System

Abstract

A system is provided.SOLUTION: A system comprising: means for directly outputting generated meeting minutes to a designated tool; means for capturing audio of a meeting in real time and converting the audio to text; means for automatically generating meeting minutes in real time using a generation AI; means for allowing a user to confirm and correct the generated meeting minutes; and means for receiving and authenticating destination information to the designated tool from the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional minutes generation systems capture the audio of a meeting and record the generated minutes in a dedicated tool, but then users have to manually copy and paste the minutes to apply them to other tools. This requires a lot of time and effort, and is a factor that reduces work efficiency. Therefore, there was a need for a way to smoothly output the minutes generated after the meeting to other tools, thereby improving work efficiency. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for directly outputting generated minutes to a specified tool, a means for capturing meeting audio in real time and converting it into text, a means for automatically generating minutes in real time using a generation AI, a means for allowing a user to check and correct the generated minutes, and a means for receiving and authenticating output destination information for the specified tool from the user. This system eliminates the need for users to manually move minutes to another tool after the meeting, significantly improving work efficiency.

[0006] The "generated minutes" are records of the contents of a meeting that are automatically generated based on the audio of the meeting.

[0007] "Designated tool" refers to the collaboration tool or project management tool designated by the user, and means the application or platform to which the minutes are directly output.

[0008] "Means" refers to a technical method or device used to achieve a specific function or process.

[0009] "Conference audio" refers to audio data spoken during a conference, and refers to sounds used to convey the contents of the conference.

[0010] "Real-time capture" refers to capturing and processing audio data simultaneously as the conference proceeds.

[0011] "Convert to text" refers to the process of converting audio data into written data.

[0012] "Generative AI" refers to a system or program that uses artificial intelligence technology to analyze data and automatically generate text.

[0013] "User" refers to the person or organization that uses this system and is the entity that manages the process of generating and outputting minutes.

[0014] "Confirmation and correction" refers to the process in which the user checks the contents of the generated minutes and edits them as necessary.

[0015] "Authentication" refers to checking the authentication information required to access the output destination tool specified by the user and allowing access.

[0016] "API" stands for Application Program Interface, and is a set of rules and tools for communicating between software programs. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to a designated tool. This system operates as follows.

[0039] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0040] The server activates a speech recognition engine based on the received conference information. The terminal then records the conference audio in real time and streams the audio data to the server. The server then receives the audio data in real time and passes it to the speech recognition engine.

[0041] A speech recognition engine on the server analyzes the recorded audio data and converts it into text in real time. The converted text data is temporarily stored on the server and broken down by speaker. The server then calls a generation AI to summarize the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are formatted into a template on the server.

[0042] The generated minutes are provided to the user as a function to check and correct them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually corrects them as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0043] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted in a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0044] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0045] A specific example is given below.

[0046] When the user clicks the "Start a new meeting" button on the device, the server activates a speech recognition engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination tool. Through this process, users can efficiently output meeting minutes to other tools, improving work efficiency.

[0047] The processing flow will be explained below.

[0048] Step 1:

[0049] A user accesses the minutes generation tool on a terminal and clicks the "Start a new meeting" button, which prompts the user to enter the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0050] Step 2:

[0051] The server configures the voice recognition engine based on the received conference information. The device then records the conference audio in real time and streams the audio data to the server.

[0052] Step 3:

[0053] The server receives the recorded voice data in real time, passes it to a speech recognition engine, and converts it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0054] Step 4:

[0055] The server passes the converted text data to the generation AI, which analyzes the text data, summarizes the meeting content, and generates minutes. The generated minutes are formatted in a template format and temporarily stored on the server.

[0056] Step 5:

[0057] The server notifies the user of the generated minutes as a URL for confirmation and correction. The user accesses this URL from their terminal to check the minutes and make corrections as necessary.

[0058] Step 6:

[0059] The user approves the changes and sends them to the server via the terminal, and the server updates the minutes with the changes received.

[0060] Step 7:

[0061] The user checks and sets the output destination tool, enters the authentication information for the output destination tool, and the device sends this to the server. The server calls the API of the output destination tool based on the authentication information and sends an authentication request.

[0062] Step 8:

[0063] If authentication is successful, the server sends the minutes to the destination tool in the specified format, for example, uploading the content to a specific section in a project management tool, or posting it to a specific channel in a messaging tool.

[0064] Step 9:

[0065] When the output is complete, the server sends a completion notification to the user, who can then check from their terminal that the output was successful.

[0066] Example 1

[0067] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0068] In today's business environment, many meetings are held, making it extremely important to efficiently generate meeting minutes. However, traditional methods require manual note-taking during meetings and then manual organization and recording of the content afterward. This takes time and effort, and can lead to reduced accuracy. Furthermore, sharing the recorded minutes with appropriate software or tools requires additional manual effort. This reduces work efficiency and increases the risk of important content being overlooked.

[0069] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0070] In this invention, the server includes: means for directly outputting the generated record to designated software; means for capturing meeting audio in real time and converting it to text; means for automatically generating minutes in real time using a generation AI; means for allowing a user to review and modify the generated minutes; means for receiving and authenticating output destination information for the designated software from the user; means for inputting meeting start information via a user interface; means for streaming audio data in real time; means for analyzing audio and segmenting the text data for each speaker; and means for formatting the summarized minutes into a template format. This allows minutes to be automatically generated from audio recorded during a meeting and instantly shared with designated software. This significantly reduces time and effort and improves the accuracy and efficiency of minutes.

[0071] A "recording" is a record of what is said or done at a meeting or other event, in text, audio, or other format.

[0072] "Software" means programs or applications that run on a digital device and provide specific functions or services.

[0073] A "voice recognition engine" refers to the algorithms and technology used to convert voice data into text data.

[0074] "Generative AI" is an artificial intelligence technology that learns patterns from large amounts of data and generates data such as text.

[0075] "User" refers to any individual or organization that uses the system or software.

[0076] A "user interface" refers to the screens and methods of operation that allow a user to interact with a computer system or software.

[0077] "Streaming" is a technology for transmitting and receiving data such as audio and video in real time.

[0078] "Analysis" is a technical process that refers to examining data to extract meaning.

[0079] A "template format" is a model of a document or data created according to a certain format.

[0080] "Authentication" is the process of verifying that a user or system is legitimate.

[0081] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to designated software. This system operates as follows.

[0082] First, when a user starts a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output software. This information is sent from the device to the server.

[0083] The server activates a voice recognition engine based on the received conference information. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the voice recognition engine. The voice recognition engine analyzes the audio data and converts it into text.

[0084] The converted text data is temporarily stored on the server and broken down by speaker. The server then calls the generation AI, which summarizes the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are then formatted into a template on the server.

[0085] The server then provides the user with the ability to check and modify the generated minutes. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually modifies them as necessary. The modifications are sent from the device to the server, and the minutes are updated.

[0086] Once the user has confirmed and set the destination software, the server sends an authentication request to the destination software's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination software in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0087] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other software, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0088] Specific examples are shown below.

[0089] For example, when a user clicks the "Start a new meeting" button on their device, the server activates a speech recognition engine and sets up a meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination software. Through this process, users can efficiently output meeting minutes to other software, improving work efficiency.

[0090] Example prompt sentence:

[0091] "The title of this meeting is 'Annual Planning Meeting' and the participants are A, B, and C. Specify the project management tool as the destination for the meeting minutes. When the meeting starts, record the audio, send it to the server, and generate and summarize the minutes in real time."

[0092] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0093] Step 1: User starts a conference

[0094] The user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output software. The input data is sent from the device to the server. The input data is basic information about the meeting, and the output is the transfer of information to the server.

[0095] Step 2: The server starts the speech recognition engine

[0096] The server starts the speech recognition engine based on the received conference information. Specifically, it sets up the conference using the title and participant list information. The input is the conference information, and the output is the initialized speech recognition engine.

[0097] Step 3: The device records the audio and streams it to the server

[0098] The terminal records the audio of the conference in real time. The recorded audio data is sent to the server via a streaming protocol (e.g., WebRTC, RTMP). The input is the recorded audio, and the output is the audio data transferred to the server.

[0099] Step 4: The server receives and analyzes the audio data.

[0100] The server receives voice data sent from the device in real time. The received voice data is passed to a voice recognition engine and converted into text data. The input is voice data, and the output is analyzed text data.

[0101] Step 5: The server passes the text data to the generation AI

[0102] The server temporarily stores the converted text data and breaks it down by speaker. This organizes it so that the content of each individual speech is clear. The organized text data is then passed to a generation AI, which instructs it to generate summarized minutes in real time. The input is organized text data, and the output is summarized minutes.

[0103] Step 6: The server formats the minutes

[0104] The server formats the minutes generated by the generation AI into a template format. Specifically, it arranges the minutes content based on a predetermined template and arranges it in an easy-to-read format. The input is a summarized minutes, and the output is the formatted minutes.

[0105] Step 7: The server notifies the user of the URL to check the minutes.

[0106] The server notifies the user of the URL where the formatted minutes are temporarily saved. The user can then access the minutes through the URL. The input is the formatted minutes, and the output is the notification URL.

[0107] Step 8: User reviews and edits the minutes

[0108] The user accesses the URL notified from the terminal and checks the generated minutes. If necessary, they manually correct the contents and send the corrections to the server. The input is correction instructions, and the output is the corrected minutes.

[0109] Step 9: The server reflects the changes

[0110] The server reflects the corrections received from the user in the minutes and updates them as the final version. The input is the corrected data, and the output is the final version of the minutes.

[0111] Step 10: The server sends the minutes to the destination

[0112] After the user confirms and sets the destination software, the server sends an authentication request to the destination software's API based on the authentication information. If authentication is successful, the server sends the minutes to the destination software in the specified format. The input is the authentication information and the final version of the minutes, and the output is the sending of the minutes to the destination software.

[0113] Step 11: Server sends notification of completion

[0114] Finally, the server notifies the user that the output is complete, allowing the user to quickly and efficiently share the meeting contents with other software. The input is the output status, and the output is the completion notification.

[0115] (Application example 1)

[0116] 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."

[0117] Quickly and accurately recording meetings and work reports held by managers and engineers in factories and efficiently sharing them with other management systems and tools is an important factor in significantly improving work efficiency. However, conventional methods require a lot of time and effort to convert speech to text and create summary minutes, which results in delays in reporting and sharing information after meetings.

[0118] 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.

[0119] In this invention, the server includes a means for directly outputting the generated information summary to a specified system, a means for capturing voice data in real time and converting it into text, and a means for automatically generating information summaries in real time using a generation AI. This allows for quick and efficient in-factory meetings and work reports, and by generating summaries in real time and checking and correcting them as necessary, it becomes possible to smoothly share information with management systems and tools.

[0120] A "generated information summary" is a summary text generated in real time from audio data using generation AI.

[0121] The term "designated system" refers to all systems that are linked as the output destination of information designated by the operator.

[0122] "Means" refers to elements or mechanisms that are installed to achieve a specific function in an invention.

[0123] "Audio data" refers to digitally recorded data of audio during meetings or work reports.

[0124] "Real-time capture" refers to the immediate recording and processing of audio data.

[0125] "Converting to text" refers to converting voice data into character string information using voice recognition technology.

[0126] "Generative AI" refers to algorithms or systems that use artificial intelligence techniques to process data and generate a specific output, in this case a summary.

[0127] "Automatically generating information summaries" refers to the process in which generative AI analyzes input data and automatically generates summary sentences.

[0128] "Operator" refers to the person who uses the system to hold meetings and report on work.

[0129] "Display device" refers to equipment for visually displaying information, such as a head-mounted display or smart glasses.

[0130] "Verify and correct" refers to verifying the content of the generated information summary and making corrections or changes as necessary.

[0131] "Speech recognition technology" refers to all technologies for analyzing human speech and converting it into text information.

[0132] "API" refers to an interface for exchanging functions and data between different software programs.

[0133] The system for realizing this invention generates and manages information summaries using advanced technologies such as a voice recognition engine and generation AI. The program processing in this system will be described in detail below.

[0134] This system is mainly composed of three elements: a server, a terminal, and an operator. The server processes the audio data and creates a summary using a generation AI, while the operator on the terminal checks and corrects the summary.

[0135] System flow:

[0136] start

[0137] 1. Starting a meeting and entering information

[0138] When an operator starts a meeting, he or she accesses the minutes generation tool interface on the terminal and clicks the "Start a new meeting" button. At this point, the operator enters the meeting title, participant list, and output destination system (e.g., a project management system). This information is sent from the terminal to the server.

[0139] 2. Audio capture and text conversion

[0140] The server starts a speech recognition engine (e.g., Google Speech Recognition API) based on the received conference information. The devices record the conference audio in real time and stream the audio data to the server. The server receives the audio data in real time and converts it into text using the speech recognition engine.

[0141] 3. Summary generation using generative AI

[0142] A generative AI (e.g., OpenAI's GPT model) on the server analyzes the text data to summarize it. It is given a prompt like this:

[0143] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0144] The generation AI logically organizes the meeting content based on text data and generates summarized information.

[0145] 4. Review and correct the generated information summary

[0146] The generated information summary is formatted in a template format on the server, and the operator can review and manually modify the summary via a terminal.

[0147] 5. Information output and sharing

[0148] When the operator confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the operator. If authentication is successful, the server sends the summarized information to the destination system in the specified format. This allows the summarized information to be automatically uploaded to, for example, a project management system using an API.

[0149] The system allows users to view and modify the generated information summary using a display device (e.g., head-mounted display or smart glasses), supporting efficient business operations within the factory.

[0150] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0151] Step 1:

[0152] Starting a meeting and entering information

[0153] When a user wants to start a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. The device sends the meeting title, participant list, and destination system (e.g., a project management system) as input to the server. The server receives this information and performs initial setup for the meeting.

[0154] Step 2:

[0155] Audio capture and text conversion

[0156] The device records the audio of the meeting in real time and streams the audio data to the server. The server receives the audio data and activates a speech recognition engine (e.g., Google Speech Recognition API) to convert the audio data into text in real time. The input is the audio data streamed in real time, and the output is character string data converted from audio to text.

[0157] Step 3:

[0158] Summary generation using generative AI

[0159] The server uses a generative AI (e.g., OpenAI's GPT model) to analyze the text data obtained in step 2 and generate a summary. By providing the generative AI with prompts such as the following, it can obtain an appropriate summary:

[0160] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0161] The input is text data obtained from a speech recognition engine, and the output is a summary generated by a generative AI.

[0162] Step 4:

[0163] Review and correct the generated information summary

[0164] The server formats the generated information summary into a template and notifies the user of the URL where it is temporarily saved. The user can access the URL from their device to check the generated information summary and manually edit it if necessary. The input is the summary generated by the generation AI, and the output is the final summary after the user has checked and edited it.

[0165] Step 5:

[0166] Output and sharing of information

[0167] When the user confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the user. If authentication is successful, the server sends the summarized information to the destination system in the specified format. For example, the summary information can be automatically uploaded to a project management system using an API. The input is the authentication information from the user and the final summary statement, and the output is the completion of sending the summary information to the destination system.

[0168] 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.

[0169] The present invention relates to a system that can efficiently generate minutes of a meeting, output them directly to a specified tool, and recognize the emotions of users and reflect them in the minutes. This system operates as follows.

