reconcile differing opinions and increase overall satisfaction.

[0125] 7. Display means:

[0126] The generated advice and minutes are visually displayed on a display installed inside the vehicle, allowing passengers to receive advice in real time and reflect it in the progress of the discussion.

[0127] Processing flow

[0128] After receiving the voice data, the server performs the following processing steps.

[0129] 1. Use a speech recognition engine to convert voice data into text data.

[0130] 2. Analyze the converted text data and generate meeting minutes.

[0131] 3. Use natural language processing technology to analyze the contents of the minutes and infer the personalities and thoughts of the participants.

[0132] 4. Based on the analysis results, advice is generated to promote consensus building.

[0133] 5. Display the generated advice on the in-car display.

[0134] Hardware and software used

[0135] Microphone: A device used to collect speech from inside the vehicle.

[0136] Edge computer: A device that temporarily stores voice data and transmits it to a cloud server (e.g., NVIDIA Jetson Nano).

[0137] Cloud server: Processes and analyzes audio data.

[0138] Speech recognition engine: Software that converts voice data into text data (e.g., Google Cloud Speech-to-Text).

[0139] Natural language processing technology: Software that analyzes text data (e.g., NLTK, Google Cloud Natural Language).

[0140] Display: An in-vehicle display device for showing advice and minutes.

[0141] Specific examples

[0142] For example, if a family is driving and discussing their next destination, the system can analyze the following prompts:

[0143] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0144] Based on this, the system generates the minutes as follows:

[0145] Father: Next time I want to go to the beach.

[0146] Mother: I think the mountains are better.

[0147] Child A: I want to go to the zoo.

[0148] Child B: An amusement park would be nice.

[0149] The analysis then generates advice that respects each individual's opinion and promotes consensus building:

[0150] "Consider going somewhere that's close to both the beach and the mountains, or plan another destination for next weekend."

[0151] In this way, the system can generate meeting minutes and provide advice in real time, supporting efficient and smooth decision-making while on the move.

[0152] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0153] Step 1:

[0154] The device collects voice data in real time from microphones installed in various locations inside the vehicle. The voice data input is the conversations of passengers and the surrounding environmental sounds, and this data is temporarily stored in digital format on the edge computer. This allows the content of the conversation to be properly recorded.

[0155] Step 2:

[0156] The device transmits the temporarily stored voice data to a cloud server via the Internet. The input data is a digitized voice file, and the output is the arrival of the voice file at the cloud server. This transmission process utilizes a stable network connection and an appropriate data transfer protocol.

[0157] Step 3:

[0158] The server receives the audio file sent to the cloud server and converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input is the audio file, and the output is the corresponding text data. The speech recognition engine analyzes the audio waveform and performs a process to identify words.

[0159] Step 4:

[0160] The server automatically generates minutes based on the converted text data. It analyzes the text data, identifies the content of each speaker's speech, and creates minutes by formatting them into a specific format (e.g., "Speaker A: ○○"). The input is the speech-recognized text data, and the output is the text of the minutes. This operation uses a text analysis algorithm to organize the order and content of the speech.

[0161] Step 5:

[0162] The server analyzes the automatically generated text data of the minutes using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. The input is the text data of the minutes, and the output is an inference about the personality and way of thinking of each participant. This process uses multiple NLP techniques, including sentiment analysis and keyword extraction.

[0163] Step 6:

[0164] The server generates advice to promote consensus building based on the inference results. The input is the inference results about each participant's personality and way of thinking, and the output is specific advice text. This operation uses a generative AI model to create the most appropriate advice at that time.

[0165] Step 7:

[0166] The terminal displays the generated advice received from the server on the in-car display. The input is text data of the advice, and the output is displayed on the display. This display function allows passengers to directly check the advice and use its content for discussion and consensus building.

[0167] Prompt Sentence Examples

[0168] For example, if a family is driving and discussing their next destination, they might use prompts like the following as input:

[0169] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0170] 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.

[0171] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0172] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0173] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0174] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0175] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0176] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0177] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0178] In this way, the present invention automatically generates meeting minutes based on the audio data of a meeting and analyzes the personalities, thoughts, and emotions of the participants to support efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0179] The above is a specific embodiment of the present invention.

[0180] The processing flow will be explained below.

[0181] Step 1:

[0182] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time, stores the audio data locally, and transmits the received audio data to a server via the Internet.

[0183] Step 2:

[0184] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). During this conversion process, the content of each speaker's speech is identified and the converted text data is classified for each speech.

[0185] Step 3:

[0186] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile approach here" and "Speaker B: I agree, but we should also consider a waterfall approach in some phases."

[0187] Step 4:

[0188] The server analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. This allows the server to infer the speaker's personality and way of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0189] Step 5:

[0190] The server uses an emotion engine to recognize the user's emotions from the voice and text data. The emotion engine analyzes the emotional tone of each utterance and labels it with an emotion, such as "excited," "calm," or "doubt."

[0191] Step 6:

[0192] The server integrates the emotion data recognized by the emotion engine with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach here," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0193] Step 7:

[0194] Based on the analysis results and emotional data, the server generates specific advice to promote consensus building. This advice becomes more appropriate by taking emotional states into account. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while appropriately monitoring emotions at each phase."

[0195] Step 8:

[0196] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice when proceeding with the meeting or making decisions.

[0197] Through these steps, the system performs a series of processes, from voice input to generating minutes, analyzing emotions, and providing advice on consensus building.

[0198] Example 2

[0199] 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."

[0200] There is a need for a system that can automatically generate minutes based on audio data from meetings and discussions, analyze the emotions as well as the personalities and ways of thinking of participants, and support efficient and harmonious consensus building. However, current systems do not adequately process audio data in real time or analyze the emotions of participants, which leaves issues with the quality and speed of consensus building.

[0201] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for temporarily storing voice data and transmitting it via the Internet, a means for converting the received voice data into text data, and a means for generating minutes based on the identified and classified text data. This enables real-time voice data processing, highly accurate text conversion, and analysis including the emotions and thoughts of participants.

[0202] "Audio data" refers to data in which audio information from meetings, discussions, etc. is recorded in digital format.

[0203] "Real time" means that processing or operations are carried out at the same speed as real time.

[0204] "Receiving" means that a system or device acquires information from the outside.

[0205] "Text data" refers to data obtained by converting voice data into character information.

[0206] "Identification and classification" refers to recognizing data or information that has specific characteristics from multiple sources and organizing it into categories.

[0207] "Minutes" are documents that record the contents of meetings and discussions.

[0208] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate natural human language.

[0209] "Emotion data" is data that includes emotional information extracted from speech or text.

[0210] "Analysis" means the detailed analysis of data or information.

[0211] "Advice" is a suggestion or guidance offered in response to a specific problem or situation.

[0212] "Display" means visually showing information to a user.

[0213] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0214] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0215] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0216] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0217] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0218] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0219] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0220] For example, when discussing how to proceed with a project, it can provide an effective approach that suits the emotional state of the person speaking and increase the sense of acceptance of the entire team. This increases satisfaction with the decisions made and strengthens team cooperation.

[0221] Examples of specific prompts are:

[0222] "Using audio data from a meeting, analyze the personalities and emotions of participants and generate advice to promote consensus building."

[0223] With these prompts, the generative AI model performs the necessary speech recognition, natural language processing, sentiment analysis, and data classification to build a program that provides appropriate advice.

[0224] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0225] Step 1:

[0226] Users use microphones to collect audio data from meetings and send audio input to the system. The input is real-time audio data from meetings, and the output is audio data sent to the terminal. The specific operation is that the microphone captures audio waveforms and converts them into digital signals.

[0227] Step 2:

[0228] The device processes the received voice data in real time, temporarily stores it in local storage, and simultaneously transmits it to a server via the Internet. The input is voice data from the user, and the output is voice data transmitted to the server via the Internet. Specifically, the device transmits the voice data to the server using data streaming technology.

[0229] Step 3:

[0230] The server processes the received voice data and converts it into text data using a voice recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The input is the received voice data, and the output is text data. Specifically, the voice recognition model analyzes the sound waves of the voice data and generates the corresponding text.

[0231] Step 4:

[0232] The server analyzes the generated text data and identifies and classifies each speaker. The input is text data generated by a speech recognition engine, and the output is classified text in the form of a meeting minutes. Specifically, the text classification algorithm analyzes keywords and context to identify the speaker.

[0233] Step 5:

[0234] The server analyzes the text data of the minutes using natural language processing technology. The input is the text data of the minutes, and the output is information including the analyzed participants' personalities and thoughts. Specific operations include sentiment analysis, keyword extraction, and language style analysis using natural language processing technology.

[0235] Step 6:

[0236] The server uses an emotion engine to recognize emotions from the voice and text data and integrate them with the text data of the minutes. The input is the analyzed text data and voice data, and the output is the minutes with added emotional context. Specifically, the emotion analysis algorithm analyzes the tone of the voice and the wording of the text.

[0237] Step 7:

[0238] The server generates specific advice to promote consensus building based on the analysis results and emotional data. The input is the integrated minutes and analytical data, and the output is advice for consensus building. Specifically, the advice generation algorithm takes into account various factors (emotional state, meeting content, participants' personalities, etc.) to construct optimal advice.

[0239] Step 8:

[0240] The server sends the generated advice to the terminal, which then visually displays it to the user. The input is the generated advice, and the output is the visually displayed advice. The specific operation is for the terminal to display the received advice to the user through a GUI.

[0241] By using the above steps, the user can achieve a series of processes from collecting voice data to integrating emotion data and finally displaying advice.

[0242] (Application example 2)

[0243] 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."

[0244] Meetings and discussions in industrial production sites are important for increasing efficiency, but preparing meeting minutes and reaching consensus while taking into account participants' emotions takes time and effort. In particular, in large-scale projects, accurately reflecting the opinions of all participants and communicating their emotions is crucial for rapid and effective decision-making. However, with existing technologies, automatic conversion of voice data alone is insufficient, making it difficult to support effective consensus-building that takes emotions into account. Therefore, the present invention aims to solve these problems and realize efficient and harmonious meeting management in industrial production sites.

[0245] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, and means for automatically generating meeting minutes at an industrial production site and supporting consensus building by analyzing and taking into account the emotions of the participants. This automates the creation of meeting minutes and enables support for consensus building that takes into account the emotions of the participants.

[0246] The "means for receiving voice input" refers to a device or system for collecting voice information from a meeting or discussion as digital data.

[0247] "Means for converting voice data into text data" refers to technology or devices that convert collected voice information into text strings.

[0248] The "means for generating minutes based on text data" is a system that organizes converted text information and automatically generates documents that describe the contents of meetings and discussions.

[0249] "Means of analyzing text data in minutes and inferring the personalities and ways of thinking of participants" is a function that analyzes the character strings in the generated minutes and infers the personalities and thought patterns of the speakers.

[0250] The "means for generating advice to promote consensus building" is a technology that automatically creates specific proposals based on analyzed data to facilitate smooth consensus building among participants.

[0251] The "means for displaying the generated advice" refers to a device or screen for visually presenting the generated advice to the user.

[0252] "Means for automatically generating meeting minutes in industrial production sites and supporting consensus building by analyzing and considering the emotions of participants" is a system that automatically records the contents of meetings held at production sites and effectively supports consensus building by taking into consideration the emotional state of participants.

[0253] The present invention provides a system for efficiently and effectively managing meetings and discussions at industrial production sites and supporting consensus building among participants. Specific embodiments of the system are described below.

[0254] First, a user uses a microphone to collect the audio of the meeting and transmits the audio input to the system. This audio input is received in real time by a terminal at the industrial production site. The received audio data is stored locally and simultaneously transmitted to a server via the Internet.

[0255] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and is classified as text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0256] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning." This is done using natural language processing libraries such as NLTK and spaCy.

[0257] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes user emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. In this process, emotion analysis models such as TextBlob and Hugging Face are used. This generates minutes with added emotional context for the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project." "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0258] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice is made more appropriate by taking into account the emotional state. It uses OpenAI's generative AI model to generate advice by inputting prompt sentences such as the following:

[0259] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[0260] The generated advice is sent from the server to the device, which then visually displays it to the user. The user can refer to this advice when conducting meetings or making decisions. For example, the advice may suggest adopting an agile methodology in the early stages of a project, and considering combining it with a waterfall methodology as the project progresses, while monitoring the emotions at each phase.

