system

Abstract

We provide the system. [Solution] A system including a speech recognition means, a generative AI means, and a means for converting the speech of meeting participants into text using speech recognition, and for the generative AI to create meeting minutes using that as input, and for highlighting important points and summaries of discussions and suggesting action items.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In conventional meetings, tasks such as creating meeting minutes, summarizing discussions, and proposing action items were performed manually, which required a significant amount of time and effort. Also, there was a possibility of misunderstandings and oversights due to being done by humans.

Means for Solving the Problems

[0005] This invention provides a system that combines speech recognition and generative AI. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this text as input to create meeting minutes. Furthermore, it highlights key points and summaries of discussions, and suggests action items. This facilitates the organization of meeting information, improves meeting productivity, reduces the burden on participants, and streamlines the decision-making process. [Brief explanation of the drawing]

[0006] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of the data processing device and smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 1 of Example 1. [Figure 12]This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment 2. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Form Example 2. [Figure 15] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 3 of Example 3. [Figure 16] This is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Form Example 3. [Figure 17] This is a sequence diagram showing the processing flow of the data processing system in Example 1 of the Form 1 when an emotion engine is combined. [Figure 18] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] This is a sequence diagram showing the processing flow of a data processing system in another embodiment. [Modes for carrying out the invention]

[0007] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0008] First, let's explain the terminology used in the following explanation.

[0009] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)), etc.

[0010] In the following embodiments, the labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0011] In the following embodiments, the labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0012] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), etc.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0014] [First Embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] As shown in FIG. 1, the 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.

[0017] 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 the "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. Also, the database 24 and the communication I / F 26 are 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).

[0018] 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. Also, the reception device 38, the output device 40, and the camera 42 are connected to the bus 52.

[0019] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0020] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0022] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0026] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0027] "Example of form 1"

[0028] In one embodiment of the present invention, a speech recognition module converts audio data input from a microphone installed in a conference room into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages. Next, a generative AI module receives the text data output from the speech recognition module as input and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' statements. "Embodiment Example 2"

[0029] As a concrete example of its use, when a meeting begins, microphones installed in the meeting room capture participants' speech, and a speech recognition module converts this into text data. A generative AI module analyzes this text data in real time and generates meeting minutes. When the meeting ends, the generative AI module creates a final version of the minutes, highlighting key points, summaries of discussions, and suggested action items. This makes it easier to organize information after the meeting and improves meeting productivity.

[0030] The following describes the processing flow for each example of the form.

[0031] "Example of form 1"

[0032] Step 1: Acquire audio data from the microphone installed in the conference room.

[0033] Step 2: The acquired audio data is input into the speech recognition module and converted into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages.

[0034] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0035] Step 4: The generative AI module analyzes the text data and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' remarks.

[0036] "Example of form 2"

[0037] Step 1: Once the meeting begins, microphones installed in the meeting room capture the participants' speech.

[0038] Step 2: Input the captured audio data into the speech recognition module and convert it into text data.

[0039] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0040] Step 4: The generative AI module analyzes the text data in real time and generates meeting minutes.

[0041] Step 5: Once the meeting concludes, the generative AI module creates a final version of the meeting minutes, highlighting key points, summaries of discussions, and suggested action items. This facilitates post-meeting information organization and improves meeting productivity.

[0042] (Example 1)

[0043] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0044] There is a need to efficiently record meeting content and provide an environment where participants can focus on the discussion. However, traditional methods are time-consuming to create meeting minutes, and there is a risk of overlooking important points and action items. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder minute-taking.

[0045] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0046] In this invention, the server includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, and a generative information processing means for analyzing the text data and automatically generating meeting minutes. This makes it possible to quickly and accurately record the contents of a meeting, create meeting minutes in multiple languages ​​without overlooking important points or action items.

[0047] "Audio data" refers to information that represents the speech of meeting participants in a digital format.

[0048] An "input device" is a device used to acquire audio data, and includes microphones and the like.

[0049] "Speech recognition means" refers to a technology or device for converting speech data into text data.

[0050] "Text data" refers to character information converted by speech recognition technology.

[0051] "Generative information processing means" refers to a technology or device for analyzing text data and automatically generating meeting minutes.

[0052] An "output device" is a device used to display or print the generated meeting minutes.

[0053] Meeting minutes are documents that record the content of a meeting, including important points and action items.

[0054] "Multiple languages" refers to a group of languages ​​with different linguistic systems, and means that speech recognition means are capable of recognizing them.

[0055] A description of embodiments for carrying out this invention will be given.

[0056] System Overview

[0057] This system is designed to efficiently record meeting content and provide an environment where participants can concentrate on the discussion. The system includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, a generative information processing means for analyzing the text data and automatically generating meeting minutes, and an output device for outputting the generated meeting minutes.

[0058] Hardware and software to be used

[0059] The microphone installed in the conference room will be used as the input device.

[0060] The speech recognition method employs deep learning-based speech recognition technology. Specifically, speech recognition software such as Google® Speech-to-Text API can be used.

[0061] The generative information processing system employs natural language processing technology. Specifically, it can use generative AI models such as OpenAI's GPT-3®.

[0062] The terminal's display or printer is used as the output device to display or print the generated meeting minutes.

[0063] Specific example

[0064] When a user says "Let's move on to the next agenda item" during a meeting, the microphone connected to the device picks up this audio.

[0065] The server uses speech recognition to convert the speech into text data that says, "Let's move on to the next topic."

[0066] The generative information processing system automatically generates meeting minutes, such as "The meeting has moved on to the next agenda item," based on this text data.

[0067] The server sends the generated meeting minutes to the terminal, and the user can view the minutes on the terminal.

[0068] Example of a prompt

[0069] "Please summarize the meeting content and create meeting minutes."

[0070] This system enables the rapid and accurate recording of meeting content, ensuring that important points and action items are not overlooked, and allows for the creation of meeting minutes in multiple languages.

[0071] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0072] Step 1:

[0073] When a user speaks in the conference room, a microphone connected to the terminal captures their voice. The input is the user's voice, and the output is digital audio data. The microphone converts the audio into an electrical signal, and the terminal prepares this to be sent to the server as digital data.

[0074] Step 2:

[0075] The server receives audio data transmitted from the terminal. The input is digital audio data, and the output is text data. The server activates speech recognition and converts the audio data into text data. In this process, speech recognition technology is used to analyze the waveform of the audio and generate the corresponding string of characters.

[0076] Step 3:

[0077] The server inputs text data obtained from the speech recognition means into the generative information processing means. The input is text data, and the output is the text of the meeting minutes. The generative information processing means analyzes the text data using natural language processing technology and automatically generates the meeting minutes. In this process, the key points of the text are extracted and the meeting minutes are constructed in a format that follows the progress of the meeting.

[0078] Step 4:

[0079] The server sends the generated meeting minutes to the terminal. The input is the text of the meeting minutes, and the output is the meeting minutes in a format viewable by the user. The terminal displays the received meeting minutes on its screen and shares them as needed, such as by printing or emailing. Users can review the meeting minutes on their terminal and reflect on the content of the meeting.

[0080] This series of processes makes it possible to record the contents of a meeting quickly and accurately, and to provide an environment where participants can concentrate on the discussion.

[0081] (Application Example 1)

[0082] Next, we will describe Application Example 1 of Form 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."

[0083] Manual record-keeping by workers for work reporting and progress management within a factory is time-consuming and labor-intensive, and presents challenges in terms of accuracy and efficiency. Furthermore, real-time monitoring of work status is difficult, making it challenging for managers to respond quickly.

[0084] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0085] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, using that as input for the generative AI to create a work record, and automatically generating logs for work progress management and quality control. This enables the automation of work reporting and real-time monitoring of work status.

[0086] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or software for accurately acquiring a worker's speech as textual information.

[0087] "Generative AI methods" refer to artificial intelligence technologies that use natural language processing techniques to automatically generate work records and progress management information based on input text data.

[0088] "Work records" are data that document the activities and progress of workers within a factory, and are useful information for improving work efficiency and quality control.

[0089] "Progress management" is a management method used to understand the progress of work and to confirm whether the work is proceeding according to plan.

[0090] "Quality control" refers to management activities carried out to maintain a consistent quality of products and services, and is a process for ensuring the accuracy and efficiency of work.

[0091] A "log" is data that shows the history and records of work, and is information used to review and analyze the work content later.

[0092] The system for carrying out this invention includes speech recognition means, generative AI means, and automatic work record generation means. The server uses speech recognition technology such as the Google Speech-to-Text API as the speech recognition means to convert the worker's speech into text data in real time. The converted text data is analyzed using natural language processing technology such as OpenAI GPT-3 as the generative AI means to automatically generate work records and progress management information.

[0093] The terminal allows workers to submit voice reports via an application installed on devices such as smartphones and tablets. Users report their work details by voice through the terminal, and this voice data is sent to a server. The server converts the voice data into text and generates work records using generative AI. The generated work records are stored in a cloud-based database and can be accessed by administrators in real time.

[0094] For example, if a worker reports "Line 1 maintenance complete, no abnormalities," the server will generate a work log such as "October 10, 2023, 14:30 Line 1 maintenance complete, no abnormalities." An example of a prompt to input to the generation AI model in this case would be, "Convert the work report into a log in the following format: Date and time, Line number, Work details, Result."

[0095] This system enables the automation of work reporting and real-time monitoring of work status, thereby improving work efficiency within the factory.

[0096] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0097] Step 1:

[0098] The user uses a terminal to report their work details verbally. The terminal acquires the user's voice data through a microphone and sends that data to a server. The input is the user's voice data, and the output is the transmission of voice data to the server.

[0099] Step 2:

[0100] The server converts received audio data into text data using speech recognition. Specifically, it analyzes the audio using the Google Speech-to-Text API and generates the corresponding text. The input is audio data, and the output is text data.

[0101] Step 3:

[0102] The server analyzes text data using generative AI methods and automatically generates work records. Using natural language processing technologies such as OpenAI GPT-3, it creates records that include work progress and results based on the text data. The input is text data, and the output is a work record.

[0103] Step 4:

[0104] The server saves the generated work records to a database in the cloud. This allows administrators to monitor the work status in real time. The input is the work records, and the output is the saving of the records to the database.

[0105] Step 5:

[0106] Administrators access a database in the cloud to review work records. This allows for efficient progress and quality control of work. Input is a request to access the database, and output is a display of work records.

[0107] (Example 2)

[0108] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0109] In meetings, accurately recording participants' remarks and efficiently organizing important information is crucial for improving meeting productivity. However, manual minute-taking is time-consuming and labor-intensive, and can lead to omissions and errors. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder the accurate transmission of information. To address these challenges, a system is needed that automatically transcribes audio data into text and extracts and organizes key information.

[0110] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0111] In this invention, the server includes means for acquiring audio data, means for converting the acquired audio data into text data, and means for analyzing the text data and generating meeting minutes. This makes it possible to record what is said during a meeting in real time and to efficiently organize important information.

[0112] "Means for acquiring audio data" refers to a device or method for capturing the speech of participants during a meeting in real time and saving it in digital format.

[0113] "Means for converting acquired audio data into text data" refers to an apparatus or method that performs the process of converting audio data into text information using speech recognition technology.

[0114] "Means for analyzing text data and generating meeting minutes" refers to a device or method that uses natural language processing technology to extract important information from text data, organize the content of a meeting, and output it as meeting minutes.

[0115] "Means for formatting generated meeting minutes into a final version and highlighting important information" refers to a device or method for organizing generated meeting minutes in an easy-to-read format and making particularly important points or action items stand out.

[0116] "Means for users to review and modify meeting minutes" refers to an interface or method that allows users to view the generated meeting minutes and modify or add to their content as needed.

[0117] This invention is a system that automatically records speeches in meetings and efficiently organizes important information. A specific embodiment of this system is described below.

[0118] The server uses audio input devices installed in the conference room to capture participants' speech in real time. Specifically, it uses multiple microphones as a typical audio input device to capture the audio signals as digital data.

[0119] Next, the server converts the acquired audio data into text data using speech recognition software. A general-purpose platform providing speech recognition technology can be used as this speech recognition software. This converts the audio data into text information.

[0120] Subsequently, the server analyzes the text data using a generative AI model and generates meeting minutes. This generative AI model can be a general-purpose generative AI platform utilizing natural language processing technology. The generated meeting minutes are formatted to highlight important points and action items.

[0121] Users can review the meeting minutes generated through their device and make corrections or additions as needed. This interface is designed to allow users to easily edit the meeting minutes.

[0122] For example, if the "new product launch plan" is discussed during a meeting, the server captures the statement, "The new product launch is scheduled for next month. We will decide on the marketing strategy in detail at the next meeting," and speech recognition software converts this to text. A generative AI model analyzes this text and generates meeting minutes like the following:

[0123] Key point: The new product is scheduled to be released next month.