[0170] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0171] Based on the conference information received by the server, the server sets up the speech recognition engine. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0172] At the same time, the server also starts an emotion engine, which analyzes the user's emotional state in real time from the recorded voice data. Once the emotion engine analyzes the emotional state, that information is set to be processed together with the text data. The server passes the converted text data and emotion data to the generation AI, which analyzes this data to summarize the meeting content and add emotion tags to the generated minutes. The minutes with the emotion tags added are formatted into a template on the server and temporarily saved.

[0173] The generated minutes are provided to the user as a function for reviewing and correcting them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, reviews the generated minutes, and manually corrects emotion tags and content as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0174] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, the server sends an HTTP POST request to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the server pastes the meeting minutes into a specific section in the board. In the case of a messaging tool, the server posts the meeting minutes to a specific channel.

[0175] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0176] As a concrete example, consider the following scenario.

[0177] When the user clicks the "Start a new meeting" button on the device, the server activates the speech recognition engine and emotion engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio, converts it into text, and passes it to the generation AI along with emotion data analyzed by the emotion engine. The generation AI processes this data to generate summarized minutes, which are then saved on the server with emotion tags. The user can review and edit the minutes and set the output tool, and the server automatically outputs the minutes to the appropriate tool. Through this process, users can efficiently create meeting minutes and output them to other tools.

[0178] The processing flow will be explained below.

[0179] Step 1:

[0180] The user accesses the minutes generation tool on the terminal and clicks the "Start a new meeting" button. The user enters the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0181] Step 2:

[0182] Based on the received conference information, the server configures the startup settings for the speech recognition engine and emotion engine. The server prepares the speech recognition module and emotion recognition module.

[0183] Step 3:

[0184] The device records the audio of the meeting in real time and streams the audio data to the server, where it is immediately transferred.

[0185] Step 4:

[0186] The server passes the received voice data to a speech recognition engine to convert it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0187] Step 5:

[0188] At the same time, the server uses an emotion engine to analyze the speaker's emotional state from the audio data in real time, and the emotion engine analyzes the emotional state and adds it to the text data as an appropriate tag.

[0189] Step 6:

[0190] The server passes the converted text data and emotion data to the generation AI, which analyzes this data and generates minutes summarizing the meeting content. The minutes with added emotion tags are formatted into a template on the server and temporarily saved.

[0191] Step 7:

[0192] The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device and checks the generated minutes. The user manually edits the emotion tags and content as necessary, and sends the edits from their device to the server.

[0193] Step 8:

[0194] The server updates the minutes with the received corrections. The user checks and sets the output destination tool for the meeting, enters the authentication information for the output destination tool, and the terminal sends this to the server.

[0195] Step 9:

[0196] The server calls the destination tool's API based on the authentication information received from the user and sends an authentication request. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted in a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0197] Step 10:

[0198] When the output is complete, the server sends a completion notification to the user. The user can then confirm that the output was performed properly from their terminal and that the minutes of the meeting were properly recorded in the specified tool.

[0199] Example 2

[0200] 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."

[0201] When creating meeting minutes, manual input and editing requires a lot of time and effort, hindering efficient business operations. Furthermore, since there is no way to reflect the emotions expressed during the meeting in the minutes, the minutes themselves may not accurately convey the content. Furthermore, since there is no system for automatically outputting the generated minutes to a specified tool, duplicate input is likely to occur, which could lead to reduced business efficiency.

[0202] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing conference audio in real time and converting it into text, means for analyzing user emotions from the audio data, and means for automatically generating minutes using a generation AI and reflecting emotional information. This makes it possible to efficiently generate minutes of a conference and create accurate minutes that reflect emotional information. In addition, by providing means for automatically outputting the generated minutes to a specific tool, it is possible to prevent double entry and improve work efficiency.

[0203] "Means for capturing meeting audio in real time and converting it into text" refers to technology or equipment used to collect speech during a meeting as audio data and instantly convert that audio data into text information.

[0204] "Means for analyzing user emotions from voice data" refers to a technology or device used to determine the speaker's emotional state in real time based on recorded voice data and extract that emotional information.

[0205] "Means for automatically generating minutes using generative AI and reflecting emotional information" refers to a technology or device that uses artificial intelligence (AI) technology to analyze collected text data and emotional information, automatically create summarized minutes based on that data, and assign emotional tags.

[0206] "Means for allowing users to review and modify the generated minutes" refers to technology or devices that provide an interface or functionality that allows users to review and modify the automatically generated minutes.

[0207] "Means for outputting meeting minutes directly to a designated tool" means the technology or device used to automatically send the generated meeting minutes to a specific collaboration or project management tool and make them immediately available within that tool.

[0208] "Means for receiving and authenticating destination information for a user-specified tool" refers to the technology or device used to receive information about a user-specified tool and perform the appropriate authentication procedures for that tool.

[0209] This invention relates to a system for efficiently generating minutes of a meeting, outputting the minutes directly to a specified tool, and recognizing and reflecting the emotions of users in the minutes. A specific embodiment of this system is described below.

[0210] First, when a user starts a meeting, they access the interface of the minutes generation tool on their terminal and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the terminal to the server.

[0211] The server configures a speech recognition engine (e.g., Google Cloud Speech-to-Text) based on the received meeting information. Next, the device records the meeting audio in real time and streams the audio data to the server. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0212] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time from the recorded audio data. Once the emotion analysis engine analyzes the emotional state, it processes that information along with the text data. The server then passes the converted text data and emotion data to a generation AI (e.g., OpenAI's GPT-3), which analyzes this data to summarize the meeting content and generate minutes with emotion tags.

[0213] An example of a prompt sentence to input to the generative AI model is as follows:

[0214] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[0215] The minutes with emotion tags are formatted into a template on the server and temporarily saved. The server then notifies the user of the URL where the minutes are saved.

[0216] The user accesses this URL from their device, checks the generated minutes, and manually edits emotion tags and content as necessary. Once edits are complete, the edited data is sent from the device to the server, and the minutes are updated.

[0217] The user then confirms and sets the destination tool. The server sends an authentication request to the destination tool's API using the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the specified tool via an HTTP POST request. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted into a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0218] Finally, when the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the meeting content with other tools. Furthermore, because the system executes all processes in real time, manual input work after the meeting is over is no longer necessary.

[0219] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0220] Step 1:

[0221] To start a meeting, a user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output destination tool. This becomes input data and is sent from the device to the server. The server receives this data and performs initial settings to start the meeting session.

[0222] Step 2:

[0223] Based on the conference information received by the server, the server configures and starts a speech recognition engine (e.g., Google Cloud Speech-to-Text). At this time, the server allocates the necessary resources and prepares to capture voice data. The input data is the conference information, and the output is the running status of the speech recognition engine.

[0224] Step 3:

[0225] The device records the audio of the meeting in real time and streams the audio data to the server. This process temporarily stores the audio captured from the device's microphone in a buffer and sends it to the server using a real-time communication protocol (e.g., WebSocket). The input is audio data, and the output is real-time streaming to the server.

[0226] Step 4:

[0227] The server passes the received voice data to a voice recognition engine and converts it into text data in real time. The voice recognition engine analyzes the voice and breaks down the text for each speaker. The input is voice data and the output is text data. At this time, the server identifies each speaker and organizes the content of their speech.

[0228] Step 5:

[0229] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time based on the recorded voice data. The server then configures the system to process the emotional analysis results together with the text data. The input is the voice data, and the output is the emotion analysis results.

[0230] Step 6:

[0231] The server passes the converted text data and emotion data to a generative AI (e.g., OpenAI's GPT-3), which summarizes the meeting content and generates minutes with emotion tags. The input is text data and emotion data, and the output is a summarized minutes. An example of a prompt sentence to input to the generative AI model is as follows:

[0232] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[0233] Step 7:

[0234] The generated minutes are formatted in a template format on the server and temporarily saved. The server notifies the user of the URL where the generated minutes are temporarily saved. The input is the minutes data, and the output is the URL of the temporarily saved file.

[0235] Step 8:

[0236] The user accesses the URL notified from the device and checks the generated minutes. If necessary, they manually correct the emotion tags and content. Once the corrections are complete, the corrected data is sent from the device to the server and the minutes are updated. The input is the user's corrected data, and the output is the updated minutes.

[0237] Step 9:

[0238] The user checks and sets the destination tool. The server sends an authentication request to the destination tool's API based on the authentication information received from the user. The input is the authentication information, and the output is the authentication status.

[0239] Step 10:

[0240] If authentication is successful, the server uses an HTTP POST request to send the minutes to the specified tool. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel. The input is the minutes data, and the output is the completion of upload to the tool.

[0241] Step 11:

[0242] When the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the contents of the meeting with other tools. The input is the output status to the tool, and the output is the completion notification to the user.

[0243] (Application example 2)

[0244] 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."

[0245] There is a need for a system that can efficiently generate meeting minutes and accurately grasp key points of discussions and changes in participants' emotions by reflecting their emotional states in real time. Furthermore, in certain on-site environments, there is a need for a method to improve work efficiency by quickly notifying managers of the meeting content and emotion analysis results. This invention was developed to solve these problems.

[0246] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0247] In this invention, the server includes means for directly outputting the generated minutes to a specified tool, means for capturing meeting audio in real time and converting it into text, means for automatically generating minutes in real time using a generation AI, means for allowing a user to confirm and correct the generated minutes, means for receiving and authenticating output destination information for the specified tool from the user, means for a factory floor worker to start a meeting using smart glasses and record audio in real time and convert it into text, means for analyzing and recording emotional states, and means for remotely notifying a manager of the automatically generated minutes and the emotional analysis results, thereby enabling the real-time generation and output of meeting minutes and emotional states.

[0248] Minutes are documents that record the contents of a meeting, including what was said, decisions made, and summaries of the agenda.

[0249] "Designated tool" refers to an external tool, such as a collaboration tool or project management tool, selected by the user to send the minutes.

[0250] A "speech recognition engine" is a software or hardware technology for converting speech into text in real time.

[0251] "Generative AI" refers to artificial intelligence technology that automatically generates meeting minutes from given data.

[0252] "Emotional state" refers to the changes in the user's emotions and state that are analyzed from the voice data.

[0253] "Smart glasses" are wearable devices that can record conference audio in real time and display information.

[0254] "Destination information" is information about the tool or location designated by the user to send the minutes.

[0255] "API" stands for Application Programming Interface and refers to the methods and tools that allow different software components to communicate with each other.

[0256] "Emotion analysis results" are information on emotional states obtained by analyzing voice data.

[0257] A "manager" is a person in charge of managing the activities of field workers and the contents of meetings within a factory or organization.

[0258] In implementing the invention, the system mainly uses a server, smart glasses, and an administrator's terminal. The detailed process of each step is described below.

[0259] First, a field worker starts a conference using the smart glasses. The smart glasses are equipped with a microphone for recording audio in real time. When the user clicks the "Start a new conference" button on the smart glasses to start the conference, conference information is sent to the server. This information includes the conference title, participant list, and output destination tool.

[0260] Based on the received information, the server activates a speech recognition engine (e.g., Google Speech-to-Text API). Then, the smart glasses stream the recorded audio data of the meeting to the server in real time. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio for each speaker and generates specific text data.

[0261] At the same time, the server activates an emotion engine (e.g., Hugging Face emotion analysis model) to analyze the user's emotional state from the voice data in real time. Once the emotion engine analyzes the emotional state, that information is processed together with the text data.

[0262] The generated text data and emotion data are passed to a generation AI (for example, OpenAI's GPT-3) on the server. The generation AI analyzes this data and summarizes the meeting content. At the same time, minutes with emotion tags are generated. These minutes are temporarily stored on the server.

[0263] Users can view the generated minutes through the smart glasses interface and manually edit emotion tags and content as needed, or administrators can view and edit the minutes from their own devices.

[0264] Finally, the minutes that have been checked and corrected are automatically sent to the destination tool (for example, a project management tool or collaboration tool). The server achieves this output by using the destination tool's API based on the authentication information received from the user and sending the minutes in the specified format.

[0265] For example, during a safety meeting at a worksite, smart glasses record workers' speech and perform text and sentiment analysis. This information is displayed in real time on a manager's dashboard, allowing them to take the necessary action quickly. An example of a prompt for the generative AI model is as follows:

[0266] Example prompt sentence:

[0267] Transcribe an audio recording of a factory floor meeting and summarize it as follows:

[0268] 1. Main agenda of the meeting.

[0269] 2. Important statements by each speaker.

[0270] 3. Analysis of emotional state.

[0271] Please include the meeting title, participant list, and destination tool.

[0272] By implementing the invention in this way, it is possible to generate and output meeting minutes and emotional states in real time, thereby improving meeting efficiency and enabling quick information sharing.

[0273] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0274] Step 1:

[0275] The user uses the smart glasses to click the "Start a new meeting" button. At this time, the meeting title, participant list, and output tool are entered, and this is sent from the device to the server. The entered data is temporarily saved on the server as meeting information.

[0276] Step 2:

[0277] Based on the received meeting information, the server launches a speech recognition engine (e.g., Google Speech-to-Text API). The audio data of the meeting recorded by the smart glasses is streamed to the server in real time. The audio data is received as input, passed to the speech recognition engine, and converted into text data. This process outputs the meeting audio as a text-formatted transcript.

[0278] Step 3:

[0279] The server runs an emotion engine (e.g., Hugging Face emotion analysis model) and processes the voice data separately for emotion analysis. The input voice data is analyzed and the emotional state is digitized. The emotion data is generated in parallel with the text data, and these data are later integrated.

[0280] Step 4:

[0281] The server passes the text data generated by the speech recognition engine and the emotion data generated by the emotion engine to a generation AI (for example, OpenAI's GPT-3). The generation AI receives these data as input and analyzes and summarizes them. This process generates a summary of the meeting minutes with emotion tags added, which are then temporarily saved. The generation of these minutes completes the integration of the data summary and emotion information.

[0282] Step 5:

[0283] The user can review the summary minutes generated through the smart glasses interface and make any necessary corrections. The correction information entered through the smart glasses is then sent to the server, and the minutes are updated. After user confirmation, the final minutes are confirmed.

[0284] Step 6:

[0285] The server sends the final minutes confirmed by the user to the specified output tool (for example, a project management tool or collaboration tool). To send the minutes, the server uses the API of each tool and performs appropriate authentication based on the user's authentication information. This allows the minutes to be uploaded to the output tool in the appropriate format.

[0286] Step 7:

[0287] Once the output process is complete, the server sends a notification to the user. The notification is displayed on the smart glasses or the administrator's device, allowing the user to confirm that the meeting content has been output correctly. At this step, all processing is complete.

[0288] 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.

[0289] 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.

[0290] 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.

[0291] [Second embodiment]

[0292] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0293] 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.

[0294] 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).

[0295] 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.

[0296] 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.

[0297] 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).

[0298] 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.

[0299] 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.

[0300] 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.

[0301] 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.

[0302] 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.

[0303] 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."

[0304] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to a designated tool. This system operates as follows.

[0305] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0306] The server activates a speech recognition engine based on the received conference information. The terminal then records the conference audio in real time and streams the audio data to the server. The server then receives the audio data in real time and passes it to the speech recognition engine.