[0261] In this way, the present invention automatically generates meeting minutes based on the audio data of meetings in industrial production sites, and supports efficient and harmonious consensus building by analyzing the personalities, ways of thinking, and emotions of the participants. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0262] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0263] Step 1:

[0264] The user collects the audio of the conference using a microphone and transmits the audio data to the system. The input is the audio data of the conference, and the output is the audio data received by the audio input device. In this step, the audio is input to the terminal in real time.

[0265] Step 2:

[0266] The device stores the received voice data locally and simultaneously transmits it to a server via the Internet. The input is the voice data received in real time, and the output is the voice data transmitted to the server. This allows the voice data to be stored and transmitted.

[0267] Step 3:

[0268] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is the voice data stored on the server, and the output is text data. The converted text data is classified by each speaker.

[0269] Step 4:

[0270] The server automatically generates minutes based on the text data. The input is classified text data, and the output is organized minutes data. In this step, each statement is converted into minutes format.

[0271] Step 5:

[0272] The server analyzes the generated text data of the minutes using natural language processing (NLP). The input is the text data of the minutes, and the output is analytical data on the participants' personalities, thoughts, emotions, etc. Specifically, NLTK and spaCy are used to extract keywords and perform sentiment analysis on the text data.

[0273] Step 6:

[0274] The server uses an emotion engine to recognize user emotions from voice and text data, and integrates the recognized emotion data with the text data of the minutes. The input is voice and text data, and the output is the minutes data with added emotional context. It uses emotion analysis models such as TextBlob and Hugging Face.

[0275] Step 7:

[0276] The server generates specific advice to promote consensus building based on the analysis results and emotion data. The input is the analyzed minutes data and emotion data, and the output is advice data to support consensus building. Using OpenAI's generative AI model, the server inputs the following prompt:

[0277] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[0278] Step 8:

[0279] The generated advice is sent from the server to the terminal, and the terminal visually displays this advice to the user. The input is the generated advice data, and the output is a visual display of the advice to the user. This allows the user to refer to the advice and proceed with the meeting or make decisions.

[0280] 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.

[0281] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0282] 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.

[0283] [Second embodiment]

[0284] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0285] 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.

[0286] 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).

[0287] 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.

[0288] 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.

[0289] 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).

[0290] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0291] 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.

[0292] 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.

[0293] 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.

[0294] 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.

[0295] 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."

[0296] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0297] First, the user applies a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives the audio input in real time and saves it locally as an audio file. The device also transmits the received audio data to a server via the Internet.

[0298] The server processes the received voice data and converts it into text using a speech recognition engine. At this time, it identifies the content of each speaker's speech, formats the converted text, and generates meeting minutes. For example, minutes could be created in the format "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0299] The server then analyzes the text data of the generated minutes using natural language processing technology. This analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of the participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0300] The server then generates advice based on the analysis results to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0301] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0302] This system automatically generates meeting minutes based on audio data from meetings and analyzes the personalities and ways of thinking of the participants, enabling efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, the system can present effective methods for consolidating differing opinions, increasing the sense of satisfaction of the entire team.

[0303] The above is a specific embodiment of the present invention.

[0304] The processing flow will be explained below.

[0305] Step 1:

[0306] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0307] Step 2:

[0308] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). At this time, it identifies the content of each speaker's speech and classifies the converted text data.

[0309] Step 3:

[0310] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0311] Step 4:

[0312] The server analyzes the generated minutes' text data using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0313] Step 5:

[0314] Based on the analysis results, the server generates specific advice to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0315] Step 6:

[0316] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice to guide the progress of the meeting and make decisions.

[0317] In this way, the system performs a series of processes from voice input to generating and analyzing minutes and providing advice on consensus building.

[0318] Example 1

[0319] 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."

[0320] Accurately and efficiently recording the contents of meetings and discussions and automatically generating meeting minutes based on that information is important, especially for large organizations and projects. However, current systems do not adequately process audio data in real time, analyze participants' personalities and ways of thinking, or generate advice to promote consensus building. This results in time-consuming meetings and decision-making, reducing productivity.

[0321] 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.

[0322] In this invention, the server includes means for locally saving voice data, means for transmitting the saved voice data to the server, means for converting the voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, and means for displaying the generated advice. This makes it possible to convert the contents of the meeting into text in real time and provide analysis and advice based on the contents.

[0323] A "means for receiving audio input" is a device or method for collecting and transmitting audio from a meeting or discussion to the system.

[0324] The "means for locally storing voice data" refers to a device or method for temporarily recording collected voice data in a storage device of the terminal.

[0325] "Means for transmitting stored audio data to a server" refers to a device or method for uploading locally stored audio data to a server via the Internet.

[0326] "Means for converting voice data into text data" refers to a device or method for converting voice data into text data using voice recognition technology.

[0327] The "means for generating minutes from text data" refers to a device or method for organizing and formatting the converted text data and compiling it into minutes of the meeting.

[0328] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to devices and methods that use natural language processing technology to analyze and infer the personalities and thought patterns of speakers from the contents of minutes.

[0329] The "means for generating advice to promote consensus building" refers to a device or method for generating specific advice to help reach consensus in a meeting based on the analysis results.

[0330] The "means for displaying the generated advice" refers to a device or method for visually presenting the generated advice to the user.

[0331] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0332] First, the user sends audio input to the system using a microphone to collect the conference audio. The device receives this audio input in real time and saves it locally as an audio file. It is desirable for the device to have a high-quality microphone and audio recording software built in. For example, a microphone built into a PC or tablet and a dedicated audio recording application can be used.

[0333] The device then transmits the saved audio data over the Internet to a server, preferably using the HTTP protocol and a secure TLS connection, where it is temporarily stored in cloud storage and then transferred to a server for analysis.

[0334] The server converts the received voice data into text data using a speech recognition service. Specifically, it uses a speech recognition engine such as Google Cloud Speech-to-Text API. This engine analyzes the voice data and generates corresponding text data. For example, minutes are created in the format "Speaker A: We should adopt an agile methodology for progressing the project" and "Speaker B: I agree, but we should also consider a waterfall methodology for some phases."

[0335] The server then analyzes the text data in the minutes using natural language processing techniques. This includes sentiment analysis, keyword extraction, and linguistic style analysis, and is performed using Python libraries such as NLTK and spaCy. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0336] Furthermore, the server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-4) to generate advice such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0337] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0338] As a concrete example, consider a situation where a user is discussing how to proceed with a project. If a user says, "We should adopt an agile methodology," during a meeting, the system recognizes the statement and records it in the minutes as "Speaker A: We should adopt an agile methodology." The system then analyzes the entire statement and determines that "Speaker A tends to value flexibility," generating advice such as, "We recommend adopting an agile methodology in the early stages of the project, and combining it with a waterfall methodology as the project progresses." These processes are implemented using Python scripts, speech recognition services, natural language processing libraries, and generative AI models.

[0339] Example prompt sentence:

[0340] 1. Please explain in detail how you convert meeting audio data into text and how you perform analytics based on that data.

[0341] 2. "Please explain in detail how you use a generative AI model to generate advice based on analysis of meeting minutes."

[0342] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0343] Step 1:

[0344] The user sends audio input to the system using a microphone to collect the conference audio. Specifically, when the user presses the record button at the start of the conference, the microphone starts recording the audio. The input is the conference audio, and the output is audio data.

[0345] Step 2:

[0346] The device receives the audio and stores it locally as an audio file in real time. Specifically, the audio recording software receives the audio stream and stores it as a .wav file. The input is an audio stream, and the output is a local audio file.

[0347] Step 3:

[0348] The device sends the stored audio data to the server over the Internet. Specifically, it uploads the audio file to the server using the HTTP protocol. A secure TLS connection is ensured during this process. The input is the local audio file, and the output is the audio data on the server.

[0349] Step 4:

[0350] The server converts the received voice data into text data using a voice recognition engine. Here, for example, the Google Cloud Speech-to-Text API is used. Specifically, the voice data is sent to the API and the returned text data is received. The input is voice data and the output is text data.

[0351] Step 5:

[0352] The server analyzes the converted text data and generates minutes. Specifically, it analyzes the text data from the speech recognition engine, identifies and organizes the statements made by each speaker. For example, it formats the minutes in a format such as "Speaker A: This is an important point" and "Speaker B: I agree." The input is text data, and the output is the minutes.

[0353] Step 6:

[0354] The server analyzes the generated text data of the minutes using natural language processing technology. Specifically, it uses Python's NLTK and spaCy to perform sentiment analysis, keyword extraction, and linguistic style analysis. For example, it extracts each speaker's speaking tendencies and keywords. The input is the minutes, and the output is the analysis results.

[0355] Step 7:

[0356] The server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-4) to create optimal advice from the analysis results. For example, it generates advice such as "We recommend taking a flexible approach." The input is the analysis results, and the output is advice.

[0357] Step 8:

[0358] The server sends the generated advice to the device, which then visually displays it to the user. Specifically, we build a UI that displays the advice on the device's display. The input is the advice, and the output is a visual confirmation for the user.

[0359] (Application example 1)

[0360] 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."

[0361] Conventional systems for automatically generating meeting minutes are designed for use in conference rooms and lack the functionality to record and analyze conversations with passengers while traveling and provide advice. Furthermore, when passengers exchange opinions or hold discussions in real time in an in-car environment, there are insufficient means to properly record and share the content. Furthermore, since there is no effective in-car display of advice for consensus building based on the content of conversations, there is a need for a system that supports efficient decision-making while traveling.

[0362] 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.

[0363] In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, means for collecting in-vehicle voice data, means for transmitting the collected voice data to a cloud server in real time, and means for displaying the minutes in the autonomously driving vehicle. This makes it possible to automatically generate minutes of discussions held by passengers while traveling, and to analyze the opinions and ways of thinking of each passenger and provide advice to promote consensus building in real time.

[0364] The "means for receiving audio input" refers to a device in which a microphone installed in the vehicle captures audio data and transmits it to the system.

[0365] The "means for converting voice data into text data" refers to a device or program that uses voice recognition technology to convert collected voice data into text information in real time.

[0366] The "means for generating minutes based on text data" refers to a device or program that analyzes the converted text data, systematically organizes the contents of the meeting or discussion, and compiles them in the form of minutes.

[0367] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to a device or program that uses natural language processing technology to analyze the contents of minutes and infer the personalities and ways of thinking of each participant based on their remarks and emotions.

[0368] The "means for generating advice to promote consensus building" is a device or program that generates specific proposals and advice for achieving better consensus building based on the results of meeting minutes and participant analysis.

[0369] The "means for displaying the generated advice" is a display device or a program for visually presenting the generated advice to the user.

[0370] "Means for collecting voice data inside the vehicle" refers to a device or system that uses microphones and sensors installed inside the vehicle to record conversations between passengers and surrounding sounds.

[0371] "Means for transmitting collected voice data to a cloud server in real time" refers to a device or program that transfers collected voice data from a device such as an edge computer to a cloud server and processes it in real time.

[0372] "Means for displaying minutes within an autonomous vehicle" refers to a device or program that provides the generated minutes and advice to the occupants through a display or various display devices within the autonomous vehicle.

[0373] The present invention provides a system for recording discussions and conversations among occupants in an autonomous vehicle, generating minutes in real time, and promoting consensus building. Specific embodiments of the system are described below.

[0374] System Configuration

[0375] The system has the following configuration:

[0376] 1. In-car audio collection means:

[0377] A microphone is installed inside the vehicle to collect passengers' comments and surrounding sounds in real time, allowing discussions to be recorded sequentially.

[0378] 2. Edge computers:

[0379] Using an edge computer (e.g., NVIDIA Jetson Nano) installed in the vehicle, the collected voice data is temporarily stored locally and transmitted to a cloud server via the Internet.

[0380] 3. Cloud Server:

[0381] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the received voice data into text data.

[0382] 4. Minutes generation method:

[0383] Based on the converted text data, the system identifies what each speaker said and generates minutes, systematically organizing the conversation and recording it in an easy-to-understand format.