[0124] Summary of discussion: Details regarding the marketing strategy will be decided at the next meeting.

[0125] Action item: Finalize the marketing strategy details at the next meeting.

[0126] An example of a prompt to input into a generative AI model might be, "Generate meeting minutes. Include key points, a summary of the discussion, and action items."

[0127] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0128] Step 1:

[0129] The server uses audio input devices installed in the conference room to capture participants' speech in real time. The input is the audio signal during the meeting, and the output is digital audio data. Specifically, it uses multiple microphones to capture the audio signal as digital data.

[0130] Step 2:

[0131] The server inputs the acquired audio data into speech recognition software and converts it into text data. The input is digital audio data, and the output is text data. Specifically, it performs a process of converting audio data into text information using speech recognition technology.

[0132] Step 3:

[0133] The server inputs text data into a generative AI model and generates meeting minutes. The input is text data, and the output is the generated meeting minutes. Specifically, it uses natural language processing techniques to analyze the text data, extract important information, and create the meeting minutes.

[0134] Step 4:

[0135] The server formats the generated meeting minutes as the final version and highlights important information. The input is the generated meeting minutes, and the output is the formatted, final version of the meeting minutes. Specifically, it organizes the content of the meeting minutes for easier reading and highlights particularly important points and action items.

[0136] Step 5:

[0137] The user reviews the meeting minutes generated through the terminal and makes corrections or additions as needed. The input is the formatted final version of the meeting minutes, and the output is the meeting minutes modified by the user. Specifically, the user views the meeting minutes and edits the content through the interface.

[0138] (Application Example 2)

[0139] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0140] In factory and other work environments, there is a need to efficiently record worker instructions and reports and to grasp the progress of work in real time. However, traditional methods require manual recording and management, which leads to decreased work efficiency. Furthermore, it is difficult to immediately grasp important points and progress of work, increasing the burden on managers.

[0141] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0142] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, and for the generative AI to create a work record using that as input, highlighting the progress of the work and important work items. This enables real-time work recording and progress management at the work site.

[0143] "Speech recognition means" refers to a technology that converts speech into text data, and is a device or software for recording a worker's speech in digital format.

[0144] "Generative AI methods" are artificial intelligence technologies that analyze input text data and generate information tailored to specific purposes, such as systems for automatically creating work records and progress highlights.

[0145] A "work record" is a text-based record of instructions and reports from workers at a work site, and serves as data for understanding the progress of work and important work items.

[0146] "Progress status" refers to information indicating how far a task has progressed, and is an essential indicator for efficient task management and planning.

[0147] "Important work items" refer to points in a task that require particular attention, or factors that significantly influence the success or failure of the task, and are information that managers should prioritize understanding.

[0148] The system for carrying out this invention is for recording workers' speech in real time and managing the progress of work in a workplace such as a factory. The server includes speech recognition means, generative AI means, and means for creating work records.

[0149] The speech recognition system captures the worker's speech and converts the audio data into text data. Specifically, speech recognition software such as the Google Speech-to-Text API can be used. The converted text data is sent to the server.

[0150] The generative AI system analyzes received text data and generates work records. Using generative AI models such as OpenAI GPT-3, it is possible to highlight work progress and important work items. The generated work records are displayed in real time on the administrator's terminal.

[0151] For example, if a worker says, "I will proceed to the next step," the speech recognition system converts this into text, and the generative AI system records "Step in progress: Proceeding to the next step" in the work log. This information is displayed on the terminal so that the administrator can check it immediately.

[0152] Examples of prompt messages include the following:

[0153] "Voice input: 'Proceed to the next step.'"

[0154] "Prompt: 'Update the work log and record that you are proceeding to the next step.'"

[0155] In this way, real-time work recording and progress management are achieved at the work site.

[0156] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0157] Step 1:

[0158] The user speaks at the work site. The voice input device captures the user's speech and sends it to the server as audio data. The input is the user's voice, and the output is audio data.

[0159] Step 2:

[0160] The server converts received audio data into text data using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to analyze the audio data and generate the corresponding text. The input is audio data, and the output is text data.

[0161] Step 3:

[0162] The server analyzes text data using generative AI methods and generates work records. Using generative AI models such as OpenAI GPT-3, it extracts work progress and important work items from the text data and creates a record. The input is text data, and the output is a work record.

[0163] Step 4:

[0164] The server sends the generated work log to the administrator's terminal. The administrator's terminal displays the received work log in real time, allowing them to check the progress of the work. The input is the work log, and the output is the display on the administrator's terminal.

[0165] Step 5:

[0166] The administrator monitors the progress of work based on the work records displayed on the terminal and issues instructions as needed. This enables efficient work management. The input is the work records displayed on the terminal, and the output is the administrator's judgment and instructions.

[0167] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0168] "Example of form 1"

[0169] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[0170] "Example of form 2"

[0171] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[0172] The following describes the processing flow for each example of the form.

[0173] "Example of form 1"

[0174] Step 1: Meeting participants speak. These speeches are entered into the system as audio data.

[0175] Step 2: The speech recognition system converts the spoken audio data into text data.

[0176] Step 3: The emotion engine recognizes emotions from the audio data of the speech. This emotion is input into the system as emotion information.

[0177] Step 4: The generative AI receives text data and emotional information as input.

[0178] Step 5: The generative AI creates meeting minutes from the text data, highlighting key points and summaries of the discussion.

[0179] Step 6: The generative AI suggests action items based on emotional information.

[0180] (Example 1)

[0181] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0182] In meetings, it is important to accurately record what participants say and create meeting minutes. However, traditional methods only record the content of what was said, making it difficult to create minutes that reflect the emotions of the participants or the progress of the meeting. Furthermore, it was difficult to appropriately grasp the emotional reactions that arose during the meeting and propose actions based on them.

[0183] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0184] In this invention, the server includes means for acquiring audio data, speech recognition means for converting the audio data into text data, generative information processing means for receiving the text data as input and automatically generating meeting minutes using natural language processing technology, and emotion recognition means for recognizing emotions from the audio data and providing that information to the generative information processing means. This makes it possible to create meeting minutes that reflect not only the content of what the participants said, but also their emotions and the progress of the meeting, thereby enabling the effective management of meetings.

[0185] "Audio data" refers to sound information acquired to record what participants say in a meeting.

[0186] "Speech recognition means" refers to technology that analyzes speech data and converts it into corresponding text data.

[0187] A "generative information processing system" is a system that receives text data as input and has the function of automatically generating meeting minutes using natural language processing technology.

[0188] "Emotion recognition means" refers to a technology that analyzes participants' emotions from audio data and provides that information to generative information processing means.

[0189] Meeting minutes are documents that record what was said, the progress of the meeting, and the feelings of the participants.

[0190] "Action items" refer to specific actions or follow-up items proposed during a meeting.

[0191] This invention is a system that efficiently records the content of discussions in meetings and automatically generates meeting minutes that reflect the emotions of the participants. The system acquires audio data, converts it into text data using speech recognition technology, and then creates meeting minutes using generative information processing technology. In addition, it analyzes the emotions of the participants using emotion recognition technology and reflects that information in the meeting minutes.

[0192] The server acquires user speech as audio data through microphones installed in the conference room. This audio data is converted into text data using speech recognition technologies such as Google Speech-to-Text API or IBM Watson® Speech to Text. The speech recognition method is capable of recognizing multiple languages.

[0193] Next, the server uses natural language processing technologies such as OpenAI's GPT-3 and Google's BERT as generative information processing tools. This allows for the automatic generation of meeting minutes based on text data, as well as management of meeting progress and summarization of participants' statements.

[0194] Furthermore, the server uses emotion recognition means to analyze the participants' emotions from the audio data. The emotion engine analyzes the tone, volume, and speed of the voice and provides the emotion information to the generative information processing means. Based on this emotion information, the generative information processing means adds information about emotions to the meeting minutes and proposes action items based on the emotion information.

[0195] For example, if a user says, "I am very dissatisfied with the slow progress of this project," the emotion engine recognizes anger from the tone of the statement. The generative information processing system records in the meeting minutes that "Participants are dissatisfied with the project's progress" and proposes "Schedule a follow-up meeting regarding project progress" as an action item.

[0196] An example of a prompt message is: "Convert meeting comments to text, analyze sentiment, and create meeting minutes. Also, suggest action items based on the speaker's sentiment."

[0197] The flow of the specific processing in Example 1 will be explained using Figure 15.

[0198] Step 1:

[0199] When a user speaks in a conference room, their voice is captured by a microphone installed in the room. The server receives the audio data transmitted from the microphone. The input is the user's voice, and the output is the audio data sent to the server. This audio data is used in subsequent processing steps.

[0200] Step 2:

[0201] The server passes the received audio data to the speech recognition system. The speech recognition system uses technologies such as the Google Speech-to-Text API or IBM Watson Speech to Text to convert the audio data into text data. The input is audio data, and the output is text data. This conversion allows the audio information to be treated as text information.

[0202] Step 3:

[0203] The server inputs text data output from the speech recognition system into a generative information processing system. The generative information processing system analyzes the text data using natural language processing technologies such as OpenAI's GPT-3 and Google's BERT, and automatically generates meeting minutes. The input is text data, and the output is meeting minutes. This process is used for managing the progress of meetings and summarizing participants' statements.

[0204] Step 4:

[0205] The server passes audio data to an emotion recognition system. The emotion recognition system analyzes the tone, volume, and speed of the voice to recognize the user's emotions. The input is audio data, and the output is emotion information. This information is used to understand the emotional state of the participants.

[0206] Step 5:

[0207] The server passes the emotion information provided by the emotion recognition means to the generative information processing means. The generative information processing means adds information about emotions to the meeting minutes based on the emotion information and proposes action items based on the emotion information. The input is emotion information, and the output is meeting minutes and action items that reflect the emotions. This process makes the meeting minutes more detailed and useful.

[0208] (Application Example 1)

[0209] Next, we will describe Application Example 1 of Form 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."

[0210] Efficiently recording worker instructions and reports and managing the progress of work is crucial in the workplace. However, conventional methods make it difficult to respond appropriately while considering workers' emotions and stress levels, resulting in insufficient improvements in work efficiency and safety.

[0211] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0212] In this invention, the server includes speech recognition means, generative information processing means, and emotion recognition means. This makes it possible to transcribe the worker's speech into text in real time, automatically generate work records, and recognize the worker's emotions to suggest appropriate actions.

[0213] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or system that converts a worker's speech into text information in real time.

[0214] "Generative information processing means" refers to technology that processes information based on text data and automatically generates work records and action items.

[0215] "Emotion recognition means" refers to a technology that analyzes the emotions of a worker from voice data and recognizes their emotional state.

[0216] A "work record" is a document that records the progress and important points of a task, generated based on the worker's spoken content.

[0217] "Action items" are specific actions or countermeasures proposed based on the worker's emotional state and the nature of their work.

[0218] The system for carrying out this invention includes speech recognition means, generative information processing means, and emotion recognition means. The server uses the Google Cloud Speech-to-Text API as the speech recognition means to convert the worker's voice into text data in real time. The generative information processing means uses an OpenAI generative AI model to automatically generate work records based on the text data and propose action items. The emotion recognition means uses IBM Watson Tone Analyzer to analyze the worker's emotions from the voice data and provides that information to the generative information processing means.

[0219] The system uses smartphones or tablets as terminals, connecting a microphone to collect voice data. Users input instructions and reports at the work site using voice, and the content is automatically converted to text and saved as a work record.

[0220] For example, if a worker says, "This task is difficult," the emotion recognition system recognizes stress, and the generative information processing system generates the action item, "The worker may be experiencing stress. We suggest a break."

[0221] An example of a prompt message is: "Transcribe the worker's statements into text, analyze their emotions, and suggest necessary actions."

[0222] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[0223] Step 1:

[0224] The terminal collects the worker's voice through a microphone. The input is the worker's voice data, and the output is that voice data. The terminal sends this voice data to the server.

[0225] Step 2:

[0226] The server uses the Google Cloud Speech-to-Text API as a speech recognition tool to convert received audio data into text data. The input is audio data, and the output is text data. This conversion allows the operator's speech to be obtained as text information.

[0227] Step 3:

[0228] The server uses IBM Watson Tone Analyzer as an emotion recognition tool to analyze the worker's emotions from text data. The input is text data, and the output is emotion information. The server provides this emotion information to a generative information processing system.

[0229] Step 4:

[0230] The server uses OpenAI's generative AI model as a generative information processing tool to automatically generate work records based on text data and sentiment information. The input is text data and sentiment information, and the output is the work record. The server saves this work record and suggests action items as needed.

[0231] Step 5:

[0232] The user reviews the generated work records and action items on their terminal. The input consists of work records and action items, while the output is the information the user reviews. Based on this, the user can consider the progress of the work and potential countermeasures.