[0307] A speech recognition engine on the server analyzes the recorded audio data and converts it into text in real time. The converted text data is temporarily stored on the server and broken down by speaker. The server then calls a generation AI to summarize the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are formatted into a template on the server.

[0308] The generated minutes are provided to the user as a function to check and correct them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually corrects them as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0309] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted in a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0310] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0311] A specific example is given below.

[0312] When the user clicks the "Start a new meeting" button on the device, the server activates a speech recognition engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination tool. Through this process, users can efficiently output meeting minutes to other tools, improving work efficiency.

[0313] The processing flow will be explained below.

[0314] Step 1:

[0315] A user accesses the minutes generation tool on a terminal and clicks the "Start a new meeting" button, which prompts the user to enter the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0316] Step 2:

[0317] The server configures the voice recognition engine based on the received conference information. The device then records the conference audio in real time and streams the audio data to the server.

[0318] Step 3:

[0319] The server receives the recorded voice data in real time, passes it to a speech recognition engine, and converts it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0320] Step 4:

[0321] The server passes the converted text data to the generation AI, which analyzes the text data, summarizes the meeting content, and generates minutes. The generated minutes are formatted in a template format and temporarily stored on the server.

[0322] Step 5:

[0323] The server notifies the user of the generated minutes as a URL for confirmation and correction. The user accesses this URL from their terminal to check the minutes and make corrections as necessary.

[0324] Step 6:

[0325] The user approves the changes and sends them to the server via the terminal, and the server updates the minutes with the changes received.

[0326] Step 7:

[0327] The user checks and sets the output destination tool, enters the authentication information for the output destination tool, and the device sends this to the server. The server calls the API of the output destination tool based on the authentication information and sends an authentication request.

[0328] Step 8:

[0329] If authentication is successful, the server sends the minutes to the destination tool in the specified format, for example, uploading the content to a specific section in a project management tool, or posting it to a specific channel in a messaging tool.

[0330] Step 9:

[0331] When the output is complete, the server sends a completion notification to the user, who can then check from their terminal that the output was successful.

[0332] Example 1

[0333] 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."

[0334] In today's business environment, many meetings are held, making it extremely important to efficiently generate meeting minutes. However, traditional methods require manual note-taking during meetings and then manual organization and recording of the content afterward. This takes time and effort, and can lead to reduced accuracy. Furthermore, sharing the recorded minutes with appropriate software or tools requires additional manual effort. This reduces work efficiency and increases the risk of important content being overlooked.

[0335] 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.

[0336] In this invention, the server includes: means for directly outputting the generated record to designated software; means for capturing meeting audio in real time and converting it to text; means for automatically generating minutes in real time using a generation AI; means for allowing a user to review and modify the generated minutes; means for receiving and authenticating output destination information for the designated software from the user; means for inputting meeting start information via a user interface; means for streaming audio data in real time; means for analyzing audio and segmenting the text data for each speaker; and means for formatting the summarized minutes into a template format. This allows minutes to be automatically generated from audio recorded during a meeting and instantly shared with designated software. This significantly reduces time and effort and improves the accuracy and efficiency of minutes.

[0337] A "recording" is a record of what is said or done at a meeting or other event, in text, audio, or other format.

[0338] "Software" means programs or applications that run on a digital device and provide specific functions or services.

[0339] A "voice recognition engine" refers to the algorithms and technology used to convert voice data into text data.

[0340] "Generative AI" is an artificial intelligence technology that learns patterns from large amounts of data and generates data such as text.

[0341] "User" refers to any individual or organization that uses the system or software.

[0342] A "user interface" refers to the screens and methods of operation that allow a user to interact with a computer system or software.

[0343] "Streaming" is a technology for transmitting and receiving data such as audio and video in real time.

[0344] "Analysis" is a technical process that refers to examining data to extract meaning.

[0345] A "template format" is a model of a document or data created according to a certain format.

[0346] "Authentication" is the process of verifying that a user or system is legitimate.

[0347] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to designated software. This system operates as follows.

[0348] First, when a user starts a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output software. This information is sent from the device to the server.

[0349] The server activates a voice recognition engine based on the received conference information. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the voice recognition engine. The voice recognition engine analyzes the audio data and converts it into text.

[0350] The converted text data is temporarily stored on the server and broken down by speaker. The server then calls the generation AI, which summarizes the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are then formatted into a template on the server.

[0351] The server then provides the user with the ability to check and modify the generated minutes. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually modifies them as necessary. The modifications are sent from the device to the server, and the minutes are updated.

[0352] Once the user has confirmed and set the destination software, the server sends an authentication request to the destination software's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination software in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0353] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other software, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0354] Specific examples are shown below.

[0355] For example, when a user clicks the "Start a new meeting" button on their device, the server activates a speech recognition engine and sets up a meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination software. Through this process, users can efficiently output meeting minutes to other software, improving work efficiency.

[0356] Example prompt sentence:

[0357] "The title of this meeting is 'Annual Planning Meeting' and the participants are A, B, and C. Specify the project management tool as the destination for the meeting minutes. When the meeting starts, record the audio, send it to the server, and generate and summarize the minutes in real time."

[0358] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0359] Step 1: User starts a conference

[0360] The user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output software. The input data is sent from the device to the server. The input data is basic information about the meeting, and the output is the transfer of information to the server.

[0361] Step 2: The server starts the speech recognition engine

[0362] The server starts the speech recognition engine based on the received conference information. Specifically, it sets up the conference using the title and participant list information. The input is the conference information, and the output is the initialized speech recognition engine.

[0363] Step 3: The device records the audio and streams it to the server

[0364] The terminal records the audio of the conference in real time. The recorded audio data is sent to the server via a streaming protocol (e.g., WebRTC, RTMP). The input is the recorded audio, and the output is the audio data transferred to the server.

[0365] Step 4: The server receives and analyzes the audio data.

[0366] The server receives voice data sent from the device in real time. The received voice data is passed to a voice recognition engine and converted into text data. The input is voice data, and the output is analyzed text data.

[0367] Step 5: The server passes the text data to the generation AI

[0368] The server temporarily stores the converted text data and breaks it down by speaker. This organizes it so that the content of each individual speech is clear. The organized text data is then passed to a generation AI, which instructs it to generate summarized minutes in real time. The input is organized text data, and the output is summarized minutes.

[0369] Step 6: The server formats the minutes

[0370] The server formats the minutes generated by the generation AI into a template format. Specifically, it arranges the minutes content based on a predetermined template and arranges it in an easy-to-read format. The input is a summarized minutes, and the output is the formatted minutes.

[0371] Step 7: The server notifies the user of the URL to check the minutes.

[0372] The server notifies the user of the URL where the formatted minutes are temporarily saved. The user can then access the minutes through the URL. The input is the formatted minutes, and the output is the notification URL.

[0373] Step 8: User reviews and edits the minutes

[0374] The user accesses the URL notified from the terminal and checks the generated minutes. If necessary, they manually correct the contents and send the corrections to the server. The input is correction instructions, and the output is the corrected minutes.

[0375] Step 9: The server reflects the changes

[0376] The server reflects the corrections received from the user in the minutes and updates them as the final version. The input is the corrected data, and the output is the final version of the minutes.

[0377] Step 10: The server sends the minutes to the destination

[0378] After the user confirms and sets the destination software, the server sends an authentication request to the destination software's API based on the authentication information. If authentication is successful, the server sends the minutes to the destination software in the specified format. The input is the authentication information and the final version of the minutes, and the output is the sending of the minutes to the destination software.

[0379] Step 11: Server sends notification of completion

[0380] Finally, the server notifies the user that the output is complete, allowing the user to quickly and efficiently share the meeting contents with other software. The input is the output status, and the output is the completion notification.

[0381] (Application example 1)

[0382] 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."

[0383] Quickly and accurately recording meetings and work reports held by managers and engineers in factories and efficiently sharing them with other management systems and tools is an important factor in significantly improving work efficiency. However, conventional methods require a lot of time and effort to convert speech to text and create summary minutes, which results in delays in reporting and sharing information after meetings.

[0384] 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.

[0385] In this invention, the server includes a means for directly outputting the generated information summary to a specified system, a means for capturing voice data in real time and converting it into text, and a means for automatically generating information summaries in real time using a generation AI. This allows for quick and efficient in-factory meetings and work reports, and by generating summaries in real time and checking and correcting them as necessary, it becomes possible to smoothly share information with management systems and tools.

[0386] A "generated information summary" is a summary text generated in real time from audio data using generation AI.

[0387] The term "designated system" refers to all systems that are linked as the output destination of information designated by the operator.

[0388] "Means" refers to elements or mechanisms that are installed to achieve a specific function in an invention.

[0389] "Audio data" refers to digitally recorded data of audio during meetings or work reports.

[0390] "Real-time capture" refers to the immediate recording and processing of audio data.

[0391] "Converting to text" refers to converting voice data into character string information using voice recognition technology.

[0392] "Generative AI" refers to algorithms or systems that use artificial intelligence techniques to process data and generate a specific output, in this case a summary.

[0393] "Automatically generating information summaries" refers to the process in which generative AI analyzes input data and automatically generates summary sentences.

[0394] "Operator" refers to the person who uses the system to hold meetings and report on work.

[0395] "Display device" refers to equipment for visually displaying information, such as a head-mounted display or smart glasses.

[0396] "Verify and correct" refers to verifying the content of the generated information summary and making corrections or changes as necessary.

[0397] "Speech recognition technology" refers to all technologies for analyzing human speech and converting it into text information.

[0398] "API" refers to an interface for exchanging functions and data between different software programs.

[0399] The system for realizing this invention generates and manages information summaries using advanced technologies such as a voice recognition engine and generation AI. The program processing in this system will be described in detail below.

[0400] This system is mainly composed of three elements: a server, a terminal, and an operator. The server processes the audio data and creates a summary using a generation AI, while the operator on the terminal checks and corrects the summary.

[0401] System flow:

[0402] start

[0403] 1. Starting a meeting and entering information

[0404] When an operator starts a meeting, he or she accesses the minutes generation tool interface on the terminal and clicks the "Start a new meeting" button. At this point, the operator enters the meeting title, participant list, and output destination system (e.g., a project management system). This information is sent from the terminal to the server.

[0405] 2. Audio capture and text conversion

[0406] The server starts a speech recognition engine (e.g., Google Speech Recognition API) based on the received conference information. The devices record the conference audio in real time and stream the audio data to the server. The server receives the audio data in real time and converts it into text using the speech recognition engine.

[0407] 3. Summary generation using generative AI

[0408] A generative AI (e.g., OpenAI's GPT model) on the server analyzes the text data to summarize it. It is given a prompt like this:

[0409] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0410] The generation AI logically organizes the meeting content based on text data and generates summarized information.

[0411] 4. Review and correct the generated information summary

[0412] The generated information summary is formatted in a template format on the server, and the operator can review and manually modify the summary via a terminal.

[0413] 5. Information output and sharing

[0414] When the operator confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the operator. If authentication is successful, the server sends the summarized information to the destination system in the specified format. This allows the summarized information to be automatically uploaded to, for example, a project management system using an API.

[0415] The system allows users to view and modify the generated information summary using a display device (e.g., head-mounted display or smart glasses), supporting efficient business operations within the factory.

[0416] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0417] Step 1:

[0418] Starting a meeting and entering information

[0419] When a user wants to start a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. The device sends the meeting title, participant list, and destination system (e.g., a project management system) as input to the server. The server receives this information and performs initial setup for the meeting.

[0420] Step 2:

[0421] Audio capture and text conversion

[0422] The device records the audio of the meeting in real time and streams the audio data to the server. The server receives the audio data and activates a speech recognition engine (e.g., Google Speech Recognition API) to convert the audio data into text in real time. The input is the audio data streamed in real time, and the output is character string data converted from audio to text.

[0423] Step 3:

[0424] Summary generation using generative AI

[0425] The server uses a generative AI (e.g., OpenAI's GPT model) to analyze the text data obtained in step 2 and generate a summary. By providing the generative AI with prompts such as the following, it can obtain an appropriate summary:

[0426] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0427] The input is text data obtained from a speech recognition engine, and the output is a summary generated by a generative AI.

[0428] Step 4:

[0429] Review and correct the generated information summary

[0430] The server formats the generated information summary into a template and notifies the user of the URL where it is temporarily saved. The user can access the URL from their device to check the generated information summary and manually edit it if necessary. The input is the summary generated by the generation AI, and the output is the final summary after the user has checked and edited it.

[0431] Step 5:

[0432] Output and sharing of information

[0433] When the user confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the user. If authentication is successful, the server sends the summarized information to the destination system in the specified format. For example, the summary information can be automatically uploaded to a project management system using an API. The input is the authentication information from the user and the final summary statement, and the output is the completion of sending the summary information to the destination system.

[0434] 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.

[0435] The present invention relates to a system that can efficiently generate minutes of a meeting, output them directly to a specified tool, and recognize the emotions of users and reflect them in the minutes. This system operates as follows.

[0436] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0437] Based on the conference information received by the server, the server sets up the speech recognition engine. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0438] At the same time, the server also starts an emotion engine, which analyzes the user's emotional state in real time from the recorded voice data. Once the emotion engine analyzes the emotional state, that information is set to be processed together with the text data. The server passes the converted text data and emotion data to the generation AI, which analyzes this data to summarize the meeting content and add emotion tags to the generated minutes. The minutes with the emotion tags added are formatted into a template on the server and temporarily saved.

[0439] The generated minutes are provided to the user as a function for reviewing and correcting them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, reviews the generated minutes, and manually corrects emotion tags and content as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0440] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, the server sends an HTTP POST request to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the server pastes the meeting minutes into a specific section in the board. In the case of a messaging tool, the server posts the meeting minutes to a specific channel.

[0441] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0442] As a concrete example, consider the following scenario.

[0443] When the user clicks the "Start a new meeting" button on the device, the server activates the speech recognition engine and emotion engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio, converts it into text, and passes it to the generation AI along with emotion data analyzed by the emotion engine. The generation AI processes this data to generate summarized minutes, which are then saved on the server with emotion tags. The user can review and edit the minutes and set the output tool, and the server automatically outputs the minutes to the appropriate tool. Through this process, users can efficiently create meeting minutes and output them to other tools.

[0444] The processing flow will be explained below.

[0445] Step 1:

[0446] The user accesses the minutes generation tool on the terminal and clicks the "Start a new meeting" button. The user enters the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0447] Step 2:

[0448] Based on the received conference information, the server configures the startup settings for the speech recognition engine and emotion engine. The server prepares the speech recognition module and emotion recognition module.

[0449] Step 3:

[0450] The device records the audio of the meeting in real time and streams the audio data to the server, where it is immediately transferred.

[0451] Step 4:

[0452] The server passes the received voice data to a speech recognition engine to convert it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0453] Step 5:

[0454] At the same time, the server uses an emotion engine to analyze the speaker's emotional state from the audio data in real time, and the emotion engine analyzes the emotional state and adds it to the text data as an appropriate tag.

[0455] Step 6:

[0456] The server passes the converted text data and emotion data to the generation AI, which analyzes this data and generates minutes summarizing the meeting content. The minutes with added emotion tags are formatted into a template on the server and temporarily saved.

[0457] Step 7:

[0458] The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device and checks the generated minutes. The user manually edits the emotion tags and content as necessary, and sends the edits from their device to the server.