[0384] 5. Natural Language Processing (NLP):

[0385] The text data in the minutes is analyzed using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. Specifically, sentiment analysis, keyword extraction, and analysis of speaking style are performed.

[0386] 6. Advice Generation Methods:

[0387] Based on the analysis results, advice is automatically generated to promote consensus building, for example by providing suggestions to reconcile differing opinions and increase overall satisfaction.

[0388] 7. Display means:

[0389] The generated advice and minutes are visually displayed on a display installed inside the vehicle, allowing passengers to receive advice in real time and reflect it in the progress of the discussion.

[0390] Processing flow

[0391] After receiving the voice data, the server performs the following processing steps.

[0392] 1. Use a speech recognition engine to convert voice data into text data.

[0393] 2. Analyze the converted text data and generate meeting minutes.

[0394] 3. Use natural language processing technology to analyze the contents of the minutes and infer the personalities and thoughts of the participants.

[0395] 4. Based on the analysis results, advice is generated to promote consensus building.

[0396] 5. Display the generated advice on the in-car display.

[0397] Hardware and software used

[0398] Microphone: A device used to collect speech from inside the vehicle.

[0399] Edge computer: A device that temporarily stores voice data and transmits it to a cloud server (e.g., NVIDIA Jetson Nano).

[0400] Cloud server: Processes and analyzes audio data.

[0401] Speech recognition engine: Software that converts voice data into text data (e.g., Google Cloud Speech-to-Text).

[0402] Natural language processing technology: Software that analyzes text data (e.g., NLTK, Google Cloud Natural Language).

[0403] Display: An in-vehicle display device for showing advice and minutes.

[0404] Specific examples

[0405] For example, if a family is driving and discussing their next destination, the system can analyze the following prompts:

[0406] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0407] Based on this, the system generates the minutes as follows:

[0408] Father: Next time I want to go to the beach.

[0409] Mother: I think the mountains are better.

[0410] Child A: I want to go to the zoo.

[0411] Child B: An amusement park would be nice.

[0412] The analysis then generates advice that respects each individual's opinion and promotes consensus building:

[0413] "Consider going somewhere that's close to both the beach and the mountains, or plan another destination for next weekend."

[0414] In this way, the system can generate meeting minutes and provide advice in real time, supporting efficient and smooth decision-making while on the move.

[0415] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0416] Step 1:

[0417] The device collects voice data in real time from microphones installed in various locations inside the vehicle. The voice data input is the conversations of passengers and the surrounding environmental sounds, and this data is temporarily stored in digital format on the edge computer. This allows the content of the conversation to be properly recorded.

[0418] Step 2:

[0419] The device transmits the temporarily stored voice data to a cloud server via the Internet. The input data is a digitized voice file, and the output is the arrival of the voice file at the cloud server. This transmission process utilizes a stable network connection and an appropriate data transfer protocol.

[0420] Step 3:

[0421] The server receives the audio file sent to the cloud server and converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input is the audio file, and the output is the corresponding text data. The speech recognition engine analyzes the audio waveform and performs a process to identify words.

[0422] Step 4:

[0423] The server automatically generates minutes based on the converted text data. It analyzes the text data, identifies the content of each speaker's speech, and creates minutes by formatting them into a specific format (e.g., "Speaker A: ○○"). The input is the speech-recognized text data, and the output is the text of the minutes. This operation uses a text analysis algorithm to organize the order and content of the speech.

[0424] Step 5:

[0425] The server analyzes the automatically generated text data of the minutes using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. The input is the text data of the minutes, and the output is an inference about the personality and way of thinking of each participant. This process uses multiple NLP techniques, including sentiment analysis and keyword extraction.

[0426] Step 6:

[0427] The server generates advice to promote consensus building based on the inference results. The input is the inference results about each participant's personality and way of thinking, and the output is specific advice text. This operation uses a generative AI model to create the most appropriate advice at that time.

[0428] Step 7:

[0429] The terminal displays the generated advice received from the server on the in-car display. The input is text data of the advice, and the output is displayed on the display. This display function allows passengers to directly check the advice and use its content for discussion and consensus building.

[0430] Prompt Sentence Examples

[0431] For example, if a family is driving and discussing their next destination, they might use prompts like the following as input:

[0432] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0433] 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.

[0434] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0435] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0436] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0437] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0438] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0439] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0440] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0441] In this way, the present invention automatically generates meeting minutes based on the audio data of a meeting and analyzes the personalities, thoughts, and emotions of the participants to support efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0442] The above is a specific embodiment of the present invention.

[0443] The processing flow will be explained below.

[0444] Step 1:

[0445] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time, stores the audio data locally, and transmits the received audio data to a server via the Internet.

[0446] Step 2:

[0447] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). During this conversion process, the content of each speaker's speech is identified and the converted text data is classified for each speech.

[0448] Step 3:

[0449] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile approach here" and "Speaker B: I agree, but we should also consider a waterfall approach in some phases."

[0450] Step 4:

[0451] The server analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. This allows the server to infer the speaker's personality and way of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0452] Step 5:

[0453] The server uses an emotion engine to recognize the user's emotions from the voice and text data. The emotion engine analyzes the emotional tone of each utterance and labels it with an emotion, such as "excited," "calm," or "doubt."

[0454] Step 6:

[0455] The server integrates the emotion data recognized by the emotion engine with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach here," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0456] Step 7:

[0457] Based on the analysis results and emotional data, the server generates specific advice to promote consensus building. This advice becomes more appropriate by taking emotional states into account. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while appropriately monitoring emotions at each phase."

[0458] Step 8:

[0459] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice when proceeding with the meeting or making decisions.

[0460] Through these steps, the system performs a series of processes, from voice input to generating minutes, analyzing emotions, and providing advice on consensus building.

[0461] Example 2

[0462] 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."

[0463] There is a need for a system that can automatically generate minutes based on audio data from meetings and discussions, analyze the emotions as well as the personalities and ways of thinking of participants, and support efficient and harmonious consensus building. However, current systems do not adequately process audio data in real time or analyze the emotions of participants, which leaves issues with the quality and speed of consensus building.

[0464] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for temporarily storing voice data and transmitting it via the Internet, a means for converting the received voice data into text data, and a means for generating minutes based on the identified and classified text data. This enables real-time voice data processing, highly accurate text conversion, and analysis including the emotions and thoughts of participants.

[0465] "Audio data" refers to data in which audio information from meetings, discussions, etc. is recorded in digital format.

[0466] "Real time" means that processing or operations are carried out at the same speed as real time.

[0467] "Receiving" means that a system or device acquires information from the outside.

[0468] "Text data" refers to data obtained by converting voice data into character information.

[0469] "Identification and classification" refers to recognizing data or information that has specific characteristics from multiple sources and organizing it into categories.

[0470] "Minutes" are documents that record the contents of meetings and discussions.

[0471] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate natural human language.

[0472] "Emotion data" is data that includes emotional information extracted from speech or text.

[0473] "Analysis" means the detailed analysis of data or information.

[0474] "Advice" is a suggestion or guidance offered in response to a specific problem or situation.

[0475] "Display" means visually showing information to a user.

[0476] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0477] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0478] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0479] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0480] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0481] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0482] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0483] For example, when discussing how to proceed with a project, it can provide an effective approach that suits the emotional state of the person speaking and increase the sense of acceptance of the entire team. This increases satisfaction with the decisions made and strengthens team cooperation.

[0484] Examples of specific prompts are:

[0485] "Using audio data from a meeting, analyze the personalities and emotions of participants and generate advice to promote consensus building."

[0486] With these prompts, the generative AI model performs the necessary speech recognition, natural language processing, sentiment analysis, and data classification to build a program that provides appropriate advice.

[0487] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0488] Step 1:

[0489] Users use microphones to collect audio data from meetings and send audio input to the system. The input is real-time audio data from meetings, and the output is audio data sent to the terminal. The specific operation is that the microphone captures audio waveforms and converts them into digital signals.

[0490] Step 2:

[0491] The device processes the received voice data in real time, temporarily stores it in local storage, and simultaneously transmits it to a server via the Internet. The input is voice data from the user, and the output is voice data transmitted to the server via the Internet. Specifically, the device transmits the voice data to the server using data streaming technology.

[0492] Step 3:

[0493] The server processes the received voice data and converts it into text data using a voice recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The input is the received voice data, and the output is text data. Specifically, the voice recognition model analyzes the sound waves of the voice data and generates the corresponding text.

[0494] Step 4:

[0495] The server analyzes the generated text data and identifies and classifies each speaker. The input is text data generated by a speech recognition engine, and the output is classified text in the form of a meeting minutes. Specifically, the text classification algorithm analyzes keywords and context to identify the speaker.

[0496] Step 5:

[0497] The server analyzes the text data of the minutes using natural language processing technology. The input is the text data of the minutes, and the output is information including the analyzed participants' personalities and thoughts. Specific operations include sentiment analysis, keyword extraction, and language style analysis using natural language processing technology.

[0498] Step 6:

[0499] The server uses an emotion engine to recognize emotions from the voice and text data and integrate them with the text data of the minutes. The input is the analyzed text data and voice data, and the output is the minutes with added emotional context. Specifically, the emotion analysis algorithm analyzes the tone of the voice and the wording of the text.

[0500] Step 7:

[0501] The server generates specific advice to promote consensus building based on the analysis results and emotional data. The input is the integrated minutes and analytical data, and the output is advice for consensus building. Specifically, the advice generation algorithm takes into account various factors (emotional state, meeting content, participants' personalities, etc.) to construct optimal advice.

[0502] Step 8:

[0503] The server sends the generated advice to the terminal, which then visually displays it to the user. The input is the generated advice, and the output is the visually displayed advice. The specific operation is for the terminal to display the received advice to the user through a GUI.

[0504] By using the above steps, the user can achieve a series of processes from collecting voice data to integrating emotion data and finally displaying advice.

[0505] (Application example 2)

[0506] 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."

[0507] Meetings and discussions in industrial production sites are important for increasing efficiency, but preparing meeting minutes and reaching consensus while taking into account participants' emotions takes time and effort. In particular, in large-scale projects, accurately reflecting the opinions of all participants and communicating their emotions is crucial for rapid and effective decision-making. However, with existing technologies, automatic conversion of voice data alone is insufficient, making it difficult to support effective consensus-building that takes emotions into account. Therefore, the present invention aims to solve these problems and realize efficient and harmonious meeting management in industrial production sites.

[0508] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, and means for automatically generating meeting minutes at an industrial production site and supporting consensus building by analyzing and taking into account the emotions of the participants. This automates the creation of meeting minutes and enables support for consensus building that takes into account the emotions of the participants.

[0509] The "means for receiving voice input" refers to a device or system for collecting voice information from a meeting or discussion as digital data.

[0510] "Means for converting voice data into text data" refers to technology or devices that convert collected voice information into text strings.

[0511] The "means for generating minutes based on text data" is a system that organizes converted text information and automatically generates documents that describe the contents of meetings and discussions.

[0512] "Means of analyzing text data in minutes and inferring the personalities and ways of thinking of participants" is a function that analyzes the character strings in the generated minutes and infers the personalities and thought patterns of the speakers.

[0513] The "means for generating advice to promote consensus building" is a technology that automatically creates specific proposals based on analyzed data to facilitate smooth consensus building among participants.

[0514] The "means for displaying the generated advice" refers to a device or screen for visually presenting the generated advice to the user.

[0515] "Means for automatically generating meeting minutes in industrial production sites and supporting consensus building by analyzing and considering the emotions of participants" is a system that automatically records the contents of meetings held at production sites and effectively supports consensus building by taking into consideration the emotional state of participants.

[0516] The present invention provides a system for efficiently and effectively managing meetings and discussions at industrial production sites and supporting consensus building among participants. Specific embodiments of the system are described below.

[0517] First, a user uses a microphone to collect the audio of the meeting and transmits the audio input to the system. This audio input is received in real time by a terminal at the industrial production site. The received audio data is stored locally and simultaneously transmitted to a server via the Internet.

[0518] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and is classified as text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0519] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning." This is done using natural language processing libraries such as NLTK and spaCy.