[0233] (Example 2)

[0234] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0235] In meetings, it is difficult to accurately record what participants say and to create minutes that reflect their emotional state. Furthermore, while it is necessary to quickly grasp important points and action items after a meeting, traditional methods are time-consuming and labor-intensive.

[0236] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0237] In this invention, the server includes means for acquiring audio data, means for converting audio data into text information, and means for recognizing emotional information. This makes it possible to accurately record the content of meetings and create meeting minutes that reflect the emotional state of the participants.

[0238] "Means for acquiring audio data" refers to a device or method for collecting the speech of meeting participants in real time.

[0239] "Means for converting audio data into text information" refers to a technology or device for analyzing acquired audio data and converting it into a corresponding string of characters.

[0240] "Means for analyzing textual information to generate meeting records" refers to a technology or device that uses converted textual information to organize the content of a meeting and record important points and the flow of discussion.

[0241] "Means for recognizing emotional information" refers to technologies or devices for analyzing and identifying emotional states from participants' voices.

[0242] "Methods for revising meeting records based on emotional information to emphasize important information and action items" refers to technologies or devices that utilize recognized emotional information to add emotional nuances to meeting records and clarify important information and next steps to take.

[0243] This invention is a system that accurately records the content of speeches and the emotional state of participants during meetings, and generates effective meeting minutes. When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time.

[0244] When the server receives audio data, it uses speech recognition software to convert the audio data into text. Specifically, a common speech recognition API can be used for speech recognition. This conversion records the speeches during the meeting in text format.

[0245] Next, the server uses a generative AI model to analyze the converted text information and generate a meeting record. This generative AI model can utilize natural language processing technology. The generative AI model extracts important points from the text information and organizes the flow of the discussion.

[0246] Furthermore, the server uses an emotion recognition engine to recognize emotional information from the audio data. The emotion recognition engine analyzes the tone, volume, and speed of the speech to determine the emotional state of the participants. This allows the server to grasp the emotional nuances contained in what is said during the meeting.

[0247] The server modifies the generated meeting minutes based on recognized sentiment information, highlighting important information and action items. Specifically, it uses sentiment information to add information to the minutes such as "the speaker may be feeling dissatisfied." It can also suggest action items based on sentiment information, such as "follow up on the speaker's dissatisfaction."

[0248] An example of a prompt might be, "Please create meeting minutes. Based on the following text data and sentiment information, highlight key points, summaries of discussions, and action items." This system ensures that meeting minutes are not merely a record of what was said, but also reflect the emotional state of the participants, enabling more effective meeting management.

[0249] The flow of the specific processing in Example 2 will be explained using Figure 17.

[0250] Step 1:

[0251] When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time. The input is the participants' speech, and the output is the audio data sent to the server. Specifically, the microphone captures the audio and transmits the data to the server over the network.

[0252] Step 2:

[0253] The server passes the received audio data to the speech recognition software. The speech recognition software converts the audio data into text information. The input is audio data, and the output is text information. As a data processing step, the audio signal is analyzed and the corresponding string is generated. Specifically, the speech recognition API analyzes the audio data and converts it into text format.

[0254] Step 3:

[0255] The server inputs text information into a generative AI model. The generative AI model analyzes the text information and generates meeting minutes. The input is text information, and the output is meeting minutes. As a data calculation, it extracts important points from the text information and organizes the flow of the discussion. Specifically, it utilizes natural language processing techniques to summarize the text data and create meeting minutes.

[0256] Step 4:

[0257] The server passes the audio data to the emotion recognition engine. The emotion recognition engine recognizes emotional information from the audio. The input is audio data, and the output is emotional information. As part of the data processing, the tone, volume, and speed of the voice are analyzed to determine the emotional state. Specifically, an emotion analysis tool analyzes the audio data and identifies emotional nuances.

[0258] Step 5:

[0259] The server provides emotional information to the generative AI model. The generative AI model modifies the meeting minutes based on the emotional information, highlighting important information and action items. The input is emotional information and the meeting minutes, and the output is the modified meeting minutes. As a data calculation, emotional information is used to add emotional nuances to the minutes and clarify action items. Specifically, the generative AI model analyzes the emotional information and adds information to the minutes such as "the speaker may be feeling dissatisfied."

[0260] Step 6:

[0261] The server provides the user with the finalized meeting transcript. The user can view the meeting transcript through their terminal and understand the content of the meeting. The input is the revised meeting transcript, and the output is the meeting transcript provided to the user. Specifically, the server sends the meeting transcript to the user's terminal, and the user views it.

[0262] (Application Example 2)

[0263] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0264] In a work environment, it is crucial to accurately record what workers say and create meeting minutes in real time. However, conventional systems have difficulty considering workers' emotions when creating meeting minutes or suggesting action items, limiting improvements in work efficiency. Furthermore, they are unable to properly recognize workers' emotions and follow up accordingly, thus necessitating improvements to the work environment.

[0265] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0266] In this invention, the server includes speech recognition means, generative AI means, and emotion recognition means. This enables the conversion of a worker's speech into text via speech recognition in the work environment, the generative AI to create meeting minutes in real time, the emotion recognition means to recognize the worker's emotions, and the suggestion of action items based on that information.

[0267] "Speech recognition means" refers to a technology that converts speech data into text data, making it possible to process spoken content in a digital format.

[0268] "Generative AI methods" refer to artificial intelligence technologies that use input text data to create meeting minutes, extract key points, and suggest action items.

[0269] "Emotion recognition means" refers to technology that analyzes the speaker's emotions from audio data and recognizes their emotional state.

[0270] "Meeting minutes" are documents that record the content of conversations during meetings or work sessions and compile them in a format that can be referenced later.

[0271] An "action item" is an item that indicates an issue identified in meeting minutes or the work environment, or the next action to be taken.

[0272] "Work environment" refers to the location and conditions in which a specific task is performed, such as a factory or office.

[0273] "Real-time" means that data processing and information provision occur immediately, resulting in a state where results are obtained without delay.

[0274] The system for carrying out this invention is centered around a server that includes speech recognition means, generative AI means, and emotion recognition means. The server receives the speech of workers as audio data through microphones installed in the work environment, such as a factory or office. The speech recognition means converts this audio data into text data.

[0275] The generative AI system creates meeting minutes in real time based on the converted text data, extracting key points and action items. Furthermore, the emotion recognition system analyzes the emotions of the participants from the audio data and provides this information to the generative AI system. This enables the generative AI system to create meeting minutes and suggest action items that take emotional information into account.

[0276] As a concrete example, during a morning meeting in a factory, when a worker expresses their opinion on a new work procedure, the server records their statement and, if the worker is feeling anxious, suggests an action item such as "Provide additional explanation to alleviate the worker's anxiety."

[0277] An example of a prompt message would be: "Analyze the worker's feelings based on this statement and reflect it in the meeting minutes. If the worker is feeling anxious, suggest a follow-up."

[0278] This system combines speech recognition using the Google Cloud Speech-to-Text API, generative AI using OpenAI's GPT model, and emotion recognition using Microsoft Azure's Emotion API. This enables more efficient communication in the work environment and allows for the creation of meeting minutes that take into account the emotions of the workers.

[0279] The flow of the specific process in Application Example 2 will be described using FIG. 18.

[0280] Step 1:

[0281] The server receives the operator's speech as voice data through the microphone installed in the working environment. This voice data serves as the input. The server collects the voice data and prepares to pass it to the next processing step.

[0282] Step 2:

[0283] The server uses voice recognition means to convert the received voice data into text data. In this process, the Google Cloud Speech-to-Text API is used to analyze the voice data and output text data. The converted text data becomes the input for the next step.

[0284] Step 3:

[0285] The server uses generative AI means to create a real-time meeting minutes based on the text data. In this process, the GPT model of OpenAI is used to extract important points and action items from the text data. The generated meeting minutes are the output.

[0286] Step 4:

[0287] The server uses emotion recognition means to analyze the operator's emotion from the voice data. In this process, the Emotion API of Microsoft Azure is used to analyze the voice data and output emotion information. The emotion information becomes the input for the next step.

[0288] Step 5:

[0289] The server provides emotional information to a generative AI system, which then reflects this emotional information in the meeting minutes. This process adds emotion-based comments and action items to the minutes based on the emotional information. The final meeting minutes are then output.

[0290] Step 6:

[0291] Users review the final meeting minutes provided by the server and perform action items as needed. They then refer to the minutes to improve their work environment and follow up.

[0292] (Other examples)

[0293] Next, other embodiments will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0294] In meetings, efficiently recording participants' remarks and quickly extracting key points and action items is crucial for enhancing meeting effectiveness. However, manual minute-taking is time-consuming and labor-intensive, and creating minutes that take participants' emotions into account is difficult. Furthermore, when multiple languages ​​are required, language barriers can hinder efficient minute-taking. A system is needed to address these challenges.

[0295] The identification processing performed by the identification processing unit 290 of the data processing device 12 in other embodiments is realized by the following means.

[0296] In this invention, the server includes: an audio data acquisition means equipped with a microphone for acquiring the speech of meeting participants; an audio recognition means that uses an acoustic model and a language model to convert the acquired audio data into text data in multiple languages; a generative AI means that analyzes the text data and inputs it as a prompt to instruct a generative AI model to automatically generate meeting minutes, and suggests key points, summaries of discussions, and action items; an emotion recognition means that analyzes audio features to recognize the user's emotions from the audio data and provides the emotion information to the generative AI means; and a means that, based on the provided emotion information, adjusts prompts to instruct the generative AI model to create meeting minutes and summarize discussions based on the emotion information, and suggests action items based on the emotion information. This enables efficient creation of meeting minutes and the suggestion of action items that take into account the emotions of the participants.

[0297] "Voice data acquisition means" refers to a device or function that uses a microphone to collect voice data in order to capture the speech of meeting participants.

[0298] "Speech recognition means" refers to a device or function that converts acquired speech data into text data in multiple languages ​​using an acoustic model and a language model.

[0299] A "generative AI means" is a device or function that analyzes text data, generates prompts to instruct a generative AI model to automatically generate meeting minutes, highlights key points and summaries of discussions, and suggests action items.

[0300] "Emotion recognition means" refers to a device or function that analyzes speech features from speech data, recognizes the user's emotions, and provides that emotion information to a generative AI means.

[0301] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate text data based on input prompts.

[0302] A "prompt" is an input sentence used to instruct a generative AI model to perform a specific task.

[0303] This invention is a system that improves the efficiency of creating meeting minutes and enables the proposal of action items considering the emotions of participants. Specific embodiments of this system are shown below.

[0304] The server uses a microphone connected to the terminal to obtain the speech of meeting participants. The audio data is sent to the server in real time. The server converts the obtained audio data into text data using speech recognition software such as Google Cloud Speech-to-Text API. At this time, an acoustic model and a language model are utilized to support multiple languages.

[0305] The converted text data is analyzed by the server, and a prompt sentence for instructing the automatic generation of meeting minutes for a generative AI model such as OpenAI's GPT-4 (registered trademark) is generated. As a specific example, a prompt sentence such as "The topic of the meeting is the progress report of the project. Please summarize the important points." is generated.

[0306] Furthermore, the server extracts audio features from the audio data and recognizes the emotions of the user using emotion recognition software such as IBM Watson Tone Analyzer. The recognized emotion information is provided to the generative AI means. The server adjusts the prompt for instructing the generative AI model to create meeting minutes or summarize discussions based on the provided emotion information. For example, when the participant shows dissatisfaction, a prompt sentence such as "Please emphasize the improvement points considering the dissatisfaction of the participant." is generated.

[0307] The generation AI model generates meeting minutes based on tailored prompts, suggesting key points and action items. The generated minutes are sent to the device for user review. Users can review the minutes displayed on their device and make corrections or provide feedback as needed.

[0308] Example prompt: The meeting agenda is a project progress report; please summarize the key points.

[0309] The flow of a specific process in another embodiment will be explained using Figure 19.

[0310] Step 1:

[0311] The terminal uses a built-in or external microphone to capture the speech of meeting participants. The captured audio data is converted to a digital format and prepared for transmission to the server. The input is the speech of the meeting participants, and the output is audio data in digital format.

[0312] Step 2:

[0313] The device transmits the acquired audio data to the server via the internet. The data is transmitted using a secure protocol (e.g., HTTPS). The input is digital audio data, and the output is the audio data sent to the server.

[0314] Step 3:

[0315] The server converts received audio data into text data using the Google Cloud Speech-to-Text API. This process leverages acoustic and language models to support multiple languages. The input is audio data sent to the server, and the output is text data.

[0316] Step 4:

[0317] The server analyzes the converted text data and generates prompts to instruct generative AI models, such as OpenAI's GPT-4, to automatically generate meeting minutes. For example, it might generate a prompt stating, "The meeting agenda is a project progress report; please summarize the key points." The input is text data, and the output is prompts.