[0459] Step 8:

[0460] The server updates the minutes with the received corrections. The user checks and sets the output destination tool for the meeting, enters the authentication information for the output destination tool, and the terminal sends this to the server.

[0461] Step 9:

[0462] The server calls the destination tool's API based on the authentication information received from the user and sends an authentication request. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted in a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0463] Step 10:

[0464] When the output is complete, the server sends a completion notification to the user. The user can then confirm that the output was performed properly from their terminal and that the minutes of the meeting were properly recorded in the specified tool.

[0465] Example 2

[0466] 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."

[0467] When creating meeting minutes, manual input and editing requires a lot of time and effort, hindering efficient business operations. Furthermore, since there is no way to reflect the emotions expressed during the meeting in the minutes, the minutes themselves may not accurately convey the content. Furthermore, since there is no system for automatically outputting the generated minutes to a specified tool, duplicate input is likely to occur, which could lead to reduced business efficiency.

[0468] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing conference audio in real time and converting it into text, means for analyzing user emotions from the audio data, and means for automatically generating minutes using a generation AI and reflecting emotional information. This makes it possible to efficiently generate minutes of a conference and create accurate minutes that reflect emotional information. In addition, by providing means for automatically outputting the generated minutes to a specific tool, it is possible to prevent double entry and improve work efficiency.

[0469] "Means for capturing meeting audio in real time and converting it into text" refers to technology or equipment used to collect speech during a meeting as audio data and instantly convert that audio data into text information.

[0470] "Means for analyzing user emotions from voice data" refers to a technology or device used to determine the speaker's emotional state in real time based on recorded voice data and extract that emotional information.

[0471] "Means for automatically generating minutes using generative AI and reflecting emotional information" refers to a technology or device that uses artificial intelligence (AI) technology to analyze collected text data and emotional information, automatically create summarized minutes based on that data, and assign emotional tags.

[0472] "Means for allowing users to review and modify the generated minutes" refers to technology or devices that provide an interface or functionality that allows users to review and modify the automatically generated minutes.

[0473] "Means for outputting meeting minutes directly to a designated tool" means the technology or device used to automatically send the generated meeting minutes to a specific collaboration or project management tool and make them immediately available within that tool.

[0474] "Means for receiving and authenticating destination information for a user-specified tool" refers to the technology or device used to receive information about a user-specified tool and perform the appropriate authentication procedures for that tool.

[0475] This invention relates to a system for efficiently generating minutes of a meeting, outputting the minutes directly to a specified tool, and recognizing and reflecting the emotions of users in the minutes. A specific embodiment of this system is described below.

[0476] First, when a user starts a meeting, they access the interface of the minutes generation tool on their terminal and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the terminal to the server.

[0477] The server configures a speech recognition engine (e.g., Google Cloud Speech-to-Text) based on the received meeting information. Next, the device records the meeting audio in real time and streams the audio data to the server. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0478] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time from the recorded audio data. Once the emotion analysis engine analyzes the emotional state, it processes that information along with the text data. The server then passes the converted text data and emotion data to a generation AI (e.g., OpenAI's GPT-3), which analyzes this data to summarize the meeting content and generate minutes with emotion tags.

[0479] An example of a prompt sentence to input to the generative AI model is as follows:

[0480] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[0481] The minutes with emotion tags are formatted into a template on the server and temporarily saved. The server then notifies the user of the URL where the minutes are saved.

[0482] The user accesses this URL from their device, checks the generated minutes, and manually edits emotion tags and content as necessary. Once edits are complete, the edited data is sent from the device to the server, and the minutes are updated.

[0483] The user then confirms and sets the destination tool. The server sends an authentication request to the destination tool's API using the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the specified tool via an HTTP POST request. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted into a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0484] Finally, when the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the meeting content with other tools. Furthermore, because the system executes all processes in real time, manual input work after the meeting is over is no longer necessary.

[0485] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0486] Step 1:

[0487] To start a meeting, a user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output destination tool. This becomes input data and is sent from the device to the server. The server receives this data and performs initial settings to start the meeting session.

[0488] Step 2:

[0489] Based on the conference information received by the server, the server configures and starts a speech recognition engine (e.g., Google Cloud Speech-to-Text). At this time, the server allocates the necessary resources and prepares to capture voice data. The input data is the conference information, and the output is the running status of the speech recognition engine.

[0490] Step 3:

[0491] The device records the audio of the meeting in real time and streams the audio data to the server. This process temporarily stores the audio captured from the device's microphone in a buffer and sends it to the server using a real-time communication protocol (e.g., WebSocket). The input is audio data, and the output is real-time streaming to the server.

[0492] Step 4:

[0493] The server passes the received voice data to a voice recognition engine and converts it into text data in real time. The voice recognition engine analyzes the voice and breaks down the text for each speaker. The input is voice data and the output is text data. At this time, the server identifies each speaker and organizes the content of their speech.

[0494] Step 5:

[0495] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time based on the recorded voice data. The server then configures the system to process the emotional analysis results together with the text data. The input is the voice data, and the output is the emotion analysis results.

[0496] Step 6:

[0497] The server passes the converted text data and emotion data to a generative AI (e.g., OpenAI's GPT-3), which summarizes the meeting content and generates minutes with emotion tags. The input is text data and emotion data, and the output is a summarized minutes. An example of a prompt sentence to input to the generative AI model is as follows:

[0498] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[0499] Step 7:

[0500] The generated minutes are formatted in a template format on the server and temporarily saved. The server notifies the user of the URL where the generated minutes are temporarily saved. The input is the minutes data, and the output is the URL of the temporarily saved file.

[0501] Step 8:

[0502] The user accesses the URL notified from the device and checks the generated minutes. If necessary, they manually correct the emotion tags and content. Once the corrections are complete, the corrected data is sent from the device to the server and the minutes are updated. The input is the user's corrected data, and the output is the updated minutes.

[0503] Step 9:

[0504] The user checks and sets the destination tool. The server sends an authentication request to the destination tool's API based on the authentication information received from the user. The input is the authentication information, and the output is the authentication status.

[0505] Step 10:

[0506] If authentication is successful, the server uses an HTTP POST request to send the minutes to the specified tool. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel. The input is the minutes data, and the output is the completion of upload to the tool.

[0507] Step 11:

[0508] When the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the contents of the meeting with other tools. The input is the output status to the tool, and the output is the completion notification to the user.

[0509] (Application example 2)

[0510] 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."

[0511] There is a need for a system that can efficiently generate meeting minutes and accurately grasp key points of discussions and changes in participants' emotions by reflecting their emotional states in real time. Furthermore, in certain on-site environments, there is a need for a method to improve work efficiency by quickly notifying managers of the meeting content and emotion analysis results. This invention was developed to solve these problems.

[0512] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0513] In this invention, the server includes means for directly outputting the generated minutes to a specified tool, means for capturing meeting audio in real time and converting it into text, means for automatically generating minutes in real time using a generation AI, means for allowing a user to confirm and correct the generated minutes, means for receiving and authenticating output destination information for the specified tool from the user, means for a factory floor worker to start a meeting using smart glasses and record audio in real time and convert it into text, means for analyzing and recording emotional states, and means for remotely notifying a manager of the automatically generated minutes and the emotional analysis results, thereby enabling the real-time generation and output of meeting minutes and emotional states.

[0514] Minutes are documents that record the contents of a meeting, including what was said, decisions made, and summaries of the agenda.

[0515] "Designated tool" refers to an external tool, such as a collaboration tool or project management tool, selected by the user to send the minutes.

[0516] A "speech recognition engine" is a software or hardware technology for converting speech into text in real time.

[0517] "Generative AI" refers to artificial intelligence technology that automatically generates meeting minutes from given data.

[0518] "Emotional state" refers to the changes in the user's emotions and state that are analyzed from the voice data.

[0519] "Smart glasses" are wearable devices that can record conference audio in real time and display information.

[0520] "Destination information" is information about the tool or location designated by the user to send the minutes.

[0521] "API" stands for Application Programming Interface and refers to the methods and tools that allow different software components to communicate with each other.

[0522] "Emotion analysis results" are information on emotional states obtained by analyzing voice data.

[0523] A "manager" is a person in charge of managing the activities of field workers and the contents of meetings within a factory or organization.

[0524] In implementing the invention, the system mainly uses a server, smart glasses, and an administrator's terminal. The detailed process of each step is described below.

[0525] First, a field worker starts a conference using the smart glasses. The smart glasses are equipped with a microphone for recording audio in real time. When the user clicks the "Start a new conference" button on the smart glasses to start the conference, conference information is sent to the server. This information includes the conference title, participant list, and output destination tool.

[0526] Based on the received information, the server activates a speech recognition engine (e.g., Google Speech-to-Text API). Then, the smart glasses stream the recorded audio data of the meeting to the server in real time. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio for each speaker and generates specific text data.

[0527] At the same time, the server activates an emotion engine (e.g., Hugging Face emotion analysis model) to analyze the user's emotional state from the voice data in real time. Once the emotion engine analyzes the emotional state, that information is processed together with the text data.

[0528] The generated text data and emotion data are passed to a generation AI (for example, OpenAI's GPT-3) on the server. The generation AI analyzes this data and summarizes the meeting content. At the same time, minutes with emotion tags are generated. These minutes are temporarily stored on the server.

[0529] Users can view the generated minutes through the smart glasses interface and manually edit emotion tags and content as needed, or administrators can view and edit the minutes from their own devices.

[0530] Finally, the minutes that have been checked and corrected are automatically sent to the destination tool (for example, a project management tool or collaboration tool). The server achieves this output by using the destination tool's API based on the authentication information received from the user and sending the minutes in the specified format.

[0531] For example, during a safety meeting at a worksite, smart glasses record workers' speech and perform text and sentiment analysis. This information is displayed in real time on a manager's dashboard, allowing them to take the necessary action quickly. An example of a prompt for the generative AI model is as follows:

[0532] Example prompt sentence:

[0533] Transcribe an audio recording of a factory floor meeting and summarize it as follows:

[0534] 1. Main agenda of the meeting.

[0535] 2. Important statements by each speaker.

[0536] 3. Analysis of emotional state.

[0537] Please include the meeting title, participant list, and destination tool.

[0538] By implementing the invention in this way, it is possible to generate and output meeting minutes and emotional states in real time, thereby improving meeting efficiency and enabling quick information sharing.

[0539] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0540] Step 1:

[0541] The user uses the smart glasses to click the "Start a new meeting" button. At this time, the meeting title, participant list, and output tool are entered, and this is sent from the device to the server. The entered data is temporarily saved on the server as meeting information.

[0542] Step 2:

[0543] Based on the received meeting information, the server launches a speech recognition engine (e.g., Google Speech-to-Text API). The audio data of the meeting recorded by the smart glasses is streamed to the server in real time. The audio data is received as input, passed to the speech recognition engine, and converted into text data. This process outputs the meeting audio as a text-formatted transcript.

[0544] Step 3:

[0545] The server runs an emotion engine (e.g., Hugging Face emotion analysis model) and processes the voice data separately for emotion analysis. The input voice data is analyzed and the emotional state is digitized. The emotion data is generated in parallel with the text data, and these data are later integrated.

[0546] Step 4:

[0547] The server passes the text data generated by the speech recognition engine and the emotion data generated by the emotion engine to a generation AI (for example, OpenAI's GPT-3). The generation AI receives these data as input and analyzes and summarizes them. This process generates a summary of the meeting minutes with emotion tags added, which are then temporarily saved. The generation of these minutes completes the integration of the data summary and emotion information.

[0548] Step 5:

[0549] The user can review the summary minutes generated through the smart glasses interface and make any necessary corrections. The correction information entered through the smart glasses is then sent to the server, and the minutes are updated. After user confirmation, the final minutes are confirmed.

[0550] Step 6:

[0551] The server sends the final minutes confirmed by the user to the specified output tool (for example, a project management tool or collaboration tool). To send the minutes, the server uses the API of each tool and performs appropriate authentication based on the user's authentication information. This allows the minutes to be uploaded to the output tool in the appropriate format.

[0552] Step 7:

[0553] Once the output process is complete, the server sends a notification to the user. The notification is displayed on the smart glasses or the administrator's device, allowing the user to confirm that the meeting content has been output correctly. At this step, all processing is complete.

[0554] 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.

[0555] 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.

[0556] 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.

[0557] [Third embodiment]

[0558] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0559] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0560] 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).

[0561] 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.

[0562] 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.

[0563] 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).

[0564] 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.

[0565] 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.

[0566] 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.

[0567] 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.

[0568] 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.

[0569] 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."

[0570] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to a designated tool. This system operates as follows.

[0571] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0572] The server activates a speech recognition engine based on the received conference information. The terminal then records the conference audio in real time and streams the audio data to the server. The server then receives the audio data in real time and passes it to the speech recognition engine.

[0573] A speech recognition engine on the server analyzes the recorded audio data and converts it into text in real time. The converted text data is temporarily stored on the server and broken down by speaker. The server then calls a generation AI to summarize the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are formatted into a template on the server.

[0574] The generated minutes are provided to the user as a function to check and correct them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually corrects them as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0575] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted in a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0576] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0577] A specific example is given below.

[0578] When the user clicks the "Start a new meeting" button on the device, the server activates a speech recognition engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination tool. Through this process, users can efficiently output meeting minutes to other tools, improving work efficiency.

[0579] The processing flow will be explained below.

[0580] Step 1:

[0581] A user accesses the minutes generation tool on a terminal and clicks the "Start a new meeting" button, which prompts the user to enter the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0582] Step 2:

[0583] The server configures the voice recognition engine based on the received conference information. The device then records the conference audio in real time and streams the audio data to the server.

[0584] Step 3:

[0585] The server receives the recorded voice data in real time, passes it to a speech recognition engine, and converts it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0586] Step 4:

[0587] The server passes the converted text data to the generation AI, which analyzes the text data, summarizes the meeting content, and generates minutes. The generated minutes are formatted in a template format and temporarily stored on the server.

[0588] Step 5:

[0589] The server notifies the user of the generated minutes as a URL for confirmation and correction. The user accesses this URL from their terminal to check the minutes and make corrections as necessary.

[0590] Step 6:

[0591] The user approves the changes and sends them to the server via the terminal, and the server updates the minutes with the changes received.

[0592] Step 7:

[0593] The user checks and sets the output destination tool, enters the authentication information for the output destination tool, and the device sends this to the server. The server calls the API of the output destination tool based on the authentication information and sends an authentication request.

[0594] Step 8:

[0595] If authentication is successful, the server sends the minutes to the destination tool in the specified format, for example, uploading the content to a specific section in a project management tool, or posting it to a specific channel in a messaging tool.

[0596] Step 9:

[0597] When the output is complete, the server sends a completion notification to the user, who can then check from their terminal that the output was successful.

[0598] Example 1

[0599] 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."

[0600] In today's business environment, many meetings are held, making it extremely important to efficiently generate meeting minutes. However, traditional methods require manual note-taking during meetings and then manual organization and recording of the content afterward. This takes time and effort, and can lead to reduced accuracy. Furthermore, sharing the recorded minutes with appropriate software or tools requires additional manual effort. This reduces work efficiency and increases the risk of important content being overlooked.

[0601] 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.