[0520] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes user emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. In this process, emotion analysis models such as TextBlob and Hugging Face are used. This generates minutes with added emotional context for the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project." "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0521] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice is made more appropriate by taking into account the emotional state. It uses OpenAI's generative AI model to generate advice by inputting prompt sentences such as the following:

[0522] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[0523] The generated advice is sent from the server to the device, which then visually displays it to the user. The user can refer to this advice when conducting meetings or making decisions. For example, the advice may suggest adopting an agile methodology in the early stages of a project, and considering combining it with a waterfall methodology as the project progresses, while monitoring the emotions at each phase.

[0524] In this way, the present invention automatically generates meeting minutes based on the audio data of meetings in industrial production sites, and supports efficient and harmonious consensus building by analyzing the personalities, ways of thinking, and emotions of the participants. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0525] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0526] Step 1:

[0527] The user collects the audio of the conference using a microphone and transmits the audio data to the system. The input is the audio data of the conference, and the output is the audio data received by the audio input device. In this step, the audio is input to the terminal in real time.

[0528] Step 2:

[0529] The device stores the received voice data locally and simultaneously transmits it to a server via the Internet. The input is the voice data received in real time, and the output is the voice data transmitted to the server. This allows the voice data to be stored and transmitted.

[0530] Step 3:

[0531] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is the voice data stored on the server, and the output is text data. The converted text data is classified by each speaker.

[0532] Step 4:

[0533] The server automatically generates minutes based on the text data. The input is classified text data, and the output is organized minutes data. In this step, each statement is converted into minutes format.

[0534] Step 5:

[0535] The server analyzes the generated text data of the minutes using natural language processing (NLP). The input is the text data of the minutes, and the output is analytical data on the participants' personalities, thoughts, emotions, etc. Specifically, NLTK and spaCy are used to extract keywords and perform sentiment analysis on the text data.

[0536] Step 6:

[0537] The server uses an emotion engine to recognize user emotions from voice and text data, and integrates the recognized emotion data with the text data of the minutes. The input is voice and text data, and the output is the minutes data with added emotional context. It uses emotion analysis models such as TextBlob and Hugging Face.

[0538] Step 7:

[0539] The server generates specific advice to promote consensus building based on the analysis results and emotion data. The input is the analyzed minutes data and emotion data, and the output is advice data to support consensus building. Using OpenAI's generative AI model, the server inputs the following prompt:

[0540] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[0541] Step 8:

[0542] The generated advice is sent from the server to the terminal, and the terminal visually displays this advice to the user. The input is the generated advice data, and the output is a visual display of the advice to the user. This allows the user to refer to the advice and proceed with the meeting or make decisions.

[0543] 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.

[0544] 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.

[0545] 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.

[0546] [Third embodiment]

[0547] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0548] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0549] 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).

[0550] 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.

[0551] 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.

[0552] 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).

[0553] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0554] 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.

[0555] 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.

[0556] 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.

[0557] 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.

[0558] 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."

[0559] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0560] First, the user applies a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives the audio input in real time and saves it locally as an audio file. The device also transmits the received audio data to a server via the Internet.

[0561] The server processes the received voice data and converts it into text using a speech recognition engine. At this time, it identifies the content of each speaker's speech, formats the converted text, and generates meeting minutes. For example, minutes could be created in the format "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0562] The server then analyzes the text data of the generated minutes using natural language processing technology. This analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of the participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0563] The server then generates advice based on the analysis results to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0564] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0565] This system automatically generates meeting minutes based on audio data from meetings and analyzes the personalities and ways of thinking of the participants, enabling efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, the system can present effective methods for consolidating differing opinions, increasing the sense of satisfaction of the entire team.

[0566] The above is a specific embodiment of the present invention.

[0567] The processing flow will be explained below.

[0568] Step 1:

[0569] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0570] Step 2:

[0571] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). At this time, it identifies the content of each speaker's speech and classifies the converted text data.

[0572] Step 3:

[0573] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0574] Step 4:

[0575] The server analyzes the generated minutes' text data using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0576] Step 5:

[0577] Based on the analysis results, the server generates specific advice to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0578] Step 6:

[0579] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice to guide the progress of the meeting and make decisions.

[0580] In this way, the system performs a series of processes from voice input to generating and analyzing minutes and providing advice on consensus building.

[0581] Example 1

[0582] 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."

[0583] Accurately and efficiently recording the contents of meetings and discussions and automatically generating meeting minutes based on that information is important, especially for large organizations and projects. However, current systems do not adequately process audio data in real time, analyze participants' personalities and ways of thinking, or generate advice to promote consensus building. This results in time-consuming meetings and decision-making, reducing productivity.

[0584] 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.

[0585] In this invention, the server includes means for locally saving voice data, means for transmitting the saved voice data to the server, means for converting the voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, and means for displaying the generated advice. This makes it possible to convert the contents of the meeting into text in real time and provide analysis and advice based on the contents.

[0586] A "means for receiving audio input" is a device or method for collecting and transmitting audio from a meeting or discussion to the system.

[0587] The "means for locally storing voice data" refers to a device or method for temporarily recording collected voice data in a storage device of the terminal.

[0588] "Means for transmitting stored audio data to a server" refers to a device or method for uploading locally stored audio data to a server via the Internet.

[0589] "Means for converting voice data into text data" refers to a device or method for converting voice data into text data using voice recognition technology.

[0590] The "means for generating minutes from text data" refers to a device or method for organizing and formatting the converted text data and compiling it into minutes of the meeting.

[0591] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to devices and methods that use natural language processing technology to analyze and infer the personalities and thought patterns of speakers from the contents of minutes.

[0592] The "means for generating advice to promote consensus building" refers to a device or method for generating specific advice to help reach consensus in a meeting based on the analysis results.

[0593] The "means for displaying the generated advice" refers to a device or method for visually presenting the generated advice to the user.

[0594] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0595] First, the user sends audio input to the system using a microphone to collect the conference audio. The device receives this audio input in real time and saves it locally as an audio file. It is desirable for the device to have a high-quality microphone and audio recording software built in. For example, a microphone built into a PC or tablet and a dedicated audio recording application can be used.

[0596] The device then transmits the saved audio data over the Internet to a server, preferably using the HTTP protocol and a secure TLS connection, where it is temporarily stored in cloud storage and then transferred to a server for analysis.

[0597] The server converts the received voice data into text data using a speech recognition service. Specifically, it uses a speech recognition engine such as Google Cloud Speech-to-Text API. This engine analyzes the voice data and generates corresponding text data. For example, minutes are created in the format "Speaker A: We should adopt an agile methodology for progressing the project" and "Speaker B: I agree, but we should also consider a waterfall methodology for some phases."

[0598] The server then analyzes the text data in the minutes using natural language processing techniques. This includes sentiment analysis, keyword extraction, and linguistic style analysis, and is performed using Python libraries such as NLTK and spaCy. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0599] Furthermore, the server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-4) to generate advice such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0600] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0601] As a concrete example, consider a situation where a user is discussing how to proceed with a project. If a user says, "We should adopt an agile methodology," during a meeting, the system recognizes the statement and records it in the minutes as "Speaker A: We should adopt an agile methodology." The system then analyzes the entire statement and determines that "Speaker A tends to value flexibility," generating advice such as, "We recommend adopting an agile methodology in the early stages of the project, and combining it with a waterfall methodology as the project progresses." These processes are implemented using Python scripts, speech recognition services, natural language processing libraries, and generative AI models.

[0602] Example prompt sentence:

[0603] 1. Please explain in detail how you convert meeting audio data into text and how you perform analytics based on that data.

[0604] 2. "Please explain in detail how you use a generative AI model to generate advice based on analysis of meeting minutes."

[0605] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0606] Step 1:

[0607] The user sends audio input to the system using a microphone to collect the conference audio. Specifically, when the user presses the record button at the start of the conference, the microphone starts recording the audio. The input is the conference audio, and the output is audio data.

[0608] Step 2:

[0609] The device receives the audio and stores it locally as an audio file in real time. Specifically, the audio recording software receives the audio stream and stores it as a .wav file. The input is an audio stream, and the output is a local audio file.

[0610] Step 3:

[0611] The device sends the stored audio data to the server over the Internet. Specifically, it uploads the audio file to the server using the HTTP protocol. A secure TLS connection is ensured during this process. The input is the local audio file, and the output is the audio data on the server.

[0612] Step 4:

[0613] The server converts the received voice data into text data using a voice recognition engine. Here, for example, the Google Cloud Speech-to-Text API is used. Specifically, the voice data is sent to the API and the returned text data is received. The input is voice data and the output is text data.

[0614] Step 5:

[0615] The server analyzes the converted text data and generates minutes. Specifically, it analyzes the text data from the speech recognition engine, identifies and organizes the statements made by each speaker. For example, it formats the minutes in a format such as "Speaker A: This is an important point" and "Speaker B: I agree." The input is text data, and the output is the minutes.

[0616] Step 6:

[0617] The server analyzes the generated text data of the minutes using natural language processing technology. Specifically, it uses Python's NLTK and spaCy to perform sentiment analysis, keyword extraction, and linguistic style analysis. For example, it extracts each speaker's speaking tendencies and keywords. The input is the minutes, and the output is the analysis results.

[0618] Step 7:

[0619] The server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-4) to create optimal advice from the analysis results. For example, it generates advice such as "We recommend taking a flexible approach." The input is the analysis results, and the output is advice.

[0620] Step 8:

[0621] The server sends the generated advice to the device, which then visually displays it to the user. Specifically, we build a UI that displays the advice on the device's display. The input is the advice, and the output is a visual confirmation for the user.

[0622] (Application example 1)

[0623] 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."

[0624] Conventional systems for automatically generating meeting minutes are designed for use in conference rooms and lack the functionality to record and analyze conversations with passengers while traveling and provide advice. Furthermore, when passengers exchange opinions or hold discussions in real time in an in-car environment, there are insufficient means to properly record and share the content. Furthermore, since there is no effective in-car display of advice for consensus building based on the content of conversations, there is a need for a system that supports efficient decision-making while traveling.

[0625] 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.

[0626] In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, means for collecting in-vehicle voice data, means for transmitting the collected voice data to a cloud server in real time, and means for displaying the minutes in the autonomously driving vehicle. This makes it possible to automatically generate minutes of discussions held by passengers while traveling, and to analyze the opinions and ways of thinking of each passenger and provide advice to promote consensus building in real time.

[0627] The "means for receiving audio input" refers to a device in which a microphone installed in the vehicle captures audio data and transmits it to the system.

[0628] The "means for converting voice data into text data" refers to a device or program that uses voice recognition technology to convert collected voice data into text information in real time.

[0629] The "means for generating minutes based on text data" refers to a device or program that analyzes the converted text data, systematically organizes the contents of the meeting or discussion, and compiles them in the form of minutes.

[0630] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to a device or program that uses natural language processing technology to analyze the contents of minutes and infer the personalities and ways of thinking of each participant based on their remarks and emotions.

[0631] The "means for generating advice to promote consensus building" is a device or program that generates specific proposals and advice for achieving better consensus building based on the results of meeting minutes and participant analysis.

[0632] The "means for displaying the generated advice" is a display device or a program for visually presenting the generated advice to the user.

[0633] "Means for collecting voice data inside the vehicle" refers to a device or system that uses microphones and sensors installed inside the vehicle to record conversations between passengers and surrounding sounds.

[0634] "Means for transmitting collected voice data to a cloud server in real time" refers to a device or program that transfers collected voice data from a device such as an edge computer to a cloud server and processes it in real time.

[0635] "Means for displaying minutes within an autonomous vehicle" refers to a device or program that provides the generated minutes and advice to the occupants through a display or various display devices within the autonomous vehicle.

[0636] The present invention provides a system for recording discussions and conversations among occupants in an autonomous vehicle, generating minutes in real time, and promoting consensus building. Specific embodiments of the system are described below.

[0637] System Configuration

[0638] The system has the following configuration:

[0639] 1. In-car audio collection means:

[0640] A microphone is installed inside the vehicle to collect passengers' comments and surrounding sounds in real time, allowing discussions to be recorded sequentially.