[0318] Step 5:

[0319] The server extracts speech features from the audio data and recognizes the user's emotions using emotion recognition software such as IBM Watson Tone Analyzer. The recognized emotion information is provided to a generative AI system. The input is audio data, and the output is emotion information.

[0320] Step 6:

[0321] The server adjusts prompts to instruct the generative AI model to create meeting minutes and summarize discussions based on the provided sentiment information. For example, if a participant expresses dissatisfaction, it might generate a prompt such as, "Consider the participant's dissatisfaction and highlight areas for improvement." The input is sentiment information, and the output is the adjusted prompt.

[0322] Step 7:

[0323] The server uses a generative AI model to generate meeting minutes based on the adjusted prompts, suggesting key points and action items. The generated meeting minutes are sent to the terminal. The input is the adjusted prompt text, and the output is the generated meeting minutes.

[0324] Step 8:

[0325] The terminal receives the generated meeting minutes sent from the server and displays them to the user. The user can review the displayed minutes and make corrections as needed. The input is the generated meeting minutes, and the output is the meeting minutes displayed to the user.

[0326] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0327] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0328] Other examples of generative AI include Gemini® (registered trademark) (Internet search). <url: https: gemini.google.com ?hl="ja">) are some examples.

[0329] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0330] [Second Embodiment]

[0331] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0332] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0333] The data processing device 12 includes a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a “computer” related to the technology of this disclosure.

[0334] Computer 22 includes a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. A database 24 and a communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0335] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0336] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0337] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0338] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0339] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0340] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0341] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0342] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0343] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0344] "Example of form 1"

[0345] In one embodiment of the present invention, a speech recognition module converts audio data input from a microphone installed in a conference room into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages. Next, a generative AI module receives the text data output from the speech recognition module as input and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' statements. "Embodiment Example 2"

[0346] As a concrete example of its use, when a meeting begins, microphones installed in the meeting room capture participants' speech, and a speech recognition module converts this into text data. A generative AI module analyzes this text data in real time and generates meeting minutes. When the meeting ends, the generative AI module creates a final version of the minutes, highlighting key points, summaries of discussions, and suggested action items. This makes it easier to organize information after the meeting and improves meeting productivity.

[0347] The following describes the processing flow for each example of the form.

[0348] "Example of form 1"

[0349] Step 1: Acquire audio data from the microphone installed in the conference room.

[0350] Step 2: The acquired audio data is input into the speech recognition module and converted into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages.

[0351] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0352] Step 4: The generative AI module analyzes the text data and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' remarks.

[0353] "Example of form 2"

[0354] Step 1: Once the meeting begins, microphones installed in the meeting room capture the participants' speech.

[0355] Step 2: Input the captured audio data into the speech recognition module and convert it into text data.

[0356] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0357] Step 4: The generative AI module analyzes the text data in real time and generates meeting minutes.

[0358] Step 5: Once the meeting concludes, the generative AI module creates a final version of the meeting minutes, highlighting key points, summaries of discussions, and suggested action items. This facilitates post-meeting information organization and improves meeting productivity.

[0359] (Example 1)

[0360] Next, we will describe Example 1 of Form Example 1. 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".

[0361] There is a need to efficiently record meeting content and provide an environment where participants can focus on the discussion. However, traditional methods are time-consuming to create meeting minutes, and there is a risk of overlooking important points and action items. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder minute-taking.

[0362] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0363] In this invention, the server includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, and a generative information processing means for analyzing the text data and automatically generating meeting minutes. This makes it possible to quickly and accurately record the contents of a meeting, create meeting minutes in multiple languages ​​without overlooking important points or action items.

[0364] "Audio data" refers to information that represents the speech of meeting participants in a digital format.

[0365] An "input device" is a device used to acquire audio data, and includes microphones and the like.

[0366] "Speech recognition means" refers to a technology or device for converting speech data into text data.

[0367] "Text data" refers to character information converted by speech recognition technology.

[0368] "Generative information processing means" refers to a technology or device for analyzing text data and automatically generating meeting minutes.

[0369] An "output device" is a device used to display or print the generated meeting minutes.

[0370] Meeting minutes are documents that record the content of a meeting, including important points and action items.

[0371] "Multiple languages" refers to a group of languages ​​with different linguistic systems, and means that speech recognition means are capable of recognizing them.

[0372] A description of embodiments for carrying out this invention will be given.

[0373] System Overview

[0374] This system is designed to efficiently record meeting content and provide an environment where participants can concentrate on the discussion. The system includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, a generative information processing means for analyzing the text data and automatically generating meeting minutes, and an output device for outputting the generated meeting minutes.

[0375] Hardware and software to be used

[0376] The microphone installed in the conference room will be used as the input device.

[0377] The speech recognition method will utilize deep learning-based speech recognition technology. Specifically, speech recognition software such as the Google Speech-to-Text API can be used.

[0378] The generative information processing system employs natural language processing technology. Specifically, generative AI models such as OpenAI's GPT-3 can be used.

[0379] The terminal's display or printer is used as the output device to display or print the generated meeting minutes.

[0380] Specific example

[0381] When a user says "Let's move on to the next agenda item" during a meeting, the microphone connected to the device picks up this audio.

[0382] The server uses speech recognition to convert the speech into text data that says, "Let's move on to the next topic."

[0383] The generative information processing system automatically generates meeting minutes, such as "The meeting has moved on to the next agenda item," based on this text data.

[0384] The server sends the generated meeting minutes to the terminal, and the user can view the minutes on the terminal.

[0385] Example of a prompt

[0386] "Please summarize the meeting content and create meeting minutes."

[0387] This system enables the rapid and accurate recording of meeting content, ensuring that important points and action items are not overlooked, and allows for the creation of meeting minutes in multiple languages.

[0388] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0389] Step 1:

[0390] When a user speaks in the conference room, a microphone connected to the terminal captures their voice. The input is the user's voice, and the output is digital audio data. The microphone converts the audio into an electrical signal, and the terminal prepares this to be sent to the server as digital data.

[0391] Step 2:

[0392] The server receives audio data transmitted from the terminal. The input is digital audio data, and the output is text data. The server activates speech recognition and converts the audio data into text data. In this process, speech recognition technology is used to analyze the waveform of the audio and generate the corresponding string of characters.

[0393] Step 3:

[0394] The server inputs text data obtained from the speech recognition means into the generative information processing means. The input is text data, and the output is the text of the meeting minutes. The generative information processing means analyzes the text data using natural language processing technology and automatically generates the meeting minutes. In this process, the key points of the text are extracted and the meeting minutes are constructed in a format that follows the progress of the meeting.

[0395] Step 4:

[0396] The server sends the generated meeting minutes to the terminal. The input is the text of the meeting minutes, and the output is the meeting minutes in a format viewable by the user. The terminal displays the received meeting minutes on its screen and shares them as needed, such as by printing or emailing. Users can review the meeting minutes on their terminal and reflect on the content of the meeting.

[0397] This series of processes makes it possible to record the contents of a meeting quickly and accurately, and to provide an environment where participants can concentrate on the discussion.

[0398] (Application Example 1)

[0399] Next, we will describe Application Example 1 of Form Example 1. 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."

[0400] Manual record-keeping by workers for work reporting and progress management within a factory is time-consuming and labor-intensive, and presents challenges in terms of accuracy and efficiency. Furthermore, real-time monitoring of work status is difficult, making it challenging for managers to respond quickly.

[0401] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0402] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, using that as input for the generative AI to create a work record, and automatically generating logs for work progress management and quality control. This enables the automation of work reporting and real-time monitoring of work status.

[0403] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or software for accurately acquiring a worker's speech as textual information.

[0404] "Generative AI methods" refer to artificial intelligence technologies that use natural language processing techniques to automatically generate work records and progress management information based on input text data.

[0405] "Work records" are data that document the activities and progress of workers within a factory, and are useful information for improving work efficiency and quality control.

[0406] "Progress management" is a management method used to understand the progress of work and to confirm whether the work is proceeding according to plan.

[0407] "Quality control" refers to management activities carried out to maintain a consistent quality of products and services, and is a process for ensuring the accuracy and efficiency of work.

[0408] A "log" is data that shows the history and records of work, and is information used to review and analyze the work content later.

[0409] The system for carrying out this invention includes speech recognition means, generative AI means, and automatic work record generation means. The server uses speech recognition technology such as the Google Speech-to-Text API as the speech recognition means to convert the worker's speech into text data in real time. The converted text data is analyzed using natural language processing technology such as OpenAI GPT-3 as the generative AI means to automatically generate work records and progress management information.

[0410] The terminal allows workers to submit voice reports via an application installed on devices such as smartphones and tablets. Users report their work details by voice through the terminal, and this voice data is sent to a server. The server converts the voice data into text and generates work records using generative AI. The generated work records are stored in a cloud-based database and can be accessed by administrators in real time.

[0411] For example, if a worker reports "Line 1 maintenance complete, no abnormalities," the server will generate a work log such as "October 10, 2023, 14:30 Line 1 maintenance complete, no abnormalities." An example of a prompt to input to the generation AI model in this case would be, "Convert the work report into a log in the following format: Date and time, Line number, Work details, Result."

[0412] This system enables the automation of work reporting and real-time monitoring of work status, thereby improving work efficiency within the factory.

[0413] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0414] Step 1:

[0415] The user uses a terminal to report their work details verbally. The terminal acquires the user's voice data through a microphone and sends that data to a server. The input is the user's voice data, and the output is the transmission of voice data to the server.

[0416] Step 2:

[0417] The server converts received audio data into text data using speech recognition. Specifically, it analyzes the audio using the Google Speech-to-Text API and generates the corresponding text. The input is audio data, and the output is text data.

[0418] Step 3:

[0419] The server analyzes text data using generative AI methods and automatically generates work records. Using natural language processing technologies such as OpenAI GPT-3, it creates records that include work progress and results based on the text data. The input is text data, and the output is a work record.

[0420] Step 4:

[0421] The server saves the generated work records to a database in the cloud. This allows administrators to monitor the work status in real time. The input is the work records, and the output is the saving of the records to the database.

[0422] Step 5:

[0423] Administrators access a database in the cloud to review work records. This allows for efficient progress and quality control of work. Input is a request to access the database, and output is a display of work records.

[0424] (Example 2)

[0425] Next, we will describe Example 2 of Form Example 2. 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".

[0426] In meetings, accurately recording participants' remarks and efficiently organizing important information is crucial for improving meeting productivity. However, manual minute-taking is time-consuming and labor-intensive, and can lead to omissions and errors. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder the accurate transmission of information. To address these challenges, a system is needed that automatically transcribes audio data into text and extracts and organizes key information.

[0427] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0428] In this invention, the server includes means for acquiring audio data, means for converting the acquired audio data into text data, and means for analyzing the text data and generating meeting minutes. This makes it possible to record what is said during a meeting in real time and to efficiently organize important information.

[0429] "Means for acquiring audio data" refers to a device or method for capturing the speech of participants during a meeting in real time and saving it in digital format.

[0430] "Means for converting acquired audio data into text data" refers to an apparatus or method that performs the process of converting audio data into text information using speech recognition technology.

[0431] "Means for analyzing text data and generating meeting minutes" refers to a device or method that uses natural language processing technology to extract important information from text data, organize the content of a meeting, and output it as meeting minutes.

[0432] "Means for formatting generated meeting minutes into a final version and highlighting important information" refers to a device or method for organizing generated meeting minutes in an easy-to-read format and making particularly important points or action items stand out.

[0433] "Means for users to review and modify meeting minutes" refers to an interface or method that allows users to view the generated meeting minutes and modify or add to their content as needed.

[0434] This invention is a system that automatically records speeches in meetings and efficiently organizes important information. A specific embodiment of this system is described below.

[0435] The server uses audio input devices installed in the conference room to capture participants' speech in real time. Specifically, it uses multiple microphones as a typical audio input device to capture the audio signals as digital data.

[0436] Next, the server converts the acquired audio data into text data using speech recognition software. A general-purpose platform providing speech recognition technology can be used as this speech recognition software. This converts the audio data into text information.

[0437] Subsequently, the server analyzes the text data using a generative AI model and generates meeting minutes. This generative AI model can be a general-purpose generative AI platform utilizing natural language processing technology. The generated meeting minutes are formatted to highlight important points and action items.

[0438] Users can review the meeting minutes generated through their device and make corrections or additions as needed. This interface is designed to allow users to easily edit the meeting minutes.

[0439] For example, if the "new product launch plan" is discussed during a meeting, the server captures the statement, "The new product launch is scheduled for next month. We will decide on the marketing strategy in detail at the next meeting," and speech recognition software converts this to text. A generative AI model analyzes this text and generates meeting minutes like the following:

[0440] Key point: The new product is scheduled to be released next month.

[0441] Summary of discussion: Details regarding the marketing strategy will be decided at the next meeting.

[0442] Action item: Finalize the marketing strategy details at the next meeting.