[0602] In this invention, the server includes: means for directly outputting the generated record to designated software; means for capturing meeting audio in real time and converting it to text; means for automatically generating minutes in real time using a generation AI; means for allowing a user to review and modify the generated minutes; means for receiving and authenticating output destination information for the designated software from the user; means for inputting meeting start information via a user interface; means for streaming audio data in real time; means for analyzing audio and segmenting the text data for each speaker; and means for formatting the summarized minutes into a template format. This allows minutes to be automatically generated from audio recorded during a meeting and instantly shared with designated software. This significantly reduces time and effort and improves the accuracy and efficiency of minutes.

[0603] A "recording" is a record of what is said or done at a meeting or other event, in text, audio, or other format.

[0604] "Software" means programs or applications that run on a digital device and provide specific functions or services.

[0605] A "voice recognition engine" refers to the algorithms and technology used to convert voice data into text data.

[0606] "Generative AI" is an artificial intelligence technology that learns patterns from large amounts of data and generates data such as text.

[0607] "User" refers to any individual or organization that uses the system or software.

[0608] A "user interface" refers to the screens and methods of operation that allow a user to interact with a computer system or software.

[0609] "Streaming" is a technology for transmitting and receiving data such as audio and video in real time.

[0610] "Analysis" is a technical process that refers to examining data to extract meaning.

[0611] A "template format" is a model of a document or data created according to a certain format.

[0612] "Authentication" is the process of verifying that a user or system is legitimate.

[0613] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to designated software. This system operates as follows.

[0614] First, when a user starts a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output software. This information is sent from the device to the server.

[0615] The server activates a voice recognition engine based on the received conference information. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the voice recognition engine. The voice recognition engine analyzes the audio data and converts it into text.

[0616] The converted text data is temporarily stored on the server and broken down by speaker. The server then calls the generation AI, which summarizes the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are then formatted into a template on the server.

[0617] The server then provides the user with the ability to check and modify the generated minutes. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually modifies them as necessary. The modifications are sent from the device to the server, and the minutes are updated.

[0618] Once the user has confirmed and set the destination software, the server sends an authentication request to the destination software's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination software in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0619] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other software, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0620] Specific examples are shown below.

[0621] For example, when a user clicks the "Start a new meeting" button on their device, the server activates a speech recognition engine and sets up a meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination software. Through this process, users can efficiently output meeting minutes to other software, improving work efficiency.

[0622] Example prompt sentence:

[0623] "The title of this meeting is 'Annual Planning Meeting' and the participants are A, B, and C. Specify the project management tool as the destination for the meeting minutes. When the meeting starts, record the audio, send it to the server, and generate and summarize the minutes in real time."

[0624] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0625] Step 1: User starts a conference

[0626] The user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output software. The input data is sent from the device to the server. The input data is basic information about the meeting, and the output is the transfer of information to the server.

[0627] Step 2: The server starts the speech recognition engine

[0628] The server starts the speech recognition engine based on the received conference information. Specifically, it sets up the conference using the title and participant list information. The input is the conference information, and the output is the initialized speech recognition engine.

[0629] Step 3: The device records the audio and streams it to the server

[0630] The terminal records the audio of the conference in real time. The recorded audio data is sent to the server via a streaming protocol (e.g., WebRTC, RTMP). The input is the recorded audio, and the output is the audio data transferred to the server.

[0631] Step 4: The server receives and analyzes the audio data.

[0632] The server receives voice data sent from the device in real time. The received voice data is passed to a voice recognition engine and converted into text data. The input is voice data, and the output is analyzed text data.

[0633] Step 5: The server passes the text data to the generation AI

[0634] The server temporarily stores the converted text data and breaks it down by speaker. This organizes it so that the content of each individual speech is clear. The organized text data is then passed to a generation AI, which instructs it to generate summarized minutes in real time. The input is organized text data, and the output is summarized minutes.

[0635] Step 6: The server formats the minutes

[0636] The server formats the minutes generated by the generation AI into a template format. Specifically, it arranges the minutes content based on a predetermined template and arranges it in an easy-to-read format. The input is a summarized minutes, and the output is the formatted minutes.

[0637] Step 7: The server notifies the user of the URL to check the minutes.

[0638] The server notifies the user of the URL where the formatted minutes are temporarily saved. The user can then access the minutes through the URL. The input is the formatted minutes, and the output is the notification URL.

[0639] Step 8: User reviews and edits the minutes

[0640] The user accesses the URL notified from the terminal and checks the generated minutes. If necessary, they manually correct the contents and send the corrections to the server. The input is correction instructions, and the output is the corrected minutes.

[0641] Step 9: The server reflects the changes

[0642] The server reflects the corrections received from the user in the minutes and updates them as the final version. The input is the corrected data, and the output is the final version of the minutes.

[0643] Step 10: The server sends the minutes to the destination

[0644] After the user confirms and sets the destination software, the server sends an authentication request to the destination software's API based on the authentication information. If authentication is successful, the server sends the minutes to the destination software in the specified format. The input is the authentication information and the final version of the minutes, and the output is the sending of the minutes to the destination software.

[0645] Step 11: Server sends notification of completion

[0646] Finally, the server notifies the user that the output is complete, allowing the user to quickly and efficiently share the meeting contents with other software. The input is the output status, and the output is the completion notification.

[0647] (Application example 1)

[0648] 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."

[0649] Quickly and accurately recording meetings and work reports held by managers and engineers in factories and efficiently sharing them with other management systems and tools is an important factor in significantly improving work efficiency. However, conventional methods require a lot of time and effort to convert speech to text and create summary minutes, which results in delays in reporting and sharing information after meetings.

[0650] 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.

[0651] In this invention, the server includes a means for directly outputting the generated information summary to a specified system, a means for capturing voice data in real time and converting it into text, and a means for automatically generating information summaries in real time using a generation AI. This allows for quick and efficient in-factory meetings and work reports, and by generating summaries in real time and checking and correcting them as necessary, it becomes possible to smoothly share information with management systems and tools.

[0652] A "generated information summary" is a summary text generated in real time from audio data using generation AI.

[0653] The term "designated system" refers to all systems that are linked as the output destination of information designated by the operator.

[0654] "Means" refers to elements or mechanisms that are installed to achieve a specific function in an invention.

[0655] "Audio data" refers to digitally recorded data of audio during meetings or work reports.

[0656] "Real-time capture" refers to the immediate recording and processing of audio data.

[0657] "Converting to text" refers to converting voice data into character string information using voice recognition technology.

[0658] "Generative AI" refers to algorithms or systems that use artificial intelligence techniques to process data and generate a specific output, in this case a summary.

[0659] "Automatically generating information summaries" refers to the process in which generative AI analyzes input data and automatically generates summary sentences.

[0660] "Operator" refers to the person who uses the system to hold meetings and report on work.

[0661] "Display device" refers to equipment for visually displaying information, such as a head-mounted display or smart glasses.

[0662] "Verify and correct" refers to verifying the content of the generated information summary and making corrections or changes as necessary.

[0663] "Speech recognition technology" refers to all technologies for analyzing human speech and converting it into text information.

[0664] "API" refers to an interface for exchanging functions and data between different software programs.

[0665] The system for realizing this invention generates and manages information summaries using advanced technologies such as a voice recognition engine and generation AI. The program processing in this system will be described in detail below.

[0666] This system is mainly composed of three elements: a server, a terminal, and an operator. The server processes the audio data and creates a summary using a generation AI, while the operator on the terminal checks and corrects the summary.

[0667] System flow:

[0668] start

[0669] 1. Starting a meeting and entering information

[0670] When an operator starts a meeting, he or she accesses the minutes generation tool interface on the terminal and clicks the "Start a new meeting" button. At this point, the operator enters the meeting title, participant list, and output destination system (e.g., a project management system). This information is sent from the terminal to the server.

[0671] 2. Audio capture and text conversion

[0672] The server starts a speech recognition engine (e.g., Google Speech Recognition API) based on the received conference information. The devices record the conference audio in real time and stream the audio data to the server. The server receives the audio data in real time and converts it into text using the speech recognition engine.

[0673] 3. Summary generation using generative AI

[0674] A generative AI (e.g., OpenAI's GPT model) on the server analyzes the text data to summarize it. It is given a prompt like this:

[0675] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0676] The generation AI logically organizes the meeting content based on text data and generates summarized information.

[0677] 4. Review and correct the generated information summary

[0678] The generated information summary is formatted in a template format on the server, and the operator can review and manually modify the summary via a terminal.

[0679] 5. Information output and sharing

[0680] When the operator confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the operator. If authentication is successful, the server sends the summarized information to the destination system in the specified format. This allows the summarized information to be automatically uploaded to, for example, a project management system using an API.

[0681] The system allows users to view and modify the generated information summary using a display device (e.g., head-mounted display or smart glasses), supporting efficient business operations within the factory.

[0682] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0683] Step 1:

[0684] Starting a meeting and entering information

[0685] When a user wants to start a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. The device sends the meeting title, participant list, and destination system (e.g., a project management system) as input to the server. The server receives this information and performs initial setup for the meeting.

[0686] Step 2:

[0687] Audio capture and text conversion

[0688] The device records the audio of the meeting in real time and streams the audio data to the server. The server receives the audio data and activates a speech recognition engine (e.g., Google Speech Recognition API) to convert the audio data into text in real time. The input is the audio data streamed in real time, and the output is character string data converted from audio to text.

[0689] Step 3:

[0690] Summary generation using generative AI

[0691] The server uses a generative AI (e.g., OpenAI's GPT model) to analyze the text data obtained in step 2 and generate a summary. By providing the generative AI with prompts such as the following, it can obtain an appropriate summary:

[0692] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0693] The input is text data obtained from a speech recognition engine, and the output is a summary generated by a generative AI.

[0694] Step 4:

[0695] Review and correct the generated information summary

[0696] The server formats the generated information summary into a template and notifies the user of the URL where it is temporarily saved. The user can access the URL from their device to check the generated information summary and manually edit it if necessary. The input is the summary generated by the generation AI, and the output is the final summary after the user has checked and edited it.

[0697] Step 5:

[0698] Output and sharing of information

[0699] When the user confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the user. If authentication is successful, the server sends the summarized information to the destination system in the specified format. For example, the summary information can be automatically uploaded to a project management system using an API. The input is the authentication information from the user and the final summary statement, and the output is the completion of sending the summary information to the destination system.

[0700] 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.

[0701] The present invention relates to a system that can efficiently generate minutes of a meeting, output them directly to a specified tool, and recognize the emotions of users and reflect them in the minutes. This system operates as follows.

[0702] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0703] Based on the conference information received by the server, the server sets up the speech recognition engine. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0704] At the same time, the server also starts an emotion engine, which analyzes the user's emotional state in real time from the recorded voice data. Once the emotion engine analyzes the emotional state, that information is set to be processed together with the text data. The server passes the converted text data and emotion data to the generation AI, which analyzes this data to summarize the meeting content and add emotion tags to the generated minutes. The minutes with the emotion tags added are formatted into a template on the server and temporarily saved.

[0705] The generated minutes are provided to the user as a function for reviewing and correcting them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, reviews the generated minutes, and manually corrects emotion tags and content as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0706] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, the server sends an HTTP POST request to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the server pastes the meeting minutes into a specific section in the board. In the case of a messaging tool, the server posts the meeting minutes to a specific channel.

[0707] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0708] As a concrete example, consider the following scenario.

[0709] When the user clicks the "Start a new meeting" button on the device, the server activates the speech recognition engine and emotion engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio, converts it into text, and passes it to the generation AI along with emotion data analyzed by the emotion engine. The generation AI processes this data to generate summarized minutes, which are then saved on the server with emotion tags. The user can review and edit the minutes and set the output tool, and the server automatically outputs the minutes to the appropriate tool. Through this process, users can efficiently create meeting minutes and output them to other tools.

[0710] The processing flow will be explained below.

[0711] Step 1:

[0712] The user accesses the minutes generation tool on the terminal and clicks the "Start a new meeting" button. The user enters the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0713] Step 2:

[0714] Based on the received conference information, the server configures the startup settings for the speech recognition engine and emotion engine. The server prepares the speech recognition module and emotion recognition module.

[0715] Step 3:

[0716] The device records the audio of the meeting in real time and streams the audio data to the server, where it is immediately transferred.

[0717] Step 4:

[0718] The server passes the received voice data to a speech recognition engine to convert it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0719] Step 5:

[0720] At the same time, the server uses an emotion engine to analyze the speaker's emotional state from the audio data in real time, and the emotion engine analyzes the emotional state and adds it to the text data as an appropriate tag.

[0721] Step 6:

[0722] The server passes the converted text data and emotion data to the generation AI, which analyzes this data and generates minutes summarizing the meeting content. The minutes with added emotion tags are formatted into a template on the server and temporarily saved.

[0723] Step 7:

[0724] The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device and checks the generated minutes. The user manually edits the emotion tags and content as necessary, and sends the edits from their device to the server.

[0725] Step 8:

[0726] The server updates the minutes with the received corrections. The user checks and sets the output destination tool for the meeting, enters the authentication information for the output destination tool, and the terminal sends this to the server.

[0727] Step 9:

[0728] The server calls the destination tool's API based on the authentication information received from the user and sends an authentication request. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted in a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0729] Step 10:

[0730] When the output is complete, the server sends a completion notification to the user. The user can then confirm that the output was performed properly from their terminal and that the minutes of the meeting were properly recorded in the specified tool.

[0731] Example 2

[0732] 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."

[0733] When creating meeting minutes, manual input and editing requires a lot of time and effort, hindering efficient business operations. Furthermore, since there is no way to reflect the emotions expressed during the meeting in the minutes, the minutes themselves may not accurately convey the content. Furthermore, since there is no system for automatically outputting the generated minutes to a specified tool, duplicate input is likely to occur, which could lead to reduced business efficiency.

[0734] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing conference audio in real time and converting it into text, means for analyzing user emotions from the audio data, and means for automatically generating minutes using a generation AI and reflecting emotional information. This makes it possible to efficiently generate minutes of a conference and create accurate minutes that reflect emotional information. In addition, by providing means for automatically outputting the generated minutes to a specific tool, it is possible to prevent double entry and improve work efficiency.

[0735] "Means for capturing meeting audio in real time and converting it into text" refers to technology or equipment used to collect speech during a meeting as audio data and instantly convert that audio data into text information.

[0736] "Means for analyzing user emotions from voice data" refers to a technology or device used to determine the speaker's emotional state in real time based on recorded voice data and extract that emotional information.

[0737] "Means for automatically generating minutes using generative AI and reflecting emotional information" refers to a technology or device that uses artificial intelligence (AI) technology to analyze collected text data and emotional information, automatically create summarized minutes based on that data, and assign emotional tags.

[0738] "Means for allowing users to review and modify the generated minutes" refers to technology or devices that provide an interface or functionality that allows users to review and modify the automatically generated minutes.

[0739] "Means for outputting meeting minutes directly to a designated tool" means the technology or device used to automatically send the generated meeting minutes to a specific collaboration or project management tool and make them immediately available within that tool.

[0740] "Means for receiving and authenticating destination information for a user-specified tool" refers to the technology or device used to receive information about a user-specified tool and perform the appropriate authentication procedures for that tool.

[0741] This invention relates to a system for efficiently generating minutes of a meeting, outputting the minutes directly to a specified tool, and recognizing and reflecting the emotions of users in the minutes. A specific embodiment of this system is described below.