[0641] 2. Edge computers:

[0642] Using an edge computer (e.g., NVIDIA Jetson Nano) installed in the vehicle, the collected voice data is temporarily stored locally and transmitted to a cloud server via the Internet.

[0643] 3. Cloud Server:

[0644] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the received voice data into text data.

[0645] 4. Minutes generation method:

[0646] Based on the converted text data, the system identifies what each speaker said and generates minutes, systematically organizing the conversation and recording it in an easy-to-understand format.

[0647] 5. Natural Language Processing (NLP):

[0648] The text data in the minutes is analyzed using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. Specifically, sentiment analysis, keyword extraction, and analysis of speaking style are performed.

[0649] 6. Advice Generation Methods:

[0650] Based on the analysis results, advice is automatically generated to promote consensus building, for example by providing suggestions to reconcile differing opinions and increase overall satisfaction.

[0651] 7. Display means:

[0652] The generated advice and minutes are visually displayed on a display installed inside the vehicle, allowing passengers to receive advice in real time and reflect it in the progress of the discussion.

[0653] Processing flow

[0654] After receiving the voice data, the server performs the following processing steps.

[0655] 1. Use a speech recognition engine to convert voice data into text data.

[0656] 2. Analyze the converted text data and generate meeting minutes.

[0657] 3. Use natural language processing technology to analyze the contents of the minutes and infer the personalities and thoughts of the participants.

[0658] 4. Based on the analysis results, advice is generated to promote consensus building.

[0659] 5. Display the generated advice on the in-car display.

[0660] Hardware and software used

[0661] Microphone: A device used to collect speech from inside the vehicle.

[0662] Edge computer: A device that temporarily stores voice data and transmits it to a cloud server (e.g., NVIDIA Jetson Nano).

[0663] Cloud server: Processes and analyzes audio data.

[0664] Speech recognition engine: Software that converts voice data into text data (e.g., Google Cloud Speech-to-Text).

[0665] Natural language processing technology: Software that analyzes text data (e.g., NLTK, Google Cloud Natural Language).

[0666] Display: An in-vehicle display device for showing advice and minutes.

[0667] Specific examples

[0668] For example, if a family is driving and discussing their next destination, the system can analyze the following prompts:

[0669] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0670] Based on this, the system generates the minutes as follows:

[0671] Father: Next time I want to go to the beach.

[0672] Mother: I think the mountains are better.

[0673] Child A: I want to go to the zoo.

[0674] Child B: An amusement park would be nice.

[0675] The analysis then generates advice that respects each individual's opinion and promotes consensus building:

[0676] "Consider going somewhere that's close to both the beach and the mountains, or plan another destination for next weekend."

[0677] In this way, the system can generate meeting minutes and provide advice in real time, supporting efficient and smooth decision-making while on the move.

[0678] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0679] Step 1:

[0680] The device collects voice data in real time from microphones installed in various locations inside the vehicle. The voice data input is the conversations of passengers and the surrounding environmental sounds, and this data is temporarily stored in digital format on the edge computer. This allows the content of the conversation to be properly recorded.

[0681] Step 2:

[0682] The device transmits the temporarily stored voice data to a cloud server via the Internet. The input data is a digitized voice file, and the output is the arrival of the voice file at the cloud server. This transmission process utilizes a stable network connection and an appropriate data transfer protocol.

[0683] Step 3:

[0684] The server receives the audio file sent to the cloud server and converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input is the audio file, and the output is the corresponding text data. The speech recognition engine analyzes the audio waveform and performs a process to identify words.

[0685] Step 4:

[0686] The server automatically generates minutes based on the converted text data. It analyzes the text data, identifies the content of each speaker's speech, and creates minutes by formatting them into a specific format (e.g., "Speaker A: ○○"). The input is the speech-recognized text data, and the output is the text of the minutes. This operation uses a text analysis algorithm to organize the order and content of the speech.

[0687] Step 5:

[0688] The server analyzes the automatically generated text data of the minutes using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. The input is the text data of the minutes, and the output is an inference about the personality and way of thinking of each participant. This process uses multiple NLP techniques, including sentiment analysis and keyword extraction.

[0689] Step 6:

[0690] The server generates advice to promote consensus building based on the inference results. The input is the inference results about each participant's personality and way of thinking, and the output is specific advice text. This operation uses a generative AI model to create the most appropriate advice at that time.

[0691] Step 7:

[0692] The terminal displays the generated advice received from the server on the in-car display. The input is text data of the advice, and the output is displayed on the display. This display function allows passengers to directly check the advice and use its content for discussion and consensus building.

[0693] Prompt Sentence Examples

[0694] For example, if a family is driving and discussing their next destination, they might use prompts like the following as input:

[0695] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0696] 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.

[0697] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0698] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0699] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0700] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0701] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0702] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0703] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0704] In this way, the present invention automatically generates meeting minutes based on the audio data of a meeting and analyzes the personalities, thoughts, and emotions of the participants to support efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0705] The above is a specific embodiment of the present invention.

[0706] The processing flow will be explained below.

[0707] Step 1:

[0708] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time, stores the audio data locally, and transmits the received audio data to a server via the Internet.

[0709] Step 2:

[0710] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). During this conversion process, the content of each speaker's speech is identified and the converted text data is classified for each speech.

[0711] Step 3:

[0712] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile approach here" and "Speaker B: I agree, but we should also consider a waterfall approach in some phases."

[0713] Step 4:

[0714] The server analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. This allows the server to infer the speaker's personality and way of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0715] Step 5:

[0716] The server uses an emotion engine to recognize the user's emotions from the voice and text data. The emotion engine analyzes the emotional tone of each utterance and labels it with an emotion, such as "excited," "calm," or "doubt."

[0717] Step 6:

[0718] The server integrates the emotion data recognized by the emotion engine with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach here," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0719] Step 7:

[0720] Based on the analysis results and emotional data, the server generates specific advice to promote consensus building. This advice becomes more appropriate by taking emotional states into account. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while appropriately monitoring emotions at each phase."

[0721] Step 8:

[0722] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice when proceeding with the meeting or making decisions.

[0723] Through these steps, the system performs a series of processes, from voice input to generating minutes, analyzing emotions, and providing advice on consensus building.

[0724] Example 2

[0725] 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."

[0726] There is a need for a system that can automatically generate minutes based on audio data from meetings and discussions, analyze the emotions as well as the personalities and ways of thinking of participants, and support efficient and harmonious consensus building. However, current systems do not adequately process audio data in real time or analyze the emotions of participants, which leaves issues with the quality and speed of consensus building.

[0727] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for temporarily storing voice data and transmitting it via the Internet, a means for converting the received voice data into text data, and a means for generating minutes based on the identified and classified text data. This enables real-time voice data processing, highly accurate text conversion, and analysis including the emotions and thoughts of participants.

[0728] "Audio data" refers to data in which audio information from meetings, discussions, etc. is recorded in digital format.

[0729] "Real time" means that processing or operations are carried out at the same speed as real time.

[0730] "Receiving" means that a system or device acquires information from the outside.

[0731] "Text data" refers to data obtained by converting voice data into character information.

[0732] "Identification and classification" refers to recognizing data or information that has specific characteristics from multiple sources and organizing it into categories.

[0733] "Minutes" are documents that record the contents of meetings and discussions.

[0734] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate natural human language.

[0735] "Emotion data" is data that includes emotional information extracted from speech or text.

[0736] "Analysis" means the detailed analysis of data or information.

[0737] "Advice" is a suggestion or guidance offered in response to a specific problem or situation.

[0738] "Display" means visually showing information to a user.

[0739] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0740] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0741] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0742] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0743] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0744] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0745] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0746] For example, when discussing how to proceed with a project, it can provide an effective approach that suits the emotional state of the person speaking and increase the sense of acceptance of the entire team. This increases satisfaction with the decisions made and strengthens team cooperation.

[0747] Examples of specific prompts are:

[0748] "Using audio data from a meeting, analyze the personalities and emotions of participants and generate advice to promote consensus building."

[0749] With these prompts, the generative AI model performs the necessary speech recognition, natural language processing, sentiment analysis, and data classification to build a program that provides appropriate advice.

[0750] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0751] Step 1:

[0752] Users use microphones to collect audio data from meetings and send audio input to the system. The input is real-time audio data from meetings, and the output is audio data sent to the terminal. The specific operation is that the microphone captures audio waveforms and converts them into digital signals.

[0753] Step 2:

[0754] The device processes the received voice data in real time, temporarily stores it in local storage, and simultaneously transmits it to a server via the Internet. The input is voice data from the user, and the output is voice data transmitted to the server via the Internet. Specifically, the device transmits the voice data to the server using data streaming technology.

[0755] Step 3:

[0756] The server processes the received voice data and converts it into text data using a voice recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The input is the received voice data, and the output is text data. Specifically, the voice recognition model analyzes the sound waves of the voice data and generates the corresponding text.

[0757] Step 4:

[0758] The server analyzes the generated text data and identifies and classifies each speaker. The input is text data generated by a speech recognition engine, and the output is classified text in the form of a meeting minutes. Specifically, the text classification algorithm analyzes keywords and context to identify the speaker.

[0759] Step 5:

[0760] The server analyzes the text data of the minutes using natural language processing technology. The input is the text data of the minutes, and the output is information including the analyzed participants' personalities and thoughts. Specific operations include sentiment analysis, keyword extraction, and language style analysis using natural language processing technology.

[0761] Step 6:

[0762] The server uses an emotion engine to recognize emotions from the voice and text data and integrate them with the text data of the minutes. The input is the analyzed text data and voice data, and the output is the minutes with added emotional context. Specifically, the emotion analysis algorithm analyzes the tone of the voice and the wording of the text.

[0763] Step 7:

[0764] The server generates specific advice to promote consensus building based on the analysis results and emotional data. The input is the integrated minutes and analytical data, and the output is advice for consensus building. Specifically, the advice generation algorithm takes into account various factors (emotional state, meeting content, participants' personalities, etc.) to construct optimal advice.

[0765] Step 8:

[0766] The server sends the generated advice to the terminal, which then visually displays it to the user. The input is the generated advice, and the output is the visually displayed advice. The specific operation is for the terminal to display the received advice to the user through a GUI.

[0767] By using the above steps, the user can achieve a series of processes from collecting voice data to integrating emotion data and finally displaying advice.

[0768] (Application example 2)

[0769] 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."

[0770] Meetings and discussions in industrial production sites are important for increasing efficiency, but preparing meeting minutes and reaching consensus while taking into account participants' emotions takes time and effort. In particular, in large-scale projects, accurately reflecting the opinions of all participants and communicating their emotions is crucial for rapid and effective decision-making. However, with existing technologies, automatic conversion of voice data alone is insufficient, making it difficult to support effective consensus-building that takes emotions into account. Therefore, the present invention aims to solve these problems and realize efficient and harmonious meeting management in industrial production sites.

[0771] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, and means for automatically generating meeting minutes at an industrial production site and supporting consensus building by analyzing and taking into account the emotions of the participants. This automates the creation of meeting minutes and enables support for consensus building that takes into account the emotions of the participants.

[0772] The "means for receiving voice input" refers to a device or system for collecting voice information from a meeting or discussion as digital data.

[0773] "Means for converting voice data into text data" refers to technology or devices that convert collected voice information into text strings.

[0774] The "means for generating minutes based on text data" is a system that organizes converted text information and automatically generates documents that describe the contents of meetings and discussions.

[0775] "Means of analyzing text data in minutes and inferring the personalities and ways of thinking of participants" is a function that analyzes the character strings in the generated minutes and infers the personalities and thought patterns of the speakers.

[0776] The "means for generating advice to promote consensus building" is a technology that automatically creates specific proposals based on analyzed data to facilitate smooth consensus building among participants.

[0777] The "means for displaying the generated advice" refers to a device or screen for visually presenting the generated advice to the user.

[0778] "Means for automatically generating meeting minutes in industrial production sites and supporting consensus building by analyzing and considering the emotions of participants" is a system that automatically records the contents of meetings held at production sites and effectively supports consensus building by taking into consideration the emotional state of participants.