[0443] An example of a prompt to input into a generative AI model might be, "Generate meeting minutes. Include key points, a summary of the discussion, and action items."

[0444] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0445] Step 1:

[0446] The server uses audio input devices installed in the conference room to capture participants' speech in real time. The input is the audio signal during the meeting, and the output is digital audio data. Specifically, it uses multiple microphones to capture the audio signal as digital data.

[0447] Step 2:

[0448] The server inputs the acquired audio data into speech recognition software and converts it into text data. The input is digital audio data, and the output is text data. Specifically, it performs a process of converting audio data into text information using speech recognition technology.

[0449] Step 3:

[0450] The server inputs text data into a generative AI model and generates meeting minutes. The input is text data, and the output is the generated meeting minutes. Specifically, it uses natural language processing techniques to analyze the text data, extract important information, and create the meeting minutes.

[0451] Step 4:

[0452] The server formats the generated meeting minutes as the final version and highlights important information. The input is the generated meeting minutes, and the output is the formatted, final version of the meeting minutes. Specifically, it organizes the content of the meeting minutes for easier reading and highlights particularly important points and action items.

[0453] Step 5:

[0454] The user reviews the meeting minutes generated through the terminal and makes corrections or additions as needed. The input is the formatted final version of the meeting minutes, and the output is the meeting minutes modified by the user. Specifically, the user views the meeting minutes and edits the content through the interface.

[0455] (Application Example 2)

[0456] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0457] In factory and other work environments, there is a need to efficiently record worker instructions and reports and to grasp the progress of work in real time. However, traditional methods require manual recording and management, which leads to decreased work efficiency. Furthermore, it is difficult to immediately grasp important points and progress of work, increasing the burden on managers.

[0458] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0459] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, and for the generative AI to create a work record using that as input, highlighting the progress of the work and important work items. This enables real-time work recording and progress management at the work site.

[0460] "Speech recognition means" refers to a technology that converts speech into text data, and is a device or software for recording a worker's speech in digital format.

[0461] "Generative AI methods" are artificial intelligence technologies that analyze input text data and generate information tailored to specific purposes, such as systems for automatically creating work records and progress highlights.

[0462] A "work record" is a text-based record of instructions and reports from workers at a work site, and serves as data for understanding the progress of work and important work items.

[0463] "Progress status" refers to information indicating how far a task has progressed, and is an essential indicator for efficient task management and planning.

[0464] "Important work items" refer to points in a task that require particular attention, or factors that significantly influence the success or failure of the task, and are information that managers should prioritize understanding.

[0465] The system for carrying out this invention is for recording workers' speech in real time and managing the progress of work in a workplace such as a factory. The server includes speech recognition means, generative AI means, and means for creating work records.

[0466] The speech recognition system captures the worker's speech and converts the audio data into text data. Specifically, speech recognition software such as the Google Speech-to-Text API can be used. The converted text data is sent to the server.

[0467] The generative AI system analyzes received text data and generates work records. Using generative AI models such as OpenAI GPT-3, it is possible to highlight work progress and important work items. The generated work records are displayed in real time on the administrator's terminal.

[0468] For example, if a worker says, "I will proceed to the next step," the speech recognition system converts this into text, and the generative AI system records "Step in progress: Proceeding to the next step" in the work log. This information is displayed on the terminal so that the administrator can check it immediately.

[0469] Examples of prompt messages include the following:

[0470] "Voice input: 'Proceed to the next step.'"

[0471] "Prompt: 'Update the work log and record that you are proceeding to the next step.'"

[0472] In this way, real-time work recording and progress management are achieved at the work site.

[0473] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0474] Step 1:

[0475] The user speaks at the work site. The voice input device captures the user's speech and sends it to the server as audio data. The input is the user's voice, and the output is audio data.

[0476] Step 2:

[0477] The server converts received audio data into text data using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to analyze the audio data and generate the corresponding text. The input is audio data, and the output is text data.

[0478] Step 3:

[0479] The server analyzes text data using generative AI methods and generates work records. Using generative AI models such as OpenAI GPT-3, it extracts work progress and important work items from the text data and creates a record. The input is text data, and the output is a work record.

[0480] Step 4:

[0481] The server sends the generated work log to the administrator's terminal. The administrator's terminal displays the received work log in real time, allowing them to check the progress of the work. The input is the work log, and the output is the display on the administrator's terminal.

[0482] Step 5:

[0483] The administrator monitors the progress of work based on the work records displayed on the terminal and issues instructions as needed. This enables efficient work management. The input is the work records displayed on the terminal, and the output is the administrator's judgment and instructions.

[0484] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0485] "Example of form 1"

[0486] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[0487] "Example of form 2"

[0488] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[0489] The following describes the processing flow for each example of the form.

[0490] "Example of form 1"

[0491] Step 1: Meeting participants speak. These speeches are entered into the system as audio data.

[0492] Step 2: The speech recognition system converts the spoken audio data into text data.

[0493] Step 3: The emotion engine recognizes emotions from the audio data of the speech. This emotion is input into the system as emotion information.

[0494] Step 4: The generative AI receives text data and emotional information as input.

[0495] Step 5: The generative AI creates meeting minutes from the text data, highlighting key points and summaries of the discussion.

[0496] Step 6: The generative AI suggests action items based on emotional information.

[0497] (Example 1)

[0498] Next, we will describe Example 1 of Form Example 1. 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".

[0499] In meetings, it is important to accurately record what participants say and create meeting minutes. However, traditional methods only record the content of what was said, making it difficult to create minutes that reflect the emotions of the participants or the progress of the meeting. Furthermore, it was difficult to appropriately grasp the emotional reactions that arose during the meeting and propose actions based on them.

[0500] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0501] In this invention, the server includes means for acquiring audio data, speech recognition means for converting the audio data into text data, generative information processing means for receiving the text data as input and automatically generating meeting minutes using natural language processing technology, and emotion recognition means for recognizing emotions from the audio data and providing that information to the generative information processing means. This makes it possible to create meeting minutes that reflect not only the content of what the participants said, but also their emotions and the progress of the meeting, thereby enabling the effective management of meetings.

[0502] "Audio data" refers to sound information acquired to record what participants say in a meeting.

[0503] "Speech recognition means" refers to technology that analyzes speech data and converts it into corresponding text data.

[0504] A "generative information processing system" is a system that receives text data as input and has the function of automatically generating meeting minutes using natural language processing technology.

[0505] "Emotion recognition means" refers to a technology that analyzes participants' emotions from audio data and provides that information to generative information processing means.

[0506] Meeting minutes are documents that record what was said, the progress of the meeting, and the feelings of the participants.

[0507] "Action items" refer to specific actions or follow-up items proposed during a meeting.

[0508] This invention is a system that efficiently records the content of discussions in meetings and automatically generates meeting minutes that reflect the emotions of the participants. The system acquires audio data, converts it into text data using speech recognition technology, and then creates meeting minutes using generative information processing technology. In addition, it analyzes the emotions of the participants using emotion recognition technology and reflects that information in the meeting minutes.

[0509] The server captures user speech as audio data through microphones installed in the conference room. This audio data is converted into text data using speech recognition technologies such as Google Speech-to-Text API or IBM Watson Speech to Text. The speech recognition technology is capable of recognizing multiple languages.

[0510] Next, the server uses natural language processing technologies such as OpenAI's GPT-3 and Google's BERT as generative information processing tools. This allows for the automatic generation of meeting minutes based on text data, as well as management of meeting progress and summarization of participants' statements.

[0511] Furthermore, the server uses emotion recognition means to analyze the participants' emotions from the audio data. The emotion engine analyzes the tone, volume, and speed of the voice and provides the emotion information to the generative information processing means. Based on this emotion information, the generative information processing means adds information about emotions to the meeting minutes and proposes action items based on the emotion information.

[0512] For example, if a user says, "I am very dissatisfied with the slow progress of this project," the emotion engine recognizes anger from the tone of the statement. The generative information processing system records in the meeting minutes that "Participants are dissatisfied with the project's progress" and proposes "Schedule a follow-up meeting regarding project progress" as an action item.

[0513] An example of a prompt message is: "Convert meeting comments to text, analyze sentiment, and create meeting minutes. Also, suggest action items based on the speaker's sentiment."

[0514] The flow of the specific processing in Example 1 will be explained using Figure 15.

[0515] Step 1:

[0516] When a user speaks in a conference room, their voice is captured by a microphone installed in the room. The server receives the audio data transmitted from the microphone. The input is the user's voice, and the output is the audio data sent to the server. This audio data is used in subsequent processing steps.

[0517] Step 2:

[0518] The server passes the received audio data to the speech recognition system. The speech recognition system uses technologies such as the Google Speech-to-Text API or IBM Watson Speech to Text to convert the audio data into text data. The input is audio data, and the output is text data. This conversion allows the audio information to be treated as text information.

[0519] Step 3:

[0520] The server inputs text data output from the speech recognition system into a generative information processing system. The generative information processing system analyzes the text data using natural language processing technologies such as OpenAI's GPT-3 and Google's BERT, and automatically generates meeting minutes. The input is text data, and the output is meeting minutes. This process is used for managing the progress of meetings and summarizing participants' statements.

[0521] Step 4:

[0522] The server passes audio data to an emotion recognition system. The emotion recognition system analyzes the tone, volume, and speed of the voice to recognize the user's emotions. The input is audio data, and the output is emotion information. This information is used to understand the emotional state of the participants.

[0523] Step 5:

[0524] The server passes the emotion information provided by the emotion recognition means to the generative information processing means. The generative information processing means adds information about emotions to the meeting minutes based on the emotion information and proposes action items based on the emotion information. The input is emotion information, and the output is meeting minutes and action items that reflect the emotions. This process makes the meeting minutes more detailed and useful.

[0525] (Application Example 1)

[0526] Next, we will describe Application Example 1 of Form Example 1. 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."

[0527] Efficiently recording worker instructions and reports and managing the progress of work is crucial in the workplace. However, conventional methods make it difficult to respond appropriately while considering workers' emotions and stress levels, resulting in insufficient improvements in work efficiency and safety.

[0528] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0529] In this invention, the server includes speech recognition means, generative information processing means, and emotion recognition means. This makes it possible to transcribe the worker's speech into text in real time, automatically generate work records, and recognize the worker's emotions to suggest appropriate actions.

[0530] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or system that converts a worker's speech into text information in real time.

[0531] "Generative information processing means" refers to technology that processes information based on text data and automatically generates work records and action items.

[0532] "Emotion recognition means" refers to a technology that analyzes the emotions of a worker from voice data and recognizes their emotional state.

[0533] A "work record" is a document that records the progress and important points of a task, generated based on the worker's spoken content.

[0534] "Action items" are specific actions or countermeasures proposed based on the worker's emotional state and the nature of their work.

[0535] The system for carrying out this invention includes speech recognition means, generative information processing means, and emotion recognition means. The server uses the Google Cloud Speech-to-Text API as the speech recognition means to convert the worker's voice into text data in real time. The generative information processing means uses an OpenAI generative AI model to automatically generate work records based on the text data and propose action items. The emotion recognition means uses IBM Watson Tone Analyzer to analyze the worker's emotions from the voice data and provides that information to the generative information processing means.

[0536] The system uses smartphones or tablets as terminals, connecting a microphone to collect voice data. Users input instructions and reports at the work site using voice, and the content is automatically converted to text and saved as a work record.

[0537] For example, if a worker says, "This task is difficult," the emotion recognition system recognizes stress, and the generative information processing system generates the action item, "The worker may be experiencing stress. We suggest a break."

[0538] An example of a prompt message is: "Transcribe the worker's statements into text, analyze their emotions, and suggest necessary actions."

[0539] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[0540] Step 1:

[0541] The terminal collects the worker's voice through a microphone. The input is the worker's voice data, and the output is that voice data. The terminal sends this voice data to the server.

[0542] Step 2:

[0543] The server uses the Google Cloud Speech-to-Text API as a speech recognition tool to convert received audio data into text data. The input is audio data, and the output is text data. This conversion allows the operator's speech to be obtained as text information.

[0544] Step 3:

[0545] The server uses IBM Watson Tone Analyzer as an emotion recognition tool to analyze the worker's emotions from text data. The input is text data, and the output is emotion information. The server provides this emotion information to a generative information processing system.

[0546] Step 4:

[0547] The server uses OpenAI's generative AI model as a generative information processing tool to automatically generate work records based on text data and sentiment information. The input is text data and sentiment information, and the output is the work record. The server saves this work record and suggests action items as needed.

[0548] Step 5:

[0549] The user reviews the generated work records and action items on their terminal. The input consists of work records and action items, while the output is the information the user reviews. Based on this, the user can consider the progress of the work and potential countermeasures.

[0550] (Example 2)

[0551] Next, we will describe Example 2 of Form Example 2. 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".