[0742] First, when a user starts a meeting, they access the interface of the minutes generation tool on their terminal and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the terminal to the server.

[0743] The server configures a speech recognition engine (e.g., Google Cloud Speech-to-Text) based on the received meeting information. Next, the device records the meeting audio in real time and streams the audio data to the server. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0744] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time from the recorded audio data. Once the emotion analysis engine analyzes the emotional state, it processes that information along with the text data. The server then passes the converted text data and emotion data to a generation AI (e.g., OpenAI's GPT-3), which analyzes this data to summarize the meeting content and generate minutes with emotion tags.

[0745] An example of a prompt sentence to input to the generative AI model is as follows:

[0746] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[0747] The minutes with emotion tags are formatted into a template on the server and temporarily saved. The server then notifies the user of the URL where the minutes are saved.

[0748] The user accesses this URL from their device, checks the generated minutes, and manually edits emotion tags and content as necessary. Once edits are complete, the edited data is sent from the device to the server, and the minutes are updated.

[0749] The user then confirms and sets the destination tool. The server sends an authentication request to the destination tool's API using the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the specified tool via an HTTP POST request. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted into a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0750] Finally, when the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the meeting content with other tools. Furthermore, because the system executes all processes in real time, manual input work after the meeting is over is no longer necessary.

[0751] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0752] Step 1:

[0753] To start a meeting, a user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output destination tool. This becomes input data and is sent from the device to the server. The server receives this data and performs initial settings to start the meeting session.

[0754] Step 2:

[0755] Based on the conference information received by the server, the server configures and starts a speech recognition engine (e.g., Google Cloud Speech-to-Text). At this time, the server allocates the necessary resources and prepares to capture voice data. The input data is the conference information, and the output is the running status of the speech recognition engine.

[0756] Step 3:

[0757] The device records the audio of the meeting in real time and streams the audio data to the server. This process temporarily stores the audio captured from the device's microphone in a buffer and sends it to the server using a real-time communication protocol (e.g., WebSocket). The input is audio data, and the output is real-time streaming to the server.

[0758] Step 4:

[0759] The server passes the received voice data to a voice recognition engine and converts it into text data in real time. The voice recognition engine analyzes the voice and breaks down the text for each speaker. The input is voice data and the output is text data. At this time, the server identifies each speaker and organizes the content of their speech.

[0760] Step 5:

[0761] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time based on the recorded voice data. The server then configures the system to process the emotional analysis results together with the text data. The input is the voice data, and the output is the emotion analysis results.

[0762] Step 6:

[0763] The server passes the converted text data and emotion data to a generative AI (e.g., OpenAI's GPT-3), which summarizes the meeting content and generates minutes with emotion tags. The input is text data and emotion data, and the output is a summarized minutes. An example of a prompt sentence to input to the generative AI model is as follows:

[0764] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[0765] Step 7:

[0766] The generated minutes are formatted in a template format on the server and temporarily saved. The server notifies the user of the URL where the generated minutes are temporarily saved. The input is the minutes data, and the output is the URL of the temporarily saved file.

[0767] Step 8:

[0768] The user accesses the URL notified from the device and checks the generated minutes. If necessary, they manually correct the emotion tags and content. Once the corrections are complete, the corrected data is sent from the device to the server and the minutes are updated. The input is the user's corrected data, and the output is the updated minutes.

[0769] Step 9:

[0770] The user checks and sets the destination tool. The server sends an authentication request to the destination tool's API based on the authentication information received from the user. The input is the authentication information, and the output is the authentication status.

[0771] Step 10:

[0772] If authentication is successful, the server uses an HTTP POST request to send the minutes to the specified tool. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel. The input is the minutes data, and the output is the completion of upload to the tool.

[0773] Step 11:

[0774] When the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the contents of the meeting with other tools. The input is the output status to the tool, and the output is the completion notification to the user.

[0775] (Application example 2)

[0776] 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."

[0777] There is a need for a system that can efficiently generate meeting minutes and accurately grasp key points of discussions and changes in participants' emotions by reflecting their emotional states in real time. Furthermore, in certain on-site environments, there is a need for a method to improve work efficiency by quickly notifying managers of the meeting content and emotion analysis results. This invention was developed to solve these problems.

[0778] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0779] In this invention, the server includes means for directly outputting the generated minutes to a specified tool, means for capturing meeting audio in real time and converting it into text, means for automatically generating minutes in real time using a generation AI, means for allowing a user to confirm and correct the generated minutes, means for receiving and authenticating output destination information for the specified tool from the user, means for a factory floor worker to start a meeting using smart glasses and record audio in real time and convert it into text, means for analyzing and recording emotional states, and means for remotely notifying a manager of the automatically generated minutes and the emotional analysis results, thereby enabling the real-time generation and output of meeting minutes and emotional states.

[0780] Minutes are documents that record the contents of a meeting, including what was said, decisions made, and summaries of the agenda.

[0781] "Designated tool" refers to an external tool, such as a collaboration tool or project management tool, selected by the user to send the minutes.

[0782] A "speech recognition engine" is a software or hardware technology for converting speech into text in real time.

[0783] "Generative AI" refers to artificial intelligence technology that automatically generates meeting minutes from given data.

[0784] "Emotional state" refers to the changes in the user's emotions and state that are analyzed from the voice data.

[0785] "Smart glasses" are wearable devices that can record conference audio in real time and display information.

[0786] "Destination information" is information about the tool or location designated by the user to send the minutes.

[0787] "API" stands for Application Programming Interface and refers to the methods and tools that allow different software components to communicate with each other.

[0788] "Emotion analysis results" are information on emotional states obtained by analyzing voice data.

[0789] A "manager" is a person in charge of managing the activities of field workers and the contents of meetings within a factory or organization.

[0790] In implementing the invention, the system mainly uses a server, smart glasses, and an administrator's terminal. The detailed process of each step is described below.

[0791] First, a field worker starts a conference using the smart glasses. The smart glasses are equipped with a microphone for recording audio in real time. When the user clicks the "Start a new conference" button on the smart glasses to start the conference, conference information is sent to the server. This information includes the conference title, participant list, and output destination tool.

[0792] Based on the received information, the server activates a speech recognition engine (e.g., Google Speech-to-Text API). Then, the smart glasses stream the recorded audio data of the meeting to the server in real time. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio for each speaker and generates specific text data.

[0793] At the same time, the server activates an emotion engine (e.g., Hugging Face emotion analysis model) to analyze the user's emotional state from the voice data in real time. Once the emotion engine analyzes the emotional state, that information is processed together with the text data.

[0794] The generated text data and emotion data are passed to a generation AI (for example, OpenAI's GPT-3) on the server. The generation AI analyzes this data and summarizes the meeting content. At the same time, minutes with emotion tags are generated. These minutes are temporarily stored on the server.

[0795] Users can view the generated minutes through the smart glasses interface and manually edit emotion tags and content as needed, or administrators can view and edit the minutes from their own devices.

[0796] Finally, the minutes that have been checked and corrected are automatically sent to the destination tool (for example, a project management tool or collaboration tool). The server achieves this output by using the destination tool's API based on the authentication information received from the user and sending the minutes in the specified format.

[0797] For example, during a safety meeting at a worksite, smart glasses record workers' speech and perform text and sentiment analysis. This information is displayed in real time on a manager's dashboard, allowing them to take the necessary action quickly. An example of a prompt for the generative AI model is as follows:

[0798] Example prompt sentence:

[0799] Transcribe an audio recording of a factory floor meeting and summarize it as follows:

[0800] 1. Main agenda of the meeting.

[0801] 2. Important statements by each speaker.

[0802] 3. Analysis of emotional state.

[0803] Please include the meeting title, participant list, and destination tool.

[0804] By implementing the invention in this way, it is possible to generate and output meeting minutes and emotional states in real time, thereby improving meeting efficiency and enabling quick information sharing.

[0805] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0806] Step 1:

[0807] The user uses the smart glasses to click the "Start a new meeting" button. At this time, the meeting title, participant list, and output tool are entered, and this is sent from the device to the server. The entered data is temporarily saved on the server as meeting information.

[0808] Step 2:

[0809] Based on the received meeting information, the server launches a speech recognition engine (e.g., Google Speech-to-Text API). The audio data of the meeting recorded by the smart glasses is streamed to the server in real time. The audio data is received as input, passed to the speech recognition engine, and converted into text data. This process outputs the meeting audio as a text-formatted transcript.

[0810] Step 3:

[0811] The server runs an emotion engine (e.g., Hugging Face emotion analysis model) and processes the voice data separately for emotion analysis. The input voice data is analyzed and the emotional state is digitized. The emotion data is generated in parallel with the text data, and these data are later integrated.

[0812] Step 4:

[0813] The server passes the text data generated by the speech recognition engine and the emotion data generated by the emotion engine to a generation AI (for example, OpenAI's GPT-3). The generation AI receives these data as input and analyzes and summarizes them. This process generates a summary of the meeting minutes with emotion tags added, which are then temporarily saved. The generation of these minutes completes the integration of the data summary and emotion information.

[0814] Step 5:

[0815] The user can review the summary minutes generated through the smart glasses interface and make any necessary corrections. The correction information entered through the smart glasses is then sent to the server, and the minutes are updated. After user confirmation, the final minutes are confirmed.

[0816] Step 6:

[0817] The server sends the final minutes confirmed by the user to the specified output tool (for example, a project management tool or collaboration tool). To send the minutes, the server uses the API of each tool and performs appropriate authentication based on the user's authentication information. This allows the minutes to be uploaded to the output tool in the appropriate format.

[0818] Step 7:

[0819] Once the output process is complete, the server sends a notification to the user. The notification is displayed on the smart glasses or the administrator's device, allowing the user to confirm that the meeting content has been output correctly. At this step, all processing is complete.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] [Fourth embodiment]

[0824] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0825] 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.

[0826] 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).

[0827] 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.

[0828] 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.

[0829] 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).

[0830] 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.

[0831] 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.

[0832] 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.

[0833] 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.

[0834] 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.

[0835] 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.

[0836] 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."

[0837] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to a designated tool. This system operates as follows.

[0838] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0839] The server activates a speech recognition engine based on the received conference information. The terminal then records the conference audio in real time and streams the audio data to the server. The server then receives the audio data in real time and passes it to the speech recognition engine.

[0840] A speech recognition engine on the server analyzes the recorded audio data and converts it into text in real time. The converted text data is temporarily stored on the server and broken down by speaker. The server then calls a generation AI to summarize the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are formatted into a template on the server.

[0841] The generated minutes are provided to the user as a function to check and correct them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually corrects them as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0842] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted in a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0843] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0844] A specific example is given below.

[0845] When the user clicks the "Start a new meeting" button on the device, the server activates a speech recognition engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination tool. Through this process, users can efficiently output meeting minutes to other tools, improving work efficiency.

[0846] The processing flow will be explained below.

[0847] Step 1:

[0848] A user accesses the minutes generation tool on a terminal and clicks the "Start a new meeting" button, which prompts the user to enter the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0849] Step 2:

[0850] The server configures the voice recognition engine based on the received conference information. The device then records the conference audio in real time and streams the audio data to the server.

[0851] Step 3:

[0852] The server receives the recorded voice data in real time, passes it to a speech recognition engine, and converts it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0853] Step 4:

[0854] The server passes the converted text data to the generation AI, which analyzes the text data, summarizes the meeting content, and generates minutes. The generated minutes are formatted in a template format and temporarily stored on the server.

[0855] Step 5:

[0856] The server notifies the user of the generated minutes as a URL for confirmation and correction. The user accesses this URL from their terminal to check the minutes and make corrections as necessary.

[0857] Step 6:

[0858] The user approves the changes and sends them to the server via the terminal, and the server updates the minutes with the changes received.

[0859] Step 7:

[0860] The user checks and sets the output destination tool, enters the authentication information for the output destination tool, and the device sends this to the server. The server calls the API of the output destination tool based on the authentication information and sends an authentication request.

[0861] Step 8:

[0862] If authentication is successful, the server sends the minutes to the destination tool in the specified format, for example, uploading the content to a specific section in a project management tool, or posting it to a specific channel in a messaging tool.

[0863] Step 9:

[0864] When the output is complete, the server sends a completion notification to the user, who can then check from their terminal that the output was successful.

[0865] Example 1

[0866] 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."

[0867] In today's business environment, many meetings are held, making it extremely important to efficiently generate meeting minutes. However, traditional methods require manual note-taking during meetings and then manual organization and recording of the content afterward. This takes time and effort, and can lead to reduced accuracy. Furthermore, sharing the recorded minutes with appropriate software or tools requires additional manual effort. This reduces work efficiency and increases the risk of important content being overlooked.

[0868] 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.

[0869] In this invention, the server includes: means for directly outputting the generated record to designated software; means for capturing meeting audio in real time and converting it to text; means for automatically generating minutes in real time using a generation AI; means for allowing a user to review and modify the generated minutes; means for receiving and authenticating output destination information for the designated software from the user; means for inputting meeting start information via a user interface; means for streaming audio data in real time; means for analyzing audio and segmenting the text data for each speaker; and means for formatting the summarized minutes into a template format. This allows minutes to be automatically generated from audio recorded during a meeting and instantly shared with designated software. This significantly reduces time and effort and improves the accuracy and efficiency of minutes.

[0870] A "recording" is a record of what is said or done at a meeting or other event, in text, audio, or other format.

[0871] "Software" means programs or applications that run on a digital device and provide specific functions or services.

[0872] A "voice recognition engine" refers to the algorithms and technology used to convert voice data into text data.

[0873] "Generative AI" is an artificial intelligence technology that learns patterns from large amounts of data and generates data such as text.

[0874] "User" refers to any individual or organization that uses the system or software.

[0875] A "user interface" refers to the screens and methods of operation that allow a user to interact with a computer system or software.

[0876] "Streaming" is a technology for transmitting and receiving data such as audio and video in real time.

[0877] "Analysis" is a technical process that refers to examining data to extract meaning.

[0878] A "template format" is a model of a document or data created according to a certain format.

[0879] "Authentication" is the process of verifying that a user or system is legitimate.

[0880] The present invention relates to a system for efficiently generating minutes of a meeting and outputting them directly to designated software. This system operates as follows.

[0881] First, when a user starts a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output software. This information is sent from the device to the server.

[0882] The server activates a voice recognition engine based on the received conference information. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the voice recognition engine. The voice recognition engine analyzes the audio data and converts it into text.

[0883] The converted text data is temporarily stored on the server and broken down by speaker. The server then calls the generation AI, which summarizes the text data from the speech recognition engine. The generation AI analyzes the text data, logically organizes the meeting content, and generates summarized minutes. The generated minutes are then formatted into a template on the server.

[0884] The server then provides the user with the ability to check and modify the generated minutes. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, checks the generated minutes, and manually modifies them as necessary. The modifications are sent from the device to the server, and the minutes are updated.

[0885] Once the user has confirmed and set the destination software, the server sends an authentication request to the destination software's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination software in the specified format. At this stage, for example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel.

[0886] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other software, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0887] Specific examples are shown below.