[0779] The present invention provides a system for efficiently and effectively managing meetings and discussions at industrial production sites and supporting consensus building among participants. Specific embodiments of the system are described below.

[0780] First, a user uses a microphone to collect the audio of the meeting and transmits the audio input to the system. This audio input is received in real time by a terminal at the industrial production site. The received audio data is stored locally and simultaneously transmitted to a server via the Internet.

[0781] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and is classified as text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0782] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning." This is done using natural language processing libraries such as NLTK and spaCy.

[0783] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes user emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. In this process, emotion analysis models such as TextBlob and Hugging Face are used. This generates minutes with added emotional context for the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project." "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0784] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice is made more appropriate by taking into account the emotional state. It uses OpenAI's generative AI model to generate advice by inputting prompt sentences such as the following:

[0785] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[0786] The generated advice is sent from the server to the device, which then visually displays it to the user. The user can refer to this advice when conducting meetings or making decisions. For example, the advice may suggest adopting an agile methodology in the early stages of a project, and considering combining it with a waterfall methodology as the project progresses, while monitoring the emotions at each phase.

[0787] In this way, the present invention automatically generates meeting minutes based on the audio data of meetings in industrial production sites, and supports efficient and harmonious consensus building by analyzing the personalities, ways of thinking, and emotions of the participants. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0788] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0789] Step 1:

[0790] The user collects the audio of the conference using a microphone and transmits the audio data to the system. The input is the audio data of the conference, and the output is the audio data received by the audio input device. In this step, the audio is input to the terminal in real time.

[0791] Step 2:

[0792] The device stores the received voice data locally and simultaneously transmits it to a server via the Internet. The input is the voice data received in real time, and the output is the voice data transmitted to the server. This allows the voice data to be stored and transmitted.

[0793] Step 3:

[0794] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is the voice data stored on the server, and the output is text data. The converted text data is classified by each speaker.

[0795] Step 4:

[0796] The server automatically generates minutes based on the text data. The input is classified text data, and the output is organized minutes data. In this step, each statement is converted into minutes format.

[0797] Step 5:

[0798] The server analyzes the generated text data of the minutes using natural language processing (NLP). The input is the text data of the minutes, and the output is analytical data on the participants' personalities, thoughts, emotions, etc. Specifically, NLTK and spaCy are used to extract keywords and perform sentiment analysis on the text data.

[0799] Step 6:

[0800] The server uses an emotion engine to recognize user emotions from voice and text data, and integrates the recognized emotion data with the text data of the minutes. The input is voice and text data, and the output is the minutes data with added emotional context. It uses emotion analysis models such as TextBlob and Hugging Face.

[0801] Step 7:

[0802] The server generates specific advice to promote consensus building based on the analysis results and emotion data. The input is the analyzed minutes data and emotion data, and the output is advice data to support consensus building. Using OpenAI's generative AI model, the server inputs the following prompt:

[0803] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[0804] Step 8:

[0805] The generated advice is sent from the server to the terminal, and the terminal visually displays this advice to the user. The input is the generated advice data, and the output is a visual display of the advice to the user. This allows the user to refer to the advice and proceed with the meeting or make decisions.

[0806] 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.

[0807] 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.

[0808] 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.

[0809] [Fourth embodiment]

[0810] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0811] 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.

[0812] 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).

[0813] 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.

[0814] 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.

[0815] 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).

[0816] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0817] 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.

[0818] 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.

[0819] 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.

[0820] 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.

[0821] 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.

[0822] 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."

[0823] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0824] First, the user applies a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives the audio input in real time and saves it locally as an audio file. The device also transmits the received audio data to a server via the Internet.

[0825] The server processes the received voice data and converts it into text using a speech recognition engine. At this time, it identifies the content of each speaker's speech, formats the converted text, and generates meeting minutes. For example, minutes could be created in the format "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0826] The server then analyzes the text data of the generated minutes using natural language processing technology. This analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of the participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0827] The server then generates advice based on the analysis results to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0828] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0829] This system automatically generates meeting minutes based on audio data from meetings and analyzes the personalities and ways of thinking of the participants, enabling efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, the system can present effective methods for consolidating differing opinions, increasing the sense of satisfaction of the entire team.

[0830] The above is a specific embodiment of the present invention.

[0831] The processing flow will be explained below.

[0832] Step 1:

[0833] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0834] Step 2:

[0835] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). At this time, it identifies the content of each speaker's speech and classifies the converted text data.

[0836] Step 3:

[0837] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile methodology for project progress" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0838] Step 4:

[0839] The server analyzes the generated minutes' text data using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0840] Step 5:

[0841] Based on the analysis results, the server generates specific advice to promote consensus building, such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0842] Step 6:

[0843] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice to guide the progress of the meeting and make decisions.

[0844] In this way, the system performs a series of processes from voice input to generating and analyzing minutes and providing advice on consensus building.

[0845] Example 1

[0846] 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."

[0847] Accurately and efficiently recording the contents of meetings and discussions and automatically generating meeting minutes based on that information is important, especially for large organizations and projects. However, current systems do not adequately process audio data in real time, analyze participants' personalities and ways of thinking, or generate advice to promote consensus building. This results in time-consuming meetings and decision-making, reducing productivity.

[0848] 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.

[0849] In this invention, the server includes means for locally saving voice data, means for transmitting the saved voice data to the server, means for converting the voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, and means for displaying the generated advice. This makes it possible to convert the contents of the meeting into text in real time and provide analysis and advice based on the contents.

[0850] A "means for receiving audio input" is a device or method for collecting and transmitting audio from a meeting or discussion to the system.

[0851] The "means for locally storing voice data" refers to a device or method for temporarily recording collected voice data in a storage device of the terminal.

[0852] "Means for transmitting stored audio data to a server" refers to a device or method for uploading locally stored audio data to a server via the Internet.

[0853] "Means for converting voice data into text data" refers to a device or method for converting voice data into text data using voice recognition technology.

[0854] The "means for generating minutes from text data" refers to a device or method for organizing and formatting the converted text data and compiling it into minutes of the meeting.

[0855] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to devices and methods that use natural language processing technology to analyze and infer the personalities and thought patterns of speakers from the contents of minutes.

[0856] The "means for generating advice to promote consensus building" refers to a device or method for generating specific advice to help reach consensus in a meeting based on the analysis results.

[0857] The "means for displaying the generated advice" refers to a device or method for visually presenting the generated advice to the user.

[0858] The present invention relates to a system that automatically generates minutes based on voice data obtained from meetings and discussions, analyzes the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0859] First, the user sends audio input to the system using a microphone to collect the conference audio. The device receives this audio input in real time and saves it locally as an audio file. It is desirable for the device to have a high-quality microphone and audio recording software built in. For example, a microphone built into a PC or tablet and a dedicated audio recording application can be used.

[0860] The device then transmits the saved audio data over the Internet to a server, preferably using the HTTP protocol and a secure TLS connection, where it is temporarily stored in cloud storage and then transferred to a server for analysis.

[0861] The server converts the received voice data into text data using a speech recognition service. Specifically, it uses a speech recognition engine such as Google Cloud Speech-to-Text API. This engine analyzes the voice data and generates corresponding text data. For example, minutes are created in the format "Speaker A: We should adopt an agile methodology for progressing the project" and "Speaker B: I agree, but we should also consider a waterfall methodology for some phases."

[0862] The server then analyzes the text data in the minutes using natural language processing techniques. This includes sentiment analysis, keyword extraction, and linguistic style analysis, and is performed using Python libraries such as NLTK and spaCy. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0863] Furthermore, the server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-4) to generate advice such as "Use agile methods in the early stages of the project, and consider combining them with waterfall methods as the project progresses."

[0864] The generated advice is sent from the server to the terminal, which then visually displays it to the user, who can use it to guide the progress of the meeting and make decisions.

[0865] As a concrete example, consider a situation where a user is discussing how to proceed with a project. If a user says, "We should adopt an agile methodology," during a meeting, the system recognizes the statement and records it in the minutes as "Speaker A: We should adopt an agile methodology." The system then analyzes the entire statement and determines that "Speaker A tends to value flexibility," generating advice such as, "We recommend adopting an agile methodology in the early stages of the project, and combining it with a waterfall methodology as the project progresses." These processes are implemented using Python scripts, speech recognition services, natural language processing libraries, and generative AI models.

[0866] Example prompt sentence:

[0867] 1. Please explain in detail how you convert meeting audio data into text and how you perform analytics based on that data.

[0868] 2. "Please explain in detail how you use a generative AI model to generate advice based on analysis of meeting minutes."

[0869] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0870] Step 1:

[0871] The user sends audio input to the system using a microphone to collect the conference audio. Specifically, when the user presses the record button at the start of the conference, the microphone starts recording the audio. The input is the conference audio, and the output is audio data.

[0872] Step 2:

[0873] The device receives the audio and stores it locally as an audio file in real time. Specifically, the audio recording software receives the audio stream and stores it as a .wav file. The input is an audio stream, and the output is a local audio file.

[0874] Step 3:

[0875] The device sends the stored audio data to the server over the Internet. Specifically, it uploads the audio file to the server using the HTTP protocol. A secure TLS connection is ensured during this process. The input is the local audio file, and the output is the audio data on the server.

[0876] Step 4:

[0877] The server converts the received voice data into text data using a voice recognition engine. Here, for example, the Google Cloud Speech-to-Text API is used. Specifically, the voice data is sent to the API and the returned text data is received. The input is voice data and the output is text data.

[0878] Step 5:

[0879] The server analyzes the converted text data and generates minutes. Specifically, it analyzes the text data from the speech recognition engine, identifies and organizes the statements made by each speaker. For example, it formats the minutes in a format such as "Speaker A: This is an important point" and "Speaker B: I agree." The input is text data, and the output is the minutes.

[0880] Step 6:

[0881] The server analyzes the generated text data of the minutes using natural language processing technology. Specifically, it uses Python's NLTK and spaCy to perform sentiment analysis, keyword extraction, and linguistic style analysis. For example, it extracts each speaker's speaking tendencies and keywords. The input is the minutes, and the output is the analysis results.

[0882] Step 7:

[0883] The server generates advice to promote consensus building based on the analysis results. Specifically, it uses a generative AI model (e.g., GPT-4) to create optimal advice from the analysis results. For example, it generates advice such as "We recommend taking a flexible approach." The input is the analysis results, and the output is advice.

[0884] Step 8:

[0885] The server sends the generated advice to the device, which then visually displays it to the user. Specifically, we build a UI that displays the advice on the device's display. The input is the advice, and the output is a visual confirmation for the user.

[0886] (Application example 1)

[0887] 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."

[0888] Conventional systems for automatically generating meeting minutes are designed for use in conference rooms and lack the functionality to record and analyze conversations with passengers while traveling and provide advice. Furthermore, when passengers exchange opinions or hold discussions in real time in an in-car environment, there are insufficient means to properly record and share the content. Furthermore, since there is no effective in-car display of advice for consensus building based on the content of conversations, there is a need for a system that supports efficient decision-making while traveling.

[0889] 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.

[0890] In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, means for collecting in-vehicle voice data, means for transmitting the collected voice data to a cloud server in real time, and means for displaying the minutes in the autonomously driving vehicle. This makes it possible to automatically generate minutes of discussions held by passengers while traveling, and to analyze the opinions and ways of thinking of each passenger and provide advice to promote consensus building in real time.

[0891] The "means for receiving audio input" refers to a device in which a microphone installed in the vehicle captures audio data and transmits it to the system.

[0892] The "means for converting voice data into text data" refers to a device or program that uses voice recognition technology to convert collected voice data into text information in real time.

[0893] The "means for generating minutes based on text data" refers to a device or program that analyzes the converted text data, systematically organizes the contents of the meeting or discussion, and compiles them in the form of minutes.

[0894] "Means for analyzing text data in minutes and inferring the personalities and ways of thinking of participants" refers to a device or program that uses natural language processing technology to analyze the contents of minutes and infer the personalities and ways of thinking of each participant based on their remarks and emotions.

[0895] The "means for generating advice to promote consensus building" is a device or program that generates specific proposals and advice for achieving better consensus building based on the results of meeting minutes and participant analysis.