[0552] In meetings, it is difficult to accurately record what participants say and to create minutes that reflect their emotional state. Furthermore, while it is necessary to quickly grasp important points and action items after a meeting, traditional methods are time-consuming and labor-intensive.

[0553] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0554] In this invention, the server includes means for acquiring audio data, means for converting audio data into text information, and means for recognizing emotional information. This makes it possible to accurately record the content of meetings and create meeting minutes that reflect the emotional state of the participants.

[0555] "Means for acquiring audio data" refers to a device or method for collecting the speech of meeting participants in real time.

[0556] "Means for converting audio data into text information" refers to a technology or device for analyzing acquired audio data and converting it into a corresponding string of characters.

[0557] "Means for analyzing textual information to generate meeting records" refers to a technology or device that uses converted textual information to organize the content of a meeting and record important points and the flow of discussion.

[0558] "Means for recognizing emotional information" refers to technologies or devices for analyzing and identifying emotional states from participants' voices.

[0559] "Methods for revising meeting records based on emotional information to emphasize important information and action items" refers to technologies or devices that utilize recognized emotional information to add emotional nuances to meeting records and clarify important information and next steps to take.

[0560] This invention is a system that accurately records the content of speeches and the emotional state of participants during meetings, and generates effective meeting minutes. When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time.

[0561] When the server receives audio data, it uses speech recognition software to convert the audio data into text. Specifically, a common speech recognition API can be used for speech recognition. This conversion records the speeches during the meeting in text format.

[0562] Next, the server uses a generative AI model to analyze the converted text information and generate a meeting record. This generative AI model can utilize natural language processing technology. The generative AI model extracts important points from the text information and organizes the flow of the discussion.

[0563] Furthermore, the server uses an emotion recognition engine to recognize emotional information from the audio data. The emotion recognition engine analyzes the tone, volume, and speed of the speech to determine the emotional state of the participants. This allows the server to grasp the emotional nuances contained in what is said during the meeting.

[0564] The server modifies the generated meeting minutes based on recognized sentiment information, highlighting important information and action items. Specifically, it uses sentiment information to add information to the minutes such as "the speaker may be feeling dissatisfied." It can also suggest action items based on sentiment information, such as "follow up on the speaker's dissatisfaction."

[0565] An example of a prompt might be, "Please create meeting minutes. Based on the following text data and sentiment information, highlight key points, summaries of discussions, and action items." This system ensures that meeting minutes are not merely a record of what was said, but also reflect the emotional state of the participants, enabling more effective meeting management.

[0566] The flow of the specific processing in Example 2 will be explained using Figure 17.

[0567] Step 1:

[0568] When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time. The input is the participants' speech, and the output is the audio data sent to the server. Specifically, the microphone captures the audio and transmits the data to the server over the network.

[0569] Step 2:

[0570] The server passes the received audio data to the speech recognition software. The speech recognition software converts the audio data into text information. The input is audio data, and the output is text information. As a data processing step, the audio signal is analyzed and the corresponding string is generated. Specifically, the speech recognition API analyzes the audio data and converts it into text format.

[0571] Step 3:

[0572] The server inputs text information into a generative AI model. The generative AI model analyzes the text information and generates meeting minutes. The input is text information, and the output is meeting minutes. As a data calculation, it extracts important points from the text information and organizes the flow of the discussion. Specifically, it utilizes natural language processing techniques to summarize the text data and create meeting minutes.

[0573] Step 4:

[0574] The server passes the audio data to the emotion recognition engine. The emotion recognition engine recognizes emotional information from the audio. The input is audio data, and the output is emotional information. As part of the data processing, the tone, volume, and speed of the voice are analyzed to determine the emotional state. Specifically, an emotion analysis tool analyzes the audio data and identifies emotional nuances.

[0575] Step 5:

[0576] The server provides emotional information to the generative AI model. The generative AI model modifies the meeting minutes based on the emotional information, highlighting important information and action items. The input is emotional information and the meeting minutes, and the output is the modified meeting minutes. As a data calculation, emotional information is used to add emotional nuances to the minutes and clarify action items. Specifically, the generative AI model analyzes the emotional information and adds information to the minutes such as "the speaker may be feeling dissatisfied."

[0577] Step 6:

[0578] The server provides the user with the finalized meeting transcript. The user can view the meeting transcript through their terminal and understand the content of the meeting. The input is the revised meeting transcript, and the output is the meeting transcript provided to the user. Specifically, the server sends the meeting transcript to the user's terminal, and the user views it.

[0579] (Application Example 2)

[0580] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0581] In a work environment, it is crucial to accurately record what workers say and create meeting minutes in real time. However, conventional systems have difficulty considering workers' emotions when creating meeting minutes or suggesting action items, limiting improvements in work efficiency. Furthermore, they are unable to properly recognize workers' emotions and follow up accordingly, thus necessitating improvements to the work environment.

[0582] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0583] In this invention, the server includes speech recognition means, generative AI means, and emotion recognition means. This enables the conversion of a worker's speech into text via speech recognition in the work environment, the generative AI to create meeting minutes in real time, the emotion recognition means to recognize the worker's emotions, and the suggestion of action items based on that information.

[0584] "Speech recognition means" refers to a technology that converts speech data into text data, making it possible to process spoken content in a digital format.

[0585] "Generative AI methods" refer to artificial intelligence technologies that use input text data to create meeting minutes, extract key points, and suggest action items.

[0586] "Emotion recognition means" refers to technology that analyzes the speaker's emotions from audio data and recognizes their emotional state.

[0587] "Meeting minutes" are documents that record the content of conversations during meetings or work sessions and compile them in a format that can be referenced later.

[0588] An "action item" is an item that indicates an issue identified in meeting minutes or the work environment, or the next action to be taken.

[0589] "Work environment" refers to the location and conditions in which a specific task is performed, such as a factory or office.

[0590] "Real-time" means that data processing and information provision occur immediately, resulting in a state where results are obtained without delay.

[0591] The system for carrying out this invention is centered around a server that includes speech recognition means, generative AI means, and emotion recognition means. The server receives the speech of workers as audio data through microphones installed in the work environment, such as a factory or office. The speech recognition means converts this audio data into text data.

[0592] The generative AI system creates meeting minutes in real time based on the converted text data, extracting key points and action items. Furthermore, the emotion recognition system analyzes the emotions of the participants from the audio data and provides this information to the generative AI system. This enables the generative AI system to create meeting minutes and suggest action items that take emotional information into account.

[0593] As a concrete example, during a morning meeting in a factory, when a worker expresses their opinion on a new work procedure, the server records their statement and, if the worker is feeling anxious, suggests an action item such as "Provide additional explanation to alleviate the worker's anxiety."

[0594] An example of a prompt message would be: "Analyze the worker's feelings based on this statement and reflect it in the meeting minutes. If the worker is feeling anxious, suggest a follow-up."

[0595] This system combines speech recognition using the Google Cloud Speech-to-Text API, generative AI using OpenAI's GPT model, and emotion recognition using Microsoft Azure's Emotion API. This enables more efficient communication in the work environment and allows for the creation of meeting minutes that take into account the emotions of the workers.

[0596] The flow of a specific process in Application Example 2 will be explained using Figure 18.

[0597] Step 1:

[0598] The server receives the worker's speech as audio data through microphones installed in the work environment. This audio data becomes the input. The server collects the audio data and prepares it for the next processing step.

[0599] Step 2:

[0600] The server uses speech recognition to convert the received audio data into text data. This process uses the Google Cloud Speech-to-Text API to analyze the audio data and output text data. The converted text data becomes the input for the next step.

[0601] Step 3:

[0602] The server uses generative AI methods to create meeting minutes in real time based on text data. This process utilizes OpenAI's GPT model to extract important points and action items from the text data. The generated meeting minutes are then output.

[0603] Step 4:

[0604] The server uses emotion recognition to analyze the worker's emotions from the audio data. This process uses the Microsoft Azure Emotion API to analyze the audio data and output emotion information. This emotion information then serves as input for the next step.

[0605] Step 5:

[0606] The server provides emotional information to a generative AI system, which then reflects this emotional information in the meeting minutes. This process adds emotion-based comments and action items to the minutes based on the emotional information. The final meeting minutes are then output.

[0607] Step 6:

[0608] Users review the final meeting minutes provided by the server and perform action items as needed. They then refer to the minutes to improve their work environment and follow up.

[0609] (Other examples)

[0610] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[0611] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0612] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0613] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[0614] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0615] [Third Embodiment]

[0616] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0617] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0618] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0619] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0620] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0621] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0622] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0623] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0624] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0625] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0626] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0627] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0628] "Example of form 1"

[0629] In one embodiment of the present invention, a speech recognition module converts audio data input from a microphone installed in a conference room into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages. Next, a generative AI module receives the text data output from the speech recognition module as input and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' statements. "Embodiment Example 2"

[0630] As a concrete example of its use, when a meeting begins, microphones installed in the meeting room capture participants' speech, and a speech recognition module converts this into text data. A generative AI module analyzes this text data in real time and generates meeting minutes. When the meeting ends, the generative AI module creates a final version of the minutes, highlighting key points, summaries of discussions, and suggested action items. This makes it easier to organize information after the meeting and improves meeting productivity.

[0631] The following describes the processing flow for each example of the form.

[0632] "Example of form 1"

[0633] Step 1: Acquire audio data from the microphone installed in the conference room.

[0634] Step 2: The acquired audio data is input into the speech recognition module and converted into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages.

[0635] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0636] Step 4: The generative AI module analyzes the text data and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' remarks.

[0637] "Example of form 2"

[0638] Step 1: Once the meeting begins, microphones installed in the meeting room capture the participants' speech.

[0639] Step 2: Input the captured audio data into the speech recognition module and convert it into text data.

[0640] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0641] Step 4: The generative AI module analyzes the text data in real time and generates meeting minutes.

[0642] Step 5: Once the meeting concludes, the generative AI module creates a final version of the meeting minutes, highlighting key points, summaries of discussions, and suggested action items. This facilitates post-meeting information organization and improves meeting productivity.

[0643] (Example 1)

[0644] Next, we will describe Embodiment 1 of Example 1. 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."

[0645] There is a need to efficiently record meeting content and provide an environment where participants can focus on the discussion. However, traditional methods are time-consuming to create meeting minutes, and there is a risk of overlooking important points and action items. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder minute-taking.

[0646] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0647] In this invention, the server includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, and a generative information processing means for analyzing the text data and automatically generating meeting minutes. This makes it possible to quickly and accurately record the contents of a meeting, create meeting minutes in multiple languages ​​without overlooking important points or action items.

[0648] "Audio data" refers to information that represents the speech of meeting participants in a digital format.

[0649] An "input device" is a device used to acquire audio data, and includes microphones and the like.

[0650] "Speech recognition means" refers to a technology or device for converting speech data into text data.

[0651] "Text data" refers to character information converted by speech recognition technology.

[0652] "Generative information processing means" refers to a technology or device for analyzing text data and automatically generating meeting minutes.

[0653] An "output device" is a device used to display or print the generated meeting minutes.

[0654] Meeting minutes are documents that record the content of a meeting, including important points and action items.

[0655] "Multiple languages" refers to a group of languages ​​with different linguistic systems, and means that speech recognition means are capable of recognizing them.

[0656] A description of embodiments for carrying out this invention will be given.

[0657] System Overview

[0658] This system is designed to efficiently record meeting content and provide an environment where participants can concentrate on the discussion. The system includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, a generative information processing means for analyzing the text data and automatically generating meeting minutes, and an output device for outputting the generated meeting minutes.

[0659] Hardware and software to be used

[0660] The microphone installed in the conference room will be used as the input device.

[0661] The speech recognition method will utilize deep learning-based speech recognition technology. Specifically, speech recognition software such as the Google Speech-to-Text API can be used.

[0662] The generative information processing system employs natural language processing technology. Specifically, generative AI models such as OpenAI's GPT-3 can be used.

[0663] The terminal's display or printer is used as the output device to display or print the generated meeting minutes.

[0664] Specific example

[0665] When a user says "Let's move on to the next agenda item" during a meeting, the microphone connected to the device picks up this audio.

[0666] The server uses speech recognition to convert the speech into text data that says, "Let's move on to the next topic."

[0667] The generative information processing system automatically generates meeting minutes, such as "The meeting has moved on to the next agenda item," based on this text data.

[0668] The server sends the generated meeting minutes to the terminal, and the user can view the minutes on the terminal.

[0669] Example of a prompt

[0670] "Please summarize the meeting content and create meeting minutes."

[0671] This system enables the rapid and accurate recording of meeting content, ensuring that important points and action items are not overlooked, and allows for the creation of meeting minutes in multiple languages.

[0672] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0673] Step 1:

[0674] When a user speaks in the conference room, a microphone connected to the terminal captures their voice. The input is the user's voice, and the output is digital audio data. The microphone converts the audio into an electrical signal, and the terminal prepares this to be sent to the server as digital data.