[0888] For example, when a user clicks the "Start a new meeting" button on their device, the server activates a speech recognition engine and sets up a meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio and converts it into text. The text data is then processed by a generation AI to generate summarized meeting minutes. After the user reviews the minutes and makes any necessary corrections, the server sends them to the destination software. Through this process, users can efficiently output meeting minutes to other software, improving work efficiency.

[0889] Example prompt sentence:

[0890] "The title of this meeting is 'Annual Planning Meeting' and the participants are A, B, and C. Specify the project management tool as the destination for the meeting minutes. When the meeting starts, record the audio, send it to the server, and generate and summarize the minutes in real time."

[0891] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0892] Step 1: User starts a conference

[0893] The user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output software. The input data is sent from the device to the server. The input data is basic information about the meeting, and the output is the transfer of information to the server.

[0894] Step 2: The server starts the speech recognition engine

[0895] The server starts the speech recognition engine based on the received conference information. Specifically, it sets up the conference using the title and participant list information. The input is the conference information, and the output is the initialized speech recognition engine.

[0896] Step 3: The device records the audio and streams it to the server

[0897] The terminal records the audio of the conference in real time. The recorded audio data is sent to the server via a streaming protocol (e.g., WebRTC, RTMP). The input is the recorded audio, and the output is the audio data transferred to the server.

[0898] Step 4: The server receives and analyzes the audio data.

[0899] The server receives voice data sent from the device in real time. The received voice data is passed to a voice recognition engine and converted into text data. The input is voice data, and the output is analyzed text data.

[0900] Step 5: The server passes the text data to the generation AI

[0901] The server temporarily stores the converted text data and breaks it down by speaker. This organizes it so that the content of each individual speech is clear. The organized text data is then passed to a generation AI, which instructs it to generate summarized minutes in real time. The input is organized text data, and the output is summarized minutes.

[0902] Step 6: The server formats the minutes

[0903] The server formats the minutes generated by the generation AI into a template format. Specifically, it arranges the minutes content based on a predetermined template and arranges it in an easy-to-read format. The input is a summarized minutes, and the output is the formatted minutes.

[0904] Step 7: The server notifies the user of the URL to check the minutes.

[0905] The server notifies the user of the URL where the formatted minutes are temporarily saved. The user can then access the minutes through the URL. The input is the formatted minutes, and the output is the notification URL.

[0906] Step 8: User reviews and edits the minutes

[0907] The user accesses the URL notified from the terminal and checks the generated minutes. If necessary, they manually correct the contents and send the corrections to the server. The input is correction instructions, and the output is the corrected minutes.

[0908] Step 9: The server reflects the changes

[0909] The server reflects the corrections received from the user in the minutes and updates them as the final version. The input is the corrected data, and the output is the final version of the minutes.

[0910] Step 10: The server sends the minutes to the destination

[0911] After the user confirms and sets the destination software, the server sends an authentication request to the destination software's API based on the authentication information. If authentication is successful, the server sends the minutes to the destination software in the specified format. The input is the authentication information and the final version of the minutes, and the output is the sending of the minutes to the destination software.

[0912] Step 11: Server sends notification of completion

[0913] Finally, the server notifies the user that the output is complete, allowing the user to quickly and efficiently share the meeting contents with other software. The input is the output status, and the output is the completion notification.

[0914] (Application example 1)

[0915] 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."

[0916] Quickly and accurately recording meetings and work reports held by managers and engineers in factories and efficiently sharing them with other management systems and tools is an important factor in significantly improving work efficiency. However, conventional methods require a lot of time and effort to convert speech to text and create summary minutes, which results in delays in reporting and sharing information after meetings.

[0917] 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.

[0918] In this invention, the server includes a means for directly outputting the generated information summary to a specified system, a means for capturing voice data in real time and converting it into text, and a means for automatically generating information summaries in real time using a generation AI. This allows for quick and efficient in-factory meetings and work reports, and by generating summaries in real time and checking and correcting them as necessary, it becomes possible to smoothly share information with management systems and tools.

[0919] A "generated information summary" is a summary text generated in real time from audio data using generation AI.

[0920] The term "designated system" refers to all systems that are linked as the output destination of information designated by the operator.

[0921] "Means" refers to elements or mechanisms that are installed to achieve a specific function in an invention.

[0922] "Audio data" refers to digitally recorded data of audio during meetings or work reports.

[0923] "Real-time capture" refers to the immediate recording and processing of audio data.

[0924] "Converting to text" refers to converting voice data into character string information using voice recognition technology.

[0925] "Generative AI" refers to algorithms or systems that use artificial intelligence techniques to process data and generate a specific output, in this case a summary.

[0926] "Automatically generating information summaries" refers to the process in which generative AI analyzes input data and automatically generates summary sentences.

[0927] "Operator" refers to the person who uses the system to hold meetings and report on work.

[0928] "Display device" refers to equipment for visually displaying information, such as a head-mounted display or smart glasses.

[0929] "Verify and correct" refers to verifying the content of the generated information summary and making corrections or changes as necessary.

[0930] "Speech recognition technology" refers to all technologies for analyzing human speech and converting it into text information.

[0931] "API" refers to an interface for exchanging functions and data between different software programs.

[0932] The system for realizing this invention generates and manages information summaries using advanced technologies such as a voice recognition engine and generation AI. The program processing in this system will be described in detail below.

[0933] This system is mainly composed of three elements: a server, a terminal, and an operator. The server processes the audio data and creates a summary using a generation AI, while the operator on the terminal checks and corrects the summary.

[0934] System flow:

[0935] start

[0936] 1. Starting a meeting and entering information

[0937] When an operator starts a meeting, he or she accesses the minutes generation tool interface on the terminal and clicks the "Start a new meeting" button. At this point, the operator enters the meeting title, participant list, and output destination system (e.g., a project management system). This information is sent from the terminal to the server.

[0938] 2. Audio capture and text conversion

[0939] The server starts a speech recognition engine (e.g., Google Speech Recognition API) based on the received conference information. The devices record the conference audio in real time and stream the audio data to the server. The server receives the audio data in real time and converts it into text using the speech recognition engine.

[0940] 3. Summary generation using generative AI

[0941] A generative AI (e.g., OpenAI's GPT model) on the server analyzes the text data to summarize it. It is given a prompt like this:

[0942] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0943] The generation AI logically organizes the meeting content based on text data and generates summarized information.

[0944] 4. Review and correct the generated information summary

[0945] The generated information summary is formatted in a template format on the server, and the operator can review and manually modify the summary via a terminal.

[0946] 5. Information output and sharing

[0947] When the operator confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the operator. If authentication is successful, the server sends the summarized information to the destination system in the specified format. This allows the summarized information to be automatically uploaded to, for example, a project management system using an API.

[0948] The system allows users to view and modify the generated information summary using a display device (e.g., head-mounted display or smart glasses), supporting efficient business operations within the factory.

[0949] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0950] Step 1:

[0951] Starting a meeting and entering information

[0952] When a user wants to start a meeting, they access the minutes generation tool interface on their device and click the "Start a new meeting" button. The device sends the meeting title, participant list, and destination system (e.g., a project management system) as input to the server. The server receives this information and performs initial setup for the meeting.

[0953] Step 2:

[0954] Audio capture and text conversion

[0955] The device records the audio of the meeting in real time and streams the audio data to the server. The server receives the audio data and activates a speech recognition engine (e.g., Google Speech Recognition API) to convert the audio data into text in real time. The input is the audio data streamed in real time, and the output is character string data converted from audio to text.

[0956] Step 3:

[0957] Summary generation using generative AI

[0958] The server uses a generative AI (e.g., OpenAI's GPT model) to analyze the text data obtained in step 2 and generate a summary. By providing the generative AI with prompts such as the following, it can obtain an appropriate summary:

[0959] "Please summarize the following conversation:\n\nToday we begin our monthly performance report meeting. Manager A will provide a progress report on Agenda Item 1..."

[0960] The input is text data obtained from a speech recognition engine, and the output is a summary generated by a generative AI.

[0961] Step 4:

[0962] Review and correct the generated information summary

[0963] The server formats the generated information summary into a template and notifies the user of the URL where it is temporarily saved. The user can access the URL from their device to check the generated information summary and manually edit it if necessary. The input is the summary generated by the generation AI, and the output is the final summary after the user has checked and edited it.

[0964] Step 5:

[0965] Output and sharing of information

[0966] When the user confirms and sets the destination system for the meeting, the server sends an authentication request to that system based on the authentication information received from the user. If authentication is successful, the server sends the summarized information to the destination system in the specified format. For example, the summary information can be automatically uploaded to a project management system using an API. The input is the authentication information from the user and the final summary statement, and the output is the completion of sending the summary information to the destination system.

[0967] 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.

[0968] The present invention relates to a system that can efficiently generate minutes of a meeting, output them directly to a specified tool, and recognize the emotions of users and reflect them in the minutes. This system operates as follows.

[0969] First, when a user starts a meeting, they access the interface of the minutes generation tool on their device and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the device to the server.

[0970] Based on the conference information received by the server, the server sets up the speech recognition engine. Next, the device records the conference audio in real time and streams the audio data to the server. The server receives the audio data in real time and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[0971] At the same time, the server also starts an emotion engine, which analyzes the user's emotional state in real time from the recorded voice data. Once the emotion engine analyzes the emotional state, that information is set to be processed together with the text data. The server passes the converted text data and emotion data to the generation AI, which analyzes this data to summarize the meeting content and add emotion tags to the generated minutes. The minutes with the emotion tags added are formatted into a template on the server and temporarily saved.

[0972] The generated minutes are provided to the user as a function for reviewing and correcting them. The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device, reviews the generated minutes, and manually corrects emotion tags and content as necessary. The corrections are sent from the device to the server, and the minutes are updated.

[0973] When a user confirms and sets the destination tool for the meeting, the server sends an authentication request to the destination tool's API based on the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, the server sends an HTTP POST request to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the server pastes the meeting minutes into a specific section in the board. In the case of a messaging tool, the server posts the meeting minutes to a specific channel.

[0974] Finally, when the output is complete, the server sends a completion notification to the user. This allows users to quickly and efficiently share the contents of the meeting with other tools, greatly improving work efficiency. In addition, because this system executes all processes in real time, there is no need to copy and paste after the meeting ends.

[0975] As a concrete example, consider the following scenario.

[0976] When the user clicks the "Start a new meeting" button on the device, the server activates the speech recognition engine and emotion engine and sets up the meeting based on the information entered by the user. The device records the meeting audio and streams it to the server, which analyzes the audio, converts it into text, and passes it to the generation AI along with emotion data analyzed by the emotion engine. The generation AI processes this data to generate summarized minutes, which are then saved on the server with emotion tags. The user can review and edit the minutes and set the output tool, and the server automatically outputs the minutes to the appropriate tool. Through this process, users can efficiently create meeting minutes and output them to other tools.

[0977] The processing flow will be explained below.

[0978] Step 1:

[0979] The user accesses the minutes generation tool on the terminal and clicks the "Start a new meeting" button. The user enters the meeting title, participant list, and output destination tool, and the terminal sends this information to the server.

[0980] Step 2:

[0981] Based on the received conference information, the server configures the startup settings for the speech recognition engine and emotion engine. The server prepares the speech recognition module and emotion recognition module.

[0982] Step 3:

[0983] The device records the audio of the meeting in real time and streams the audio data to the server, where it is immediately transferred.

[0984] Step 4:

[0985] The server passes the received voice data to a speech recognition engine to convert it into text data. The speech recognition engine analyzes the voice and breaks down the text for each speaker.

[0986] Step 5:

[0987] At the same time, the server uses an emotion engine to analyze the speaker's emotional state from the audio data in real time, and the emotion engine analyzes the emotional state and adds it to the text data as an appropriate tag.

[0988] Step 6:

[0989] The server passes the converted text data and emotion data to the generation AI, which analyzes this data and generates minutes summarizing the meeting content. The minutes with added emotion tags are formatted into a template on the server and temporarily saved.

[0990] Step 7:

[0991] The server notifies the user of the URL where the generated minutes are temporarily saved. The user accesses this URL from their device and checks the generated minutes. The user manually edits the emotion tags and content as necessary, and sends the edits from their device to the server.

[0992] Step 8:

[0993] The server updates the minutes with the received corrections. The user checks and sets the output destination tool for the meeting, enters the authentication information for the output destination tool, and the terminal sends this to the server.

[0994] Step 9:

[0995] The server calls the destination tool's API based on the authentication information received from the user and sends an authentication request. If authentication is successful, the server sends the meeting minutes to the destination tool in the specified format. For example, in the case of a collaboration tool, an HTTP POST request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted in a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[0996] Step 10:

[0997] When the output is complete, the server sends a completion notification to the user. The user can then confirm that the output was performed properly from their terminal and that the minutes of the meeting were properly recorded in the specified tool.

[0998] Example 2

[0999] 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."

[1000] When creating meeting minutes, manual input and editing requires a lot of time and effort, hindering efficient business operations. Furthermore, since there is no way to reflect the emotions expressed during the meeting in the minutes, the minutes themselves may not accurately convey the content. Furthermore, since there is no system for automatically outputting the generated minutes to a specified tool, duplicate input is likely to occur, which could lead to reduced business efficiency.

[1001] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for capturing conference audio in real time and converting it into text, means for analyzing user emotions from the audio data, and means for automatically generating minutes using a generation AI and reflecting emotional information. This makes it possible to efficiently generate minutes of a conference and create accurate minutes that reflect emotional information. In addition, by providing means for automatically outputting the generated minutes to a specific tool, it is possible to prevent double entry and improve work efficiency.

[1002] "Means for capturing meeting audio in real time and converting it into text" refers to technology or equipment used to collect speech during a meeting as audio data and instantly convert that audio data into text information.

[1003] "Means for analyzing user emotions from voice data" refers to a technology or device used to determine the speaker's emotional state in real time based on recorded voice data and extract that emotional information.

[1004] "Means for automatically generating minutes using generative AI and reflecting emotional information" refers to a technology or device that uses artificial intelligence (AI) technology to analyze collected text data and emotional information, automatically create summarized minutes based on that data, and assign emotional tags.

[1005] "Means for allowing users to review and modify the generated minutes" refers to technology or devices that provide an interface or functionality that allows users to review and modify the automatically generated minutes.

[1006] "Means for outputting meeting minutes directly to a designated tool" means the technology or device used to automatically send the generated meeting minutes to a specific collaboration or project management tool and make them immediately available within that tool.

[1007] "Means for receiving and authenticating destination information for a user-specified tool" refers to the technology or device used to receive information about a user-specified tool and perform the appropriate authentication procedures for that tool.

[1008] This invention relates to a system for efficiently generating minutes of a meeting, outputting the minutes directly to a specified tool, and recognizing and reflecting the emotions of users in the minutes. A specific embodiment of this system is described below.

[1009] First, when a user starts a meeting, they access the interface of the minutes generation tool on their terminal and click the "Start a new meeting" button. At this time, the user enters the meeting title, participant list, and output destination tool (e.g., collaboration tool or project management tool). This information is sent from the terminal to the server.

[1010] The server configures a speech recognition engine (e.g., Google Cloud Speech-to-Text) based on the received meeting information. Next, the device records the meeting audio in real time and streams the audio data to the server. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio and breaks down the text for each speaker.