[0896] The "means for displaying the generated advice" is a display device or a program for visually presenting the generated advice to the user.

[0897] "Means for collecting voice data inside the vehicle" refers to a device or system that uses microphones and sensors installed inside the vehicle to record conversations between passengers and surrounding sounds.

[0898] "Means for transmitting collected voice data to a cloud server in real time" refers to a device or program that transfers collected voice data from a device such as an edge computer to a cloud server and processes it in real time.

[0899] "Means for displaying minutes within an autonomous vehicle" refers to a device or program that provides the generated minutes and advice to the occupants through a display or various display devices within the autonomous vehicle.

[0900] The present invention provides a system for recording discussions and conversations among occupants in an autonomous vehicle, generating minutes in real time, and promoting consensus building. Specific embodiments of the system are described below.

[0901] System Configuration

[0902] The system has the following configuration:

[0903] 1. In-car audio collection means:

[0904] A microphone is installed inside the vehicle to collect passengers' comments and surrounding sounds in real time, allowing discussions to be recorded sequentially.

[0905] 2. Edge computers:

[0906] Using an edge computer (e.g., NVIDIA Jetson Nano) installed in the vehicle, the collected voice data is temporarily stored locally and transmitted to a cloud server via the Internet.

[0907] 3. Cloud Server:

[0908] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the received voice data into text data.

[0909] 4. Minutes generation method:

[0910] Based on the converted text data, the system identifies what each speaker said and generates minutes, systematically organizing the conversation and recording it in an easy-to-understand format.

[0911] 5. Natural Language Processing (NLP):

[0912] The text data in the minutes is analyzed using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. Specifically, sentiment analysis, keyword extraction, and analysis of speaking style are performed.

[0913] 6. Advice Generation Methods:

[0914] Based on the analysis results, advice is automatically generated to promote consensus building, for example by providing suggestions to reconcile differing opinions and increase overall satisfaction.

[0915] 7. Display means:

[0916] The generated advice and minutes are visually displayed on a display installed inside the vehicle, allowing passengers to receive advice in real time and reflect it in the progress of the discussion.

[0917] Processing flow

[0918] After receiving the voice data, the server performs the following processing steps.

[0919] 1. Use a speech recognition engine to convert voice data into text data.

[0920] 2. Analyze the converted text data and generate meeting minutes.

[0921] 3. Use natural language processing technology to analyze the contents of the minutes and infer the personalities and thoughts of the participants.

[0922] 4. Based on the analysis results, advice is generated to promote consensus building.

[0923] 5. Display the generated advice on the in-car display.

[0924] Hardware and software used

[0925] Microphone: A device used to collect speech from inside the vehicle.

[0926] Edge computer: A device that temporarily stores voice data and transmits it to a cloud server (e.g., NVIDIA Jetson Nano).

[0927] Cloud server: Processes and analyzes audio data.

[0928] Speech recognition engine: Software that converts voice data into text data (e.g., Google Cloud Speech-to-Text).

[0929] Natural language processing technology: Software that analyzes text data (e.g., NLTK, Google Cloud Natural Language).

[0930] Display: An in-vehicle display device for showing advice and minutes.

[0931] Specific examples

[0932] For example, if a family is driving and discussing their next destination, the system can analyze the following prompts:

[0933] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0934] Based on this, the system generates the minutes as follows:

[0935] Father: Next time I want to go to the beach.

[0936] Mother: I think the mountains are better.

[0937] Child A: I want to go to the zoo.

[0938] Child B: An amusement park would be nice.

[0939] The analysis then generates advice that respects each individual's opinion and promotes consensus building:

[0940] "Consider going somewhere that's close to both the beach and the mountains, or plan another destination for next weekend."

[0941] In this way, the system can generate meeting minutes and provide advice in real time, supporting efficient and smooth decision-making while on the move.

[0942] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0943] Step 1:

[0944] The device collects voice data in real time from microphones installed in various locations inside the vehicle. The voice data input is the conversations of passengers and the surrounding environmental sounds, and this data is temporarily stored in digital format on the edge computer. This allows the content of the conversation to be properly recorded.

[0945] Step 2:

[0946] The device transmits the temporarily stored voice data to a cloud server via the Internet. The input data is a digitized voice file, and the output is the arrival of the voice file at the cloud server. This transmission process utilizes a stable network connection and an appropriate data transfer protocol.

[0947] Step 3:

[0948] The server receives the audio file sent to the cloud server and converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The input is the audio file, and the output is the corresponding text data. The speech recognition engine analyzes the audio waveform and performs a process to identify words.

[0949] Step 4:

[0950] The server automatically generates minutes based on the converted text data. It analyzes the text data, identifies the content of each speaker's speech, and creates minutes by formatting them into a specific format (e.g., "Speaker A: ○○"). The input is the speech-recognized text data, and the output is the text of the minutes. This operation uses a text analysis algorithm to organize the order and content of the speech.

[0951] Step 5:

[0952] The server analyzes the automatically generated text data of the minutes using natural language processing technology (e.g., NLTK, Google Cloud Natural Language) to infer the personality and way of thinking of each participant. The input is the text data of the minutes, and the output is an inference about the personality and way of thinking of each participant. This process uses multiple NLP techniques, including sentiment analysis and keyword extraction.

[0953] Step 6:

[0954] The server generates advice to promote consensus building based on the inference results. The input is the inference results about each participant's personality and way of thinking, and the output is specific advice text. This operation uses a generative AI model to create the most appropriate advice at that time.

[0955] Step 7:

[0956] The terminal displays the generated advice received from the server on the in-car display. The input is text data of the advice, and the output is displayed on the display. This display function allows passengers to directly check the advice and use its content for discussion and consensus building.

[0957] Prompt Sentence Examples

[0958] For example, if a family is driving and discussing their next destination, they might use prompts like the following as input:

[0959] A family is discussing where to go for their next drive. The father says, "I want to go to the beach next time," while the mother replies, "The mountains would be better." The children also have their own opinions.

[0960] 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.

[0961] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[0962] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[0963] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[0964] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0965] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0966] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[0967] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[0968] In this way, the present invention automatically generates meeting minutes based on the audio data of a meeting and analyzes the personalities, thoughts, and emotions of the participants to support efficient and harmonious consensus building. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[0969] The above is a specific embodiment of the present invention.

[0970] The processing flow will be explained below.

[0971] Step 1:

[0972] Users use a microphone to input audio during meetings and discussions. The device receives this audio input in real time, stores the audio data locally, and transmits the received audio data to a server via the Internet.

[0973] Step 2:

[0974] The server receives the voice data sent from the device and converts it into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text API). During this conversion process, the content of each speaker's speech is identified and the converted text data is classified for each speech.

[0975] Step 3:

[0976] The server generates meeting minutes organized by each statement based on the text data generated by speech recognition. For example, minutes might be created in a format such as "Speaker A: We should adopt an agile approach here" and "Speaker B: I agree, but we should also consider a waterfall approach in some phases."

[0977] Step 4:

[0978] The server analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. This allows the server to infer the speaker's personality and way of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[0979] Step 5:

[0980] The server uses an emotion engine to recognize the user's emotions from the voice and text data. The emotion engine analyzes the emotional tone of each utterance and labels it with an emotion, such as "excited," "calm," or "doubt."

[0981] Step 6:

[0982] The server integrates the emotion data recognized by the emotion engine with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach here," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[0983] Step 7:

[0984] Based on the analysis results and emotional data, the server generates specific advice to promote consensus building. This advice becomes more appropriate by taking emotional states into account. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while appropriately monitoring emotions at each phase."

[0985] Step 8:

[0986] The server sends the generated advice to the terminal, which then displays it to the user. The user can refer to this advice when proceeding with the meeting or making decisions.

[0987] Through these steps, the system performs a series of processes, from voice input to generating minutes, analyzing emotions, and providing advice on consensus building.

[0988] Example 2

[0989] 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."

[0990] There is a need for a system that can automatically generate minutes based on audio data from meetings and discussions, analyze the emotions as well as the personalities and ways of thinking of participants, and support efficient and harmonious consensus building. However, current systems do not adequately process audio data in real time or analyze the emotions of participants, which leaves issues with the quality and speed of consensus building.

[0991] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for temporarily storing voice data and transmitting it via the Internet, a means for converting the received voice data into text data, and a means for generating minutes based on the identified and classified text data. This enables real-time voice data processing, highly accurate text conversion, and analysis including the emotions and thoughts of participants.

[0992] "Audio data" refers to data in which audio information from meetings, discussions, etc. is recorded in digital format.

[0993] "Real time" means that processing or operations are carried out at the same speed as real time.

[0994] "Receiving" means that a system or device acquires information from the outside.

[0995] "Text data" refers to data obtained by converting voice data into character information.

[0996] "Identification and classification" refers to recognizing data or information that has specific characteristics from multiple sources and organizing it into categories.

[0997] "Minutes" are documents that record the contents of meetings and discussions.

[0998] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate natural human language.

[0999] "Emotion data" is data that includes emotional information extracted from speech or text.

[1000] "Analysis" means the detailed analysis of data or information.

[1001] "Advice" is a suggestion or guidance offered in response to a specific problem or situation.

[1002] "Display" means visually showing information to a user.

[1003] The present invention relates to a system that automatically generates minutes based on audio data from meetings and discussions, analyzes the emotions as well as the personalities and ways of thinking of participants, and supports consensus building. Specific embodiments of the system are described below.

[1004] First, the user uses a microphone to collect the audio of the meeting and sends the audio input to the system. The device receives this audio input in real time and stores it locally, while simultaneously transmitting the audio data to a server via the Internet.

[1005] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and classifies the text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[1006] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the personalities and ways of thinking of participants, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning."

[1007] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes the user's emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. This generates minutes that include the emotional context of the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project," or "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[1008] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice becomes more appropriate by taking into account the emotional state. For example, it might suggest, "It is advisable to adopt an agile methodology in the early stages of the project, and consider combining it with a waterfall methodology as the project progresses, while monitoring emotions at each phase."

[1009] The generated advice is sent from the server to the terminal, which then visually displays it to the user, allowing the user to proceed with the meeting and make decisions while referring to the advice.

[1010] For example, when discussing how to proceed with a project, it can provide an effective approach that suits the emotional state of the person speaking and increase the sense of acceptance of the entire team. This increases satisfaction with the decisions made and strengthens team cooperation.

[1011] Examples of specific prompts are:

[1012] "Using audio data from a meeting, analyze the personalities and emotions of participants and generate advice to promote consensus building."

[1013] With these prompts, the generative AI model performs the necessary speech recognition, natural language processing, sentiment analysis, and data classification to build a program that provides appropriate advice.

[1014] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1015] Step 1:

[1016] Users use microphones to collect audio data from meetings and send audio input to the system. The input is real-time audio data from meetings, and the output is audio data sent to the terminal. The specific operation is that the microphone captures audio waveforms and converts them into digital signals.

[1017] Step 2:

[1018] The device processes the received voice data in real time, temporarily stores it in local storage, and simultaneously transmits it to a server via the Internet. The input is voice data from the user, and the output is voice data transmitted to the server via the Internet. Specifically, the device transmits the voice data to the server using data streaming technology.

[1019] Step 3:

[1020] The server processes the received voice data and converts it into text data using a voice recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The input is the received voice data, and the output is text data. Specifically, the voice recognition model analyzes the sound waves of the voice data and generates the corresponding text.

[1021] Step 4:

[1022] The server analyzes the generated text data and identifies and classifies each speaker. The input is text data generated by a speech recognition engine, and the output is classified text in the form of a meeting minutes. Specifically, the text classification algorithm analyzes keywords and context to identify the speaker.

[1023] Step 5:

[1024] The server analyzes the text data of the minutes using natural language processing technology. The input is the text data of the minutes, and the output is information including the analyzed participants' personalities and thoughts. Specific operations include sentiment analysis, keyword extraction, and language style analysis using natural language processing technology.

[1025] Step 6:

[1026] The server uses an emotion engine to recognize emotions from the voice and text data and integrate them with the text data of the minutes. The input is the analyzed text data and voice data, and the output is the minutes with added emotional context. Specifically, the emotion analysis algorithm analyzes the tone of the voice and the wording of the text.