[0675] Step 2:

[0676] The server receives audio data transmitted from the terminal. The input is digital audio data, and the output is text data. The server activates speech recognition and converts the audio data into text data. In this process, speech recognition technology is used to analyze the waveform of the audio and generate the corresponding string of characters.

[0677] Step 3:

[0678] The server inputs text data obtained from the speech recognition means into the generative information processing means. The input is text data, and the output is the text of the meeting minutes. The generative information processing means analyzes the text data using natural language processing technology and automatically generates the meeting minutes. In this process, the key points of the text are extracted and the meeting minutes are constructed in a format that follows the progress of the meeting.

[0679] Step 4:

[0680] The server sends the generated meeting minutes to the terminal. The input is the text of the meeting minutes, and the output is the meeting minutes in a format viewable by the user. The terminal displays the received meeting minutes on its screen and shares them as needed, such as by printing or emailing. Users can review the meeting minutes on their terminal and reflect on the content of the meeting.

[0681] This series of processes makes it possible to record the contents of a meeting quickly and accurately, and to provide an environment where participants can concentrate on the discussion.

[0682] (Application Example 1)

[0683] Next, we will describe Application Example 1 of Form Example 1. 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."

[0684] Manual record-keeping by workers for work reporting and progress management within a factory is time-consuming and labor-intensive, and presents challenges in terms of accuracy and efficiency. Furthermore, real-time monitoring of work status is difficult, making it challenging for managers to respond quickly.

[0685] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0686] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, using that as input for the generative AI to create a work record, and automatically generating logs for work progress management and quality control. This enables the automation of work reporting and real-time monitoring of work status.

[0687] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or software for accurately acquiring a worker's speech as textual information.

[0688] "Generative AI methods" refer to artificial intelligence technologies that use natural language processing techniques to automatically generate work records and progress management information based on input text data.

[0689] "Work records" are data that document the activities and progress of workers within a factory, and are useful information for improving work efficiency and quality control.

[0690] "Progress management" is a management method used to understand the progress of work and to confirm whether the work is proceeding according to plan.

[0691] "Quality control" refers to management activities carried out to maintain a consistent quality of products and services, and is a process for ensuring the accuracy and efficiency of work.

[0692] A "log" is data that shows the history and records of work, and is information used to review and analyze the work content later.

[0693] The system for carrying out this invention includes speech recognition means, generative AI means, and automatic work record generation means. The server uses speech recognition technology such as the Google Speech-to-Text API as the speech recognition means to convert the worker's speech into text data in real time. The converted text data is analyzed using natural language processing technology such as OpenAI GPT-3 as the generative AI means to automatically generate work records and progress management information.

[0694] The terminal allows workers to submit voice reports via an application installed on devices such as smartphones and tablets. Users report their work details by voice through the terminal, and this voice data is sent to a server. The server converts the voice data into text and generates work records using generative AI. The generated work records are stored in a cloud-based database and can be accessed by administrators in real time.

[0695] For example, if a worker reports "Line 1 maintenance complete, no abnormalities," the server will generate a work log such as "October 10, 2023, 14:30 Line 1 maintenance complete, no abnormalities." An example of a prompt to input to the generation AI model in this case would be, "Convert the work report into a log in the following format: Date and time, Line number, Work details, Result."

[0696] This system enables the automation of work reporting and real-time monitoring of work status, thereby improving work efficiency within the factory.

[0697] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0698] Step 1:

[0699] The user uses a terminal to report their work details verbally. The terminal acquires the user's voice data through a microphone and sends that data to a server. The input is the user's voice data, and the output is the transmission of voice data to the server.

[0700] Step 2:

[0701] The server converts received audio data into text data using speech recognition. Specifically, it analyzes the audio using the Google Speech-to-Text API and generates the corresponding text. The input is audio data, and the output is text data.

[0702] Step 3:

[0703] The server analyzes text data using generative AI methods and automatically generates work records. Using natural language processing technologies such as OpenAI GPT-3, it creates records that include work progress and results based on the text data. The input is text data, and the output is a work record.

[0704] Step 4:

[0705] The server saves the generated work records to a database in the cloud. This allows administrators to monitor the work status in real time. The input is the work records, and the output is the saving of the records to the database.

[0706] Step 5:

[0707] Administrators access a database in the cloud to review work records. This allows for efficient progress and quality control of work. Input is a request to access the database, and output is a display of work records.

[0708] (Example 2)

[0709] Next, we will describe Example 2 of the morphological example. 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."

[0710] In meetings, accurately recording participants' remarks and efficiently organizing important information is crucial for improving meeting productivity. However, manual minute-taking is time-consuming and labor-intensive, and can lead to omissions and errors. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder the accurate transmission of information. To address these challenges, a system is needed that automatically transcribes audio data into text and extracts and organizes key information.

[0711] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0712] In this invention, the server includes means for acquiring audio data, means for converting the acquired audio data into text data, and means for analyzing the text data and generating meeting minutes. This makes it possible to record what is said during a meeting in real time and to efficiently organize important information.

[0713] "Means for acquiring audio data" refers to a device or method for capturing the speech of participants during a meeting in real time and saving it in digital format.

[0714] "Means for converting acquired audio data into text data" refers to an apparatus or method that performs the process of converting audio data into text information using speech recognition technology.

[0715] "Means for analyzing text data and generating meeting minutes" refers to a device or method that uses natural language processing technology to extract important information from text data, organize the content of a meeting, and output it as meeting minutes.

[0716] "Means for formatting generated meeting minutes into a final version and highlighting important information" refers to a device or method for organizing generated meeting minutes in an easy-to-read format and making particularly important points or action items stand out.

[0717] "Means for users to review and modify meeting minutes" refers to an interface or method that allows users to view the generated meeting minutes and modify or add to their content as needed.

[0718] This invention is a system that automatically records speeches in meetings and efficiently organizes important information. A specific embodiment of this system is described below.

[0719] The server uses audio input devices installed in the conference room to capture participants' speech in real time. Specifically, it uses multiple microphones as a typical audio input device to capture the audio signals as digital data.

[0720] Next, the server converts the acquired audio data into text data using speech recognition software. A general-purpose platform providing speech recognition technology can be used as this speech recognition software. This converts the audio data into text information.

[0721] Subsequently, the server analyzes the text data using a generative AI model and generates meeting minutes. This generative AI model can be a general-purpose generative AI platform utilizing natural language processing technology. The generated meeting minutes are formatted to highlight important points and action items.

[0722] Users can review the meeting minutes generated through their device and make corrections or additions as needed. This interface is designed to allow users to easily edit the meeting minutes.

[0723] For example, if the "new product launch plan" is discussed during a meeting, the server captures the statement, "The new product launch is scheduled for next month. We will decide on the marketing strategy in detail at the next meeting," and speech recognition software converts this to text. A generative AI model analyzes this text and generates meeting minutes like the following:

[0724] Key point: The new product is scheduled to be released next month.

[0725] Summary of discussion: Details regarding the marketing strategy will be decided at the next meeting.

[0726] Action item: Finalize the marketing strategy details at the next meeting.

[0727] An example of a prompt to input into a generative AI model might be, "Generate meeting minutes. Include key points, a summary of the discussion, and action items."

[0728] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0729] Step 1:

[0730] The server uses audio input devices installed in the conference room to capture participants' speech in real time. The input is the audio signal during the meeting, and the output is digital audio data. Specifically, it uses multiple microphones to capture the audio signal as digital data.

[0731] Step 2:

[0732] The server inputs the acquired audio data into speech recognition software and converts it into text data. The input is digital audio data, and the output is text data. Specifically, it performs a process of converting audio data into text information using speech recognition technology.

[0733] Step 3:

[0734] The server inputs text data into a generative AI model and generates meeting minutes. The input is text data, and the output is the generated meeting minutes. Specifically, it uses natural language processing techniques to analyze the text data, extract important information, and create the meeting minutes.

[0735] Step 4:

[0736] The server formats the generated meeting minutes as the final version and highlights important information. The input is the generated meeting minutes, and the output is the formatted, final version of the meeting minutes. Specifically, it organizes the content of the meeting minutes for easier reading and highlights particularly important points and action items.

[0737] Step 5:

[0738] The user reviews the meeting minutes generated through the terminal and makes corrections or additions as needed. The input is the formatted final version of the meeting minutes, and the output is the meeting minutes modified by the user. Specifically, the user views the meeting minutes and edits the content through the interface.

[0739] (Application Example 2)

[0740] Next, we will describe application example 2 of form example 2. 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."

[0741] In factory and other work environments, there is a need to efficiently record worker instructions and reports and to grasp the progress of work in real time. However, traditional methods require manual recording and management, which leads to decreased work efficiency. Furthermore, it is difficult to immediately grasp important points and progress of work, increasing the burden on managers.

[0742] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0743] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, and for the generative AI to create a work record using that as input, highlighting the progress of the work and important work items. This enables real-time work recording and progress management at the work site.

[0744] "Speech recognition means" refers to a technology that converts speech into text data, and is a device or software for recording a worker's speech in digital format.

[0745] "Generative AI methods" are artificial intelligence technologies that analyze input text data and generate information tailored to specific purposes, such as systems for automatically creating work records and progress highlights.

[0746] A "work record" is a text-based record of instructions and reports from workers at a work site, and serves as data for understanding the progress of work and important work items.

[0747] "Progress status" refers to information indicating how far a task has progressed, and is an essential indicator for efficient task management and planning.

[0748] "Important work items" refer to points in a task that require particular attention, or factors that significantly influence the success or failure of the task, and are information that managers should prioritize understanding.

[0749] The system for carrying out this invention is for recording workers' speech in real time and managing the progress of work in a workplace such as a factory. The server includes speech recognition means, generative AI means, and means for creating work records.

[0750] The speech recognition system captures the worker's speech and converts the audio data into text data. Specifically, speech recognition software such as the Google Speech-to-Text API can be used. The converted text data is sent to the server.

[0751] The generative AI system analyzes received text data and generates work records. Using generative AI models such as OpenAI GPT-3, it is possible to highlight work progress and important work items. The generated work records are displayed in real time on the administrator's terminal.

[0752] For example, if a worker says, "I will proceed to the next step," the speech recognition system converts this into text, and the generative AI system records "Step in progress: Proceeding to the next step" in the work log. This information is displayed on the terminal so that the administrator can check it immediately.

[0753] Examples of prompt messages include the following:

[0754] "Voice input: 'Proceed to the next step.'"

[0755] "Prompt: 'Update the work log and record that you are proceeding to the next step.'"

[0756] In this way, real-time work recording and progress management are achieved at the work site.

[0757] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0758] Step 1:

[0759] The user speaks at the work site. The voice input device captures the user's speech and sends it to the server as audio data. The input is the user's voice, and the output is audio data.

[0760] Step 2:

[0761] The server converts received audio data into text data using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to analyze the audio data and generate the corresponding text. The input is audio data, and the output is text data.

[0762] Step 3:

[0763] The server analyzes text data using generative AI methods and generates work records. Using generative AI models such as OpenAI GPT-3, it extracts work progress and important work items from the text data and creates a record. The input is text data, and the output is a work record.

[0764] Step 4:

[0765] The server sends the generated work log to the administrator's terminal. The administrator's terminal displays the received work log in real time, allowing them to check the progress of the work. The input is the work log, and the output is the display on the administrator's terminal.

[0766] Step 5:

[0767] The administrator monitors the progress of work based on the work records displayed on the terminal and issues instructions as needed. This enables efficient work management. The input is the work records displayed on the terminal, and the output is the administrator's judgment and instructions.

[0768] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0769] "Example of form 1"

[0770] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[0771] "Example of form 2"

[0772] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[0773] The following describes the processing flow for each example of the form.

[0774] "Example of form 1"

[0775] Step 1: Meeting participants speak. These speeches are entered into the system as audio data.

[0776] Step 2: The speech recognition system converts the spoken audio data into text data.

[0777] Step 3: The emotion engine recognizes emotions from the audio data of the speech. This emotion is input into the system as emotion information.

[0778] Step 4: The generative AI receives text data and emotional information as input.

[0779] Step 5: The generative AI creates meeting minutes from the text data, highlighting key points and summaries of the discussion.

[0780] Step 6: The generative AI suggests action items based on emotional information.

[0781] (Example 1)

[0782] Next, we will describe Embodiment 1 of Example 1. 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."

[0783] In meetings, it is important to accurately record what participants say and create meeting minutes. However, traditional methods only record the content of what was said, making it difficult to create minutes that reflect the emotions of the participants or the progress of the meeting. Furthermore, it was difficult to appropriately grasp the emotional reactions that arose during the meeting and propose actions based on them.