[1011] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time from the recorded audio data. Once the emotion analysis engine analyzes the emotional state, it processes that information along with the text data. The server then passes the converted text data and emotion data to a generation AI (e.g., OpenAI's GPT-3), which analyzes this data to summarize the meeting content and generate minutes with emotion tags.

[1012] An example of a prompt sentence to input to the generative AI model is as follows:

[1013] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[1014] The minutes with emotion tags are formatted into a template on the server and temporarily saved. The server then notifies the user of the URL where the minutes are saved.

[1015] The user accesses this URL from their device, checks the generated minutes, and manually edits emotion tags and content as necessary. Once edits are complete, the edited data is sent from the device to the server, and the minutes are updated.

[1016] The user then confirms and sets the destination tool. The server sends an authentication request to the destination tool's API using the authentication information received from the user. If authentication is successful, the server sends the meeting minutes to the specified tool via an HTTP POST request. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the meeting minutes data. In the case of a project management tool, the meeting minutes are pasted into a specific section within the board. In the case of a messaging tool, the meeting minutes are posted to a specific channel.

[1017] Finally, when the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the meeting content with other tools. Furthermore, because the system executes all processes in real time, manual input work after the meeting is over is no longer necessary.

[1018] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1019] Step 1:

[1020] To start a meeting, a user accesses the minutes generation tool interface on their device and clicks the "Start a new meeting" button. At this time, the user inputs the meeting title, participant list, and output destination tool. This becomes input data and is sent from the device to the server. The server receives this data and performs initial settings to start the meeting session.

[1021] Step 2:

[1022] Based on the conference information received by the server, the server configures and starts a speech recognition engine (e.g., Google Cloud Speech-to-Text). At this time, the server allocates the necessary resources and prepares to capture voice data. The input data is the conference information, and the output is the running status of the speech recognition engine.

[1023] Step 3:

[1024] The device records the audio of the meeting in real time and streams the audio data to the server. This process temporarily stores the audio captured from the device's microphone in a buffer and sends it to the server using a real-time communication protocol (e.g., WebSocket). The input is audio data, and the output is real-time streaming to the server.

[1025] Step 4:

[1026] The server passes the received voice data to a voice recognition engine and converts it into text data in real time. The voice recognition engine analyzes the voice and breaks down the text for each speaker. The input is voice data and the output is text data. At this time, the server identifies each speaker and organizes the content of their speech.

[1027] Step 5:

[1028] At the same time, the server launches an emotion analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the user's emotional state in real time based on the recorded voice data. The server then configures the system to process the emotional analysis results together with the text data. The input is the voice data, and the output is the emotion analysis results.

[1029] Step 6:

[1030] The server passes the converted text data and emotion data to a generative AI (e.g., OpenAI's GPT-3), which summarizes the meeting content and generates minutes with emotion tags. The input is text data and emotion data, and the output is a summarized minutes. An example of a prompt sentence to input to the generative AI model is as follows:

[1031] Prompt: "Please summarize the following meeting content and sentiment data and create a transcript including sentiment tags. The meeting content is as follows: [Text data of the meeting content] The sentiment data is as follows: [Sentiment data]"

[1032] Step 7:

[1033] The generated minutes are formatted in a template format on the server and temporarily saved. The server notifies the user of the URL where the generated minutes are temporarily saved. The input is the minutes data, and the output is the URL of the temporarily saved file.

[1034] Step 8:

[1035] The user accesses the URL notified from the device and checks the generated minutes. If necessary, they manually correct the emotion tags and content. Once the corrections are complete, the corrected data is sent from the device to the server and the minutes are updated. The input is the user's corrected data, and the output is the updated minutes.

[1036] Step 9:

[1037] The user checks and sets the destination tool. The server sends an authentication request to the destination tool's API based on the authentication information received from the user. The input is the authentication information, and the output is the authentication status.

[1038] Step 10:

[1039] If authentication is successful, the server uses an HTTP POST request to send the minutes to the specified tool. For example, in the case of a collaboration tool, a request is sent to the meeting space and page to upload the minutes data. In the case of a project management tool, the minutes are pasted into a specific section within the board. In the case of a messaging tool, the minutes are posted to a specific channel. The input is the minutes data, and the output is the completion of upload to the tool.

[1040] Step 11:

[1041] When the output is complete, the server sends a completion notification to the user, allowing users to quickly and efficiently share the contents of the meeting with other tools. The input is the output status to the tool, and the output is the completion notification to the user.

[1042] (Application example 2)

[1043] 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."

[1044] There is a need for a system that can efficiently generate meeting minutes and accurately grasp key points of discussions and changes in participants' emotions by reflecting their emotional states in real time. Furthermore, in certain on-site environments, there is a need for a method to improve work efficiency by quickly notifying managers of the meeting content and emotion analysis results. This invention was developed to solve these problems.

[1045] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1046] In this invention, the server includes means for directly outputting the generated minutes to a specified tool, means for capturing meeting audio in real time and converting it into text, means for automatically generating minutes in real time using a generation AI, means for allowing a user to confirm and correct the generated minutes, means for receiving and authenticating output destination information for the specified tool from the user, means for a factory floor worker to start a meeting using smart glasses and record audio in real time and convert it into text, means for analyzing and recording emotional states, and means for remotely notifying a manager of the automatically generated minutes and the emotional analysis results, thereby enabling the real-time generation and output of meeting minutes and emotional states.

[1047] Minutes are documents that record the contents of a meeting, including what was said, decisions made, and summaries of the agenda.

[1048] "Designated tool" refers to an external tool, such as a collaboration tool or project management tool, selected by the user to send the minutes.

[1049] A "speech recognition engine" is a software or hardware technology for converting speech into text in real time.

[1050] "Generative AI" refers to artificial intelligence technology that automatically generates meeting minutes from given data.

[1051] "Emotional state" refers to the changes in the user's emotions and state that are analyzed from the voice data.

[1052] "Smart glasses" are wearable devices that can record conference audio in real time and display information.

[1053] "Destination information" is information about the tool or location designated by the user to send the minutes.

[1054] "API" stands for Application Programming Interface and refers to the methods and tools that allow different software components to communicate with each other.

[1055] "Emotion analysis results" are information on emotional states obtained by analyzing voice data.

[1056] A "manager" is a person in charge of managing the activities of field workers and the contents of meetings within a factory or organization.

[1057] In implementing the invention, the system mainly uses a server, smart glasses, and an administrator's terminal. The detailed process of each step is described below.

[1058] First, a field worker starts a conference using the smart glasses. The smart glasses are equipped with a microphone for recording audio in real time. When the user clicks the "Start a new conference" button on the smart glasses to start the conference, conference information is sent to the server. This information includes the conference title, participant list, and output destination tool.

[1059] Based on the received information, the server activates a speech recognition engine (e.g., Google Speech-to-Text API). Then, the smart glasses stream the recorded audio data of the meeting to the server in real time. The server receives the audio data and passes it to the speech recognition engine to convert it into text data. The speech recognition engine analyzes the audio for each speaker and generates specific text data.

[1060] At the same time, the server activates an emotion engine (e.g., Hugging Face emotion analysis model) to analyze the user's emotional state from the voice data in real time. Once the emotion engine analyzes the emotional state, that information is processed together with the text data.

[1061] The generated text data and emotion data are passed to a generation AI (for example, OpenAI's GPT-3) on the server. The generation AI analyzes this data and summarizes the meeting content. At the same time, minutes with emotion tags are generated. These minutes are temporarily stored on the server.

[1062] Users can view the generated minutes through the smart glasses interface and manually edit emotion tags and content as needed, or administrators can view and edit the minutes from their own devices.

[1063] Finally, the minutes that have been checked and corrected are automatically sent to the destination tool (for example, a project management tool or collaboration tool). The server achieves this output by using the destination tool's API based on the authentication information received from the user and sending the minutes in the specified format.

[1064] For example, during a safety meeting at a worksite, smart glasses record workers' speech and perform text and sentiment analysis. This information is displayed in real time on a manager's dashboard, allowing them to take the necessary action quickly. An example of a prompt for the generative AI model is as follows:

[1065] Example prompt sentence:

[1066] Transcribe an audio recording of a factory floor meeting and summarize it as follows:

[1067] 1. Main agenda of the meeting.

[1068] 2. Important statements by each speaker.

[1069] 3. Analysis of emotional state.

[1070] Please include the meeting title, participant list, and destination tool.

[1071] By implementing the invention in this way, it is possible to generate and output meeting minutes and emotional states in real time, thereby improving meeting efficiency and enabling quick information sharing.

[1072] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1073] Step 1:

[1074] The user uses the smart glasses to click the "Start a new meeting" button. At this time, the meeting title, participant list, and output tool are entered, and this is sent from the device to the server. The entered data is temporarily saved on the server as meeting information.

[1075] Step 2:

[1076] Based on the received meeting information, the server launches a speech recognition engine (e.g., Google Speech-to-Text API). The audio data of the meeting recorded by the smart glasses is streamed to the server in real time. The audio data is received as input, passed to the speech recognition engine, and converted into text data. This process outputs the meeting audio as a text-formatted transcript.

[1077] Step 3:

[1078] The server runs an emotion engine (e.g., Hugging Face emotion analysis model) and processes the voice data separately for emotion analysis. The input voice data is analyzed and the emotional state is digitized. The emotion data is generated in parallel with the text data, and these data are later integrated.

[1079] Step 4:

[1080] The server passes the text data generated by the speech recognition engine and the emotion data generated by the emotion engine to a generation AI (for example, OpenAI's GPT-3). The generation AI receives these data as input and analyzes and summarizes them. This process generates a summary of the meeting minutes with emotion tags added, which are then temporarily saved. The generation of these minutes completes the integration of the data summary and emotion information.

[1081] Step 5:

[1082] The user can review the summary minutes generated through the smart glasses interface and make any necessary corrections. The correction information entered through the smart glasses is then sent to the server, and the minutes are updated. After user confirmation, the final minutes are confirmed.

[1083] Step 6:

[1084] The server sends the final minutes confirmed by the user to the specified output tool (for example, a project management tool or collaboration tool). To send the minutes, the server uses the API of each tool and performs appropriate authentication based on the user's authentication information. This allows the minutes to be uploaded to the output tool in the appropriate format.

[1085] Step 7:

[1086] Once the output process is complete, the server sends a notification to the user. The notification is displayed on the smart glasses or the administrator's device, allowing the user to confirm that the meeting content has been output correctly. At this step, all processing is complete.

[1087] 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.

[1088] 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.

[1089] 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.

[1090] 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.

[1091] 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.

[1092] 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.

[1093] 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).

[1094] 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.

[1095] 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."

[1096] 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.

[1097] 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).

[1098] 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.

[1099] 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.

[1100] 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.

[1101] 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.

[1102] 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.

[1103] 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.

[1104] 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.

[1105] 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.

[1106] 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.

[1107] 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.

[1108] The following is further disclosed regarding the above embodiment.

[1109] (Claim 1)

[1110] a means for outputting the generated minutes directly to a specified tool;

[1111] A means to capture and convert meeting audio in real time into text;

[1112] A means for automatically generating meeting minutes in real time using generative AI;

[1113] means for allowing a user to review and correct the generated minutes;

[1114] means for receiving and authenticating output destination information for a tool designated by a user;

[1115] A system including:

[1116] (Claim 2)

[1117] 10. The system of claim 1, wherein a speech recognition engine is used as a means for capturing conference audio and converting it to text in real time.

[1118] (Claim 3)

[1119] The system according to claim 1, further comprising means for automatically outputting the generated minutes to a specific tool using an API.

[1120] "Example 1"

[1121] (Claim 1)

[1122] means for outputting the generated records directly to designated software;

[1123] A means to capture and convert meeting audio in real time into text;

[1124] A means for automatically generating meeting minutes in real time using generative AI;

[1125] means for allowing a user to review and correct the generated minutes;

[1126] means for receiving and authenticating output destination information for the software designated by the user;

[1127] means for inputting conference start information via a user interface;

[1128] means for streaming audio data in real time;

[1129] a means for analyzing the speech and segmenting the text data by speaker;

[1130] A means for formatting the summarized minutes into a template format;

[1131] A system including:

[1132] (Claim 2)

[1133] 10. The system of claim 1, wherein a speech recognition engine is used as a means for capturing conference audio and converting it to text in real time.

[1134] (Claim 3)

[1135] The system according to claim 1, further comprising means for automatically outputting the generated minutes to specific software using an API.

[1136] "Application Example 1"

[1137] (Claim 1)

[1138] means for outputting the generated information summary directly to a designated system;

[1139] a means for capturing and converting voice data into text in real time;

[1140] A means for automatically generating information summaries in real time using generative AI;

[1141] means for allowing an operator to review and modify the generated information summary;

[1142] means for receiving and authenticating output destination information for a system designated by an operator;

[1143] means for enabling the generated information summary to be viewed and modified on a display device;

[1144] A system including:

[1145] (Claim 2)

[1146] 10. The system of claim 1, wherein speech recognition technology is used as a means for capturing voice data and converting it to text in real time.

[1147] (Claim 3)

[1148] 10. The system of claim 1, further comprising means for automatically outputting the generated information summary to a specific system using an API.

[1149] "Example 2: Combining Emotion Engines"

[1150] (Claim 1)

[1151] A means to capture and convert meeting audio in real time into text;

[1152] A means for analyzing user emotions from voice data;

[1153] A method for automatically generating minutes using generative AI and reflecting emotional information.

[1154] means for allowing a user to review and correct the generated minutes;

[1155] A means for outputting the minutes directly to a designated tool;

[1156] means for receiving and authenticating output destination information for a tool designated by a user;

[1157] A system including:

[1158] (Claim 2)

[1159] 10. The system of claim 1, wherein a speech recognition engine is used as a means for capturing conference audio and converting it to text in real time.

[1160] (Claim 3)

[1161] The system according to claim 1, further comprising means for automatically outputting the generated minutes to a specific tool using an API.

[1162] "Application example 2 when combining emotion engines"

[1163] (Claim 1)

[1164] a means for outputting the generated minutes directly to a specified tool;

[1165] A means to capture and convert meeting audio in real time into text;

[1166] A means for automatically generating meeting minutes in real time using generative AI;

[1167] means for allowing a user to review and correct the generated minutes;

[1168] means for receiving and authenticating output destination information for a tool designated by a user;

[1169] A means for factory floor workers to use smart glasses to initiate meetings and record and transcribe audio in real time;

[1170] a means for analyzing and recording emotional states;

[1171] A means for remotely notifying an administrator of the automatically generated meeting minutes and sentiment analysis results; and

[1172] A system including:

[1173] (Claim 2)

[1174] 10. The system of claim 1, wherein a speech recognition engine is used as a means for capturing conference audio and converting it to text in real time.

[1175] (Claim 3)

[1176] The system according to claim 1, further comprising means for automatically outputting the generated minutes to a specific tool using an API. [Explanation of symbols]

[1177] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for outputting the generated minutes directly to a specified tool; A means to capture and convert meeting audio in real time into text; A means for automatically generating meeting minutes in real time using generative AI; means for allowing a user to review and correct the generated minutes; means for receiving and authenticating output destination information for a tool designated by a user; A system including:

2. 10. The system of claim 1, wherein a speech recognition engine is used as a means for capturing conference audio and converting it into text in real time.

3. The system according to claim 1 , further comprising means for automatically outputting the generated minutes to a specific tool using an API.

Images