[1027] Step 7:

[1028] The server generates specific advice to promote consensus building based on the analysis results and emotional data. The input is the integrated minutes and analytical data, and the output is advice for consensus building. Specifically, the advice generation algorithm takes into account various factors (emotional state, meeting content, participants' personalities, etc.) to construct optimal advice.

[1029] Step 8:

[1030] The server sends the generated advice to the terminal, which then visually displays it to the user. The input is the generated advice, and the output is the visually displayed advice. The specific operation is for the terminal to display the received advice to the user through a GUI.

[1031] By using the above steps, the user can achieve a series of processes from collecting voice data to integrating emotion data and finally displaying advice.

[1032] (Application example 2)

[1033] 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."

[1034] Meetings and discussions in industrial production sites are important for increasing efficiency, but preparing meeting minutes and reaching consensus while taking into account participants' emotions takes time and effort. In particular, in large-scale projects, accurately reflecting the opinions of all participants and communicating their emotions is crucial for rapid and effective decision-making. However, with existing technologies, automatic conversion of voice data alone is insufficient, making it difficult to support effective consensus-building that takes emotions into account. Therefore, the present invention aims to solve these problems and realize efficient and harmonious meeting management in industrial production sites.

[1035] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving voice input, means for converting voice data into text data, means for generating minutes based on the text data, means for analyzing the text data of the minutes and inferring the personalities and ways of thinking of the participants, means for generating advice to promote consensus building, means for displaying the generated advice, and means for automatically generating meeting minutes at an industrial production site and supporting consensus building by analyzing and taking into account the emotions of the participants. This automates the creation of meeting minutes and enables support for consensus building that takes into account the emotions of the participants.

[1036] The "means for receiving voice input" refers to a device or system for collecting voice information from a meeting or discussion as digital data.

[1037] "Means for converting voice data into text data" refers to technology or devices that convert collected voice information into text strings.

[1038] The "means for generating minutes based on text data" is a system that organizes converted text information and automatically generates documents that describe the contents of meetings and discussions.

[1039] "Means of analyzing text data in minutes and inferring the personalities and ways of thinking of participants" is a function that analyzes the character strings in the generated minutes and infers the personalities and thought patterns of the speakers.

[1040] The "means for generating advice to promote consensus building" is a technology that automatically creates specific proposals based on analyzed data to facilitate smooth consensus building among participants.

[1041] The "means for displaying the generated advice" refers to a device or screen for visually presenting the generated advice to the user.

[1042] "Means for automatically generating meeting minutes in industrial production sites and supporting consensus building by analyzing and considering the emotions of participants" is a system that automatically records the contents of meetings held at production sites and effectively supports consensus building by taking into consideration the emotional state of participants.

[1043] The present invention provides a system for efficiently and effectively managing meetings and discussions at industrial production sites and supporting consensus building among participants. Specific embodiments of the system are described below.

[1044] First, a user uses a microphone to collect the audio of the meeting and transmits the audio input to the system. This audio input is received in real time by a terminal at the industrial production site. The received audio data is stored locally and simultaneously transmitted to a server via the Internet.

[1045] The server processes the received voice data and converts it into text using a speech recognition engine. For example, Google Cloud Speech-to-Text API is used for this voice recognition. The converted text data identifies the content of each speaker's speech and is classified as text data. This generates minutes of the meeting. For example, minutes of the meeting may be created in a format such as "Speaker A: We should adopt an agile methodology to progress the project" and "Speaker B: I agree, but we should also consider the waterfall methodology for some phases."

[1046] The server then analyzes the text data of the generated minutes using natural language processing (NLP) technology. The analysis includes sentiment analysis, keyword extraction, and linguistic style analysis. For example, it can infer the participants' personalities and ways of thinking, such as "Speaker A tends to value flexibility" or "Speaker B tends to value planning." This is done using natural language processing libraries such as NLTK and spaCy.

[1047] Furthermore, this invention incorporates an emotion engine. The emotion engine recognizes user emotions from voice data and text data. The emotion data recognized by the emotion engine is integrated with the text data of the minutes. In this process, emotion analysis models such as TextBlob and Hugging Face are used. This generates minutes with added emotional context for the comments. For example, the format might be something like, "Speaker A (excited): We should adopt an agile approach to progress the project." "Speaker B (calm): I agree, but we should also consider a waterfall approach in some phases."

[1048] The server then generates specific advice to promote consensus building based on the analysis results and emotional data. This advice is made more appropriate by taking into account the emotional state. It uses OpenAI's generative AI model to generate advice by inputting prompt sentences such as the following:

[1049] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[1050] The generated advice is sent from the server to the device, which then visually displays it to the user. The user can refer to this advice when conducting meetings or making decisions. For example, the advice may suggest adopting an agile methodology in the early stages of a project, and considering combining it with a waterfall methodology as the project progresses, while monitoring the emotions at each phase.

[1051] In this way, the present invention automatically generates meeting minutes based on the audio data of meetings in industrial production sites, and supports efficient and harmonious consensus building by analyzing the personalities, ways of thinking, and emotions of the participants. As a specific example, when discussing how to proceed with a project, it can provide an effective approach according to the emotional state of the speaker, increasing the sense of agreement of the entire team.

[1052] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1053] Step 1:

[1054] The user collects the audio of the conference using a microphone and transmits the audio data to the system. The input is the audio data of the conference, and the output is the audio data received by the audio input device. In this step, the audio is input to the terminal in real time.

[1055] Step 2:

[1056] The device stores the received voice data locally and simultaneously transmits it to a server via the Internet. The input is the voice data received in real time, and the output is the voice data transmitted to the server. This allows the voice data to be stored and transmitted.

[1057] Step 3:

[1058] The server converts the received voice data into text data using the Google Cloud Speech-to-Text API. The input is the voice data stored on the server, and the output is text data. The converted text data is classified by each speaker.

[1059] Step 4:

[1060] The server automatically generates minutes based on the text data. The input is classified text data, and the output is organized minutes data. In this step, each statement is converted into minutes format.

[1061] Step 5:

[1062] The server analyzes the generated text data of the minutes using natural language processing (NLP). The input is the text data of the minutes, and the output is analytical data on the participants' personalities, thoughts, emotions, etc. Specifically, NLTK and spaCy are used to extract keywords and perform sentiment analysis on the text data.

[1063] Step 6:

[1064] The server uses an emotion engine to recognize user emotions from voice and text data, and integrates the recognized emotion data with the text data of the minutes. The input is voice and text data, and the output is the minutes data with added emotional context. It uses emotion analysis models such as TextBlob and Hugging Face.

[1065] Step 7:

[1066] The server generates specific advice to promote consensus building based on the analysis results and emotion data. The input is the analyzed minutes data and emotion data, and the output is advice data to support consensus building. Using OpenAI's generative AI model, the server inputs the following prompt:

[1067] "Given the following meeting transcription: 'Participant A: We should adopt an agile approach for the project. Participant B: I agree but we should also consider traditional methods in certain phases.', and considering the participants agreement are generally 'enthusiastic', provide concrete advice to promote."

[1068] Step 8:

[1069] The generated advice is sent from the server to the terminal, and the terminal visually displays this advice to the user. The input is the generated advice data, and the output is a visual display of the advice to the user. This allows the user to refer to the advice and proceed with the meeting or make decisions.

[1070] 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.

[1071] 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.

[1072] 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.

[1073] 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.

[1074] 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.

[1075] 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.

[1076] 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).

[1077] 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.

[1078] 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."

[1079] 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.

[1080] 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).

[1081] 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.

[1082] 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.

[1083] 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.

[1084] 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.

[1085] 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.

[1086] 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.

[1087] 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.

[1088] 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.

[1089] 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.

[1090] 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.

[1091] The following is further disclosed regarding the above embodiment.

[1092] (Claim 1)

[1093] means for receiving audio input;

[1094] means for converting voice data into text data;

[1095] A means for generating minutes based on text data;

[1096] A method for analyzing the text data of minutes to infer the personalities and thoughts of participants,

[1097] a means for generating advice to facilitate consensus building;

[1098] a means for displaying the generated advice;

[1099] A system including:

[1100] (Claim 2)

[1101] 10. The system of claim 1, further comprising means for transmitting audio data in real time.

[1102] (Claim 3)

[1103] 10. The system of claim 1, further comprising means for analyzing the text data using natural language processing techniques.

[1104] "Example 1"

[1105] (Claim 1)

[1106] means for receiving audio input;

[1107] a means for locally storing the audio data;

[1108] means for transmitting the stored voice data to a server;

[1109] means for converting voice data into text data;

[1110] A means for generating minutes based on text data;

[1111] A method for analyzing the text data of minutes to infer the personalities and thoughts of participants,

[1112] a means for generating advice to facilitate consensus building;

[1113] a means for displaying the generated advice;

[1114] A system including:

[1115] (Claim 2)

[1116] 10. The system of claim 1, further comprising means for transmitting audio data in real time.

[1117] (Claim 3)

[1118] 10. The system of claim 1, further comprising means for analyzing the text data using natural language processing techniques.

[1119] "Application Example 1"

[1120] (Claim 1)

[1121] means for receiving audio input;

[1122] means for converting voice data into text data;

[1123] A means for generating minutes based on text data;

[1124] A method for analyzing the text data of minutes to infer the personalities and thoughts of participants,

[1125] a means for generating advice to facilitate consensus building;

[1126] a means for displaying the generated advice;

[1127] a means for collecting in-vehicle voice data;

[1128] a means for transmitting the collected voice data to a cloud server in real time;

[1129] a means for displaying the minutes in the autonomous vehicle;

[1130] A system including:

[1131] (Claim 2)

[1132] 10. The system of claim 1, further comprising means for transmitting audio data in real time.

[1133] (Claim 3)

[1134] 10. The system of claim 1, further comprising means for analyzing the text data using natural language processing techniques.

[1135] "Example 2: Combining Emotion Engines"

[1136] (Claim 1)

[1137] means for collecting audio data;

[1138] means for receiving audio data in real time;

[1139] means for temporarily storing and transmitting the audio data via the Internet;

[1140] means for converting received voice data into text data;

[1141] A means for identifying and classifying text data by speaker;

[1142] A means for generating minutes based on the identified and classified text data;

[1143] A means for using natural language processing technology to analyze text data of the minutes;

[1144] A means of inferring the participants' personalities and ways of thinking from the analyzed text data,

[1145] means for recognizing emotions from speech data and text data;

[1146] a means for integrating the recognized emotion data with the text data of the minutes;

[1147] A means for generating advice that promotes consensus building based on the analysis results and emotion data;

[1148] a means for displaying the generated advice in real time;

[1149] A system including:

[1150] (Claim 2)

[1151] 10. The system of claim 1, further comprising means for transmitting audio data in real time.

[1152] (Claim 3)

[1153] 10. The system of claim 1, further comprising means for analyzing the text data using natural language processing techniques to infer information including the participants' personalities, thoughts, and emotions.

[1154] "Application example 2 when combining emotion engines"

[1155] (Claim 1)

[1156] means for receiving audio input;

[1157] means for converting voice data into text data;

[1158] A means for generating minutes based on text data;

[1159] A method for analyzing the text data of minutes to infer the personalities and thoughts of participants,

[1160] a means for generating advice to facilitate consensus building;

[1161] a means for displaying the generated advice;

[1162] A method to automatically generate meeting minutes in industrial production sites and support consensus building by analyzing and considering the emotions of participants.

[1163] A system including:

[1164] (Claim 2)

[1165] 10. The system of claim 1, further comprising means for transmitting audio data in real time.

[1166] (Claim 3)

[1167] 10. The system of claim 1, further comprising means for analyzing the text data using natural language processing techniques. [Explanation of symbols]

[1168] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving audio input; means for converting voice data into text data; A means for generating minutes based on text data; A method for analyzing the text data of minutes to infer the personalities and thoughts of participants, a means for generating advice to facilitate consensus building; a means for displaying the generated advice; A system including:

2. 10. The system of claim 1, further comprising means for transmitting audio data in real time.

3. 10. The system of claim 1, further comprising means for analyzing the text data using natural language processing techniques.

Images