[0784] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0785] In this invention, the server includes means for acquiring audio data, speech recognition means for converting the audio data into text data, generative information processing means for receiving the text data as input and automatically generating meeting minutes using natural language processing technology, and emotion recognition means for recognizing emotions from the audio data and providing that information to the generative information processing means. This makes it possible to create meeting minutes that reflect not only the content of what the participants said, but also their emotions and the progress of the meeting, thereby enabling the effective management of meetings.

[0786] "Audio data" refers to sound information acquired to record what participants say in a meeting.

[0787] "Speech recognition means" refers to technology that analyzes speech data and converts it into corresponding text data.

[0788] A "generative information processing system" is a system that receives text data as input and has the function of automatically generating meeting minutes using natural language processing technology.

[0789] "Emotion recognition means" refers to a technology that analyzes participants' emotions from audio data and provides that information to generative information processing means.

[0790] Meeting minutes are documents that record what was said, the progress of the meeting, and the feelings of the participants.

[0791] "Action items" refer to specific actions or follow-up items proposed during a meeting.

[0792] This invention is a system that efficiently records the content of discussions in meetings and automatically generates meeting minutes that reflect the emotions of the participants. The system acquires audio data, converts it into text data using speech recognition technology, and then creates meeting minutes using generative information processing technology. In addition, it analyzes the emotions of the participants using emotion recognition technology and reflects that information in the meeting minutes.

[0793] The server captures user speech as audio data through microphones installed in the conference room. This audio data is converted into text data using speech recognition technologies such as Google Speech-to-Text API or IBM Watson Speech to Text. The speech recognition technology is capable of recognizing multiple languages.

[0794] Next, the server uses natural language processing technologies such as OpenAI's GPT-3 and Google's BERT as generative information processing tools. This allows for the automatic generation of meeting minutes based on text data, as well as management of meeting progress and summarization of participants' statements.

[0795] Furthermore, the server uses emotion recognition means to analyze the participants' emotions from the audio data. The emotion engine analyzes the tone, volume, and speed of the voice and provides the emotion information to the generative information processing means. Based on this emotion information, the generative information processing means adds information about emotions to the meeting minutes and proposes action items based on the emotion information.

[0796] For example, if a user says, "I am very dissatisfied with the slow progress of this project," the emotion engine recognizes anger from the tone of the statement. The generative information processing system records in the meeting minutes that "Participants are dissatisfied with the project's progress" and proposes "Schedule a follow-up meeting regarding project progress" as an action item.

[0797] An example of a prompt message is: "Convert meeting comments to text, analyze sentiment, and create meeting minutes. Also, suggest action items based on the speaker's sentiment."

[0798] The flow of the specific processing in Example 1 will be explained using Figure 15.

[0799] Step 1:

[0800] When a user speaks in a conference room, their voice is captured by a microphone installed in the room. The server receives the audio data transmitted from the microphone. The input is the user's voice, and the output is the audio data sent to the server. This audio data is used in subsequent processing steps.

[0801] Step 2:

[0802] The server passes the received audio data to the speech recognition system. The speech recognition system uses technologies such as the Google Speech-to-Text API or IBM Watson Speech to Text to convert the audio data into text data. The input is audio data, and the output is text data. This conversion allows the audio information to be treated as text information.

[0803] Step 3:

[0804] The server inputs text data output from the speech recognition system into a generative information processing system. The generative information processing system analyzes the text data using natural language processing technologies such as OpenAI's GPT-3 and Google's BERT, and automatically generates meeting minutes. The input is text data, and the output is meeting minutes. This process is used for managing the progress of meetings and summarizing participants' statements.

[0805] Step 4:

[0806] The server passes audio data to an emotion recognition system. The emotion recognition system analyzes the tone, volume, and speed of the voice to recognize the user's emotions. The input is audio data, and the output is emotion information. This information is used to understand the emotional state of the participants.

[0807] Step 5:

[0808] The server passes the emotion information provided by the emotion recognition means to the generative information processing means. The generative information processing means adds information about emotions to the meeting minutes based on the emotion information and proposes action items based on the emotion information. The input is emotion information, and the output is meeting minutes and action items that reflect the emotions. This process makes the meeting minutes more detailed and useful.

[0809] (Application Example 1)

[0810] Next, we will describe Application Example 1 of Form Example 1. 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."

[0811] Efficiently recording worker instructions and reports and managing the progress of work is crucial in the workplace. However, conventional methods make it difficult to respond appropriately while considering workers' emotions and stress levels, resulting in insufficient improvements in work efficiency and safety.

[0812] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0813] In this invention, the server includes speech recognition means, generative information processing means, and emotion recognition means. This makes it possible to transcribe the worker's speech into text in real time, automatically generate work records, and recognize the worker's emotions to suggest appropriate actions.

[0814] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or system that converts a worker's speech into text information in real time.

[0815] "Generative information processing means" refers to technology that processes information based on text data and automatically generates work records and action items.

[0816] "Emotion recognition means" refers to a technology that analyzes the emotions of a worker from voice data and recognizes their emotional state.

[0817] A "work record" is a document that records the progress and important points of a task, generated based on the worker's spoken content.

[0818] "Action items" are specific actions or countermeasures proposed based on the worker's emotional state and the nature of their work.

[0819] The system for carrying out this invention includes speech recognition means, generative information processing means, and emotion recognition means. The server uses the Google Cloud Speech-to-Text API as the speech recognition means to convert the worker's voice into text data in real time. The generative information processing means uses an OpenAI generative AI model to automatically generate work records based on the text data and propose action items. The emotion recognition means uses IBM Watson Tone Analyzer to analyze the worker's emotions from the voice data and provides that information to the generative information processing means.

[0820] The system uses smartphones or tablets as terminals, connecting a microphone to collect voice data. Users input instructions and reports at the work site using voice, and the content is automatically converted to text and saved as a work record.

[0821] For example, if a worker says, "This task is difficult," the emotion recognition system recognizes stress, and the generative information processing system generates the action item, "The worker may be experiencing stress. We suggest a break."

[0822] An example of a prompt message is: "Transcribe the worker's statements into text, analyze their emotions, and suggest necessary actions."

[0823] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[0824] Step 1:

[0825] The terminal collects the worker's voice through a microphone. The input is the worker's voice data, and the output is that voice data. The terminal sends this voice data to the server.

[0826] Step 2:

[0827] The server uses the Google Cloud Speech-to-Text API as a speech recognition tool to convert received audio data into text data. The input is audio data, and the output is text data. This conversion allows the operator's speech to be obtained as text information.

[0828] Step 3:

[0829] The server uses IBM Watson Tone Analyzer as an emotion recognition tool to analyze the worker's emotions from text data. The input is text data, and the output is emotion information. The server provides this emotion information to a generative information processing system.

[0830] Step 4:

[0831] The server uses OpenAI's generative AI model as a generative information processing tool to automatically generate work records based on text data and sentiment information. The input is text data and sentiment information, and the output is the work record. The server saves this work record and suggests action items as needed.

[0832] Step 5:

[0833] The user reviews the generated work records and action items on their terminal. The input consists of work records and action items, while the output is the information the user reviews. Based on this, the user can consider the progress of the work and potential countermeasures.

[0834] (Example 2)

[0835] Next, we will describe Example 2 of the morphological example. 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."

[0836] In meetings, it is difficult to accurately record what participants say and to create minutes that reflect their emotional state. Furthermore, while it is necessary to quickly grasp important points and action items after a meeting, traditional methods are time-consuming and labor-intensive.

[0837] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0838] In this invention, the server includes means for acquiring audio data, means for converting audio data into text information, and means for recognizing emotional information. This makes it possible to accurately record the content of meetings and create meeting minutes that reflect the emotional state of the participants.

[0839] "Means for acquiring audio data" refers to a device or method for collecting the speech of meeting participants in real time.

[0840] "Means for converting audio data into text information" refers to a technology or device for analyzing acquired audio data and converting it into a corresponding string of characters.

[0841] "Means for analyzing textual information to generate meeting records" refers to a technology or device that uses converted textual information to organize the content of a meeting and record important points and the flow of discussion.

[0842] "Means for recognizing emotional information" refers to technologies or devices for analyzing and identifying emotional states from participants' voices.

[0843] "Methods for revising meeting records based on emotional information to emphasize important information and action items" refers to technologies or devices that utilize recognized emotional information to add emotional nuances to meeting records and clarify important information and next steps to take.

[0844] This invention is a system that accurately records the content of speeches and the emotional state of participants during meetings, and generates effective meeting minutes. When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time.

[0845] When the server receives audio data, it uses speech recognition software to convert the audio data into text. Specifically, a common speech recognition API can be used for speech recognition. This conversion records the speeches during the meeting in text format.

[0846] Next, the server uses a generative AI model to analyze the converted text information and generate a meeting record. This generative AI model can utilize natural language processing technology. The generative AI model extracts important points from the text information and organizes the flow of the discussion.

[0847] Furthermore, the server uses an emotion recognition engine to recognize emotional information from the audio data. The emotion recognition engine analyzes the tone, volume, and speed of the speech to determine the emotional state of the participants. This allows the server to grasp the emotional nuances contained in what is said during the meeting.

[0848] The server modifies the generated meeting minutes based on recognized sentiment information, highlighting important information and action items. Specifically, it uses sentiment information to add information to the minutes such as "the speaker may be feeling dissatisfied." It can also suggest action items based on sentiment information, such as "follow up on the speaker's dissatisfaction."

[0849] An example of a prompt might be, "Please create meeting minutes. Based on the following text data and sentiment information, highlight key points, summaries of discussions, and action items." This system ensures that meeting minutes are not merely a record of what was said, but also reflect the emotional state of the participants, enabling more effective meeting management.

[0850] The flow of the specific processing in Example 2 will be explained using Figure 17.

[0851] Step 1:

[0852] When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time. The input is the participants' speech, and the output is the audio data sent to the server. Specifically, the microphone captures the audio and transmits the data to the server over the network.

[0853] Step 2:

[0854] The server passes the received audio data to the speech recognition software. The speech recognition software converts the audio data into text information. The input is audio data, and the output is text information. As a data processing step, the audio signal is analyzed and the corresponding string is generated. Specifically, the speech recognition API analyzes the audio data and converts it into text format.

[0855] Step 3:

[0856] The server inputs text information into a generative AI model. The generative AI model analyzes the text information and generates meeting minutes. The input is text information, and the output is meeting minutes. As a data calculation, it extracts important points from the text information and organizes the flow of the discussion. Specifically, it utilizes natural language processing techniques to summarize the text data and create meeting minutes.

[0857] Step 4:

[0858] The server passes the audio data to the emotion recognition engine. The emotion recognition engine recognizes emotional information from the audio. The input is audio data, and the output is emotional information. As part of the data processing, the tone, volume, and speed of the voice are analyzed to determine the emotional state. Specifically, an emotion analysis tool analyzes the audio data and identifies emotional nuances.

[0859] Step 5:

[0860] The server provides emotional information to the generative AI model. The generative AI model modifies the meeting minutes based on the emotional information, highlighting important information and action items. The input is emotional information and the meeting minutes, and the output is the modified meeting minutes. As a data calculation, emotional information is used to add emotional nuances to the minutes and clarify action items. Specifically, the generative AI model analyzes the emotional information and adds information to the minutes such as "the speaker may be feeling dissatisfied."

[0861] Step 6:

[0862] The server provides the user with the finalized meeting transcript. The user can view the meeting transcript through their terminal and understand the content of the meeting. The input is the revised meeting transcript, and the output is the meeting transcript provided to the user. Specifically, the server sends the meeting transcript to the user's terminal, and the user views it.

[0863] (Application Example 2)

[0864] Next, we will describe application example 2 of form example 2. 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."

[0865] In a work environment, it is crucial to accurately record what workers say and create meeting minutes in real time. However, conventional systems have difficulty considering workers' emotions when creating meeting minutes or suggesting action items, limiting improvements in work efficiency. Furthermore, they are unable to properly recognize workers' emotions and follow up accordingly, thus necessitating improvements to the work environment.

[0866] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0867] In this invention, the server includes speech recognition means, generative AI means, and emotion recognition means. This enables the conversion of a worker's speech into text via speech recognition in the work environment, the generative AI to create meeting minutes in real time, the emotion recognition means to recognize the worker's emotions, and the suggestion of action items based on that information.

[0868] "Speech recognition means" refers to a technology that converts speech data into text data, making it possible to process spoken content in a digital format.

[0869] "Generative AI methods" refer to artificial intelligence technologies that use input text data to create meeting minutes, extract key points, and suggest action items.

[0870] "Emotion recognition means" refers to technology that analyzes the speaker's emotions from audio data and recognizes their emotional state.

[0871] "Meeting minutes" are documents that record the content of conversations during meetings or work sessions and compile them in a format that can be referenced later.

[0872] An "action item" is an item that indicates an issue identified in meeting minutes or the work environment, or the next action to be taken.

[0873] "Work environment" refers to the location and conditions in which a specific task is performed, such as a factory or office.

[0874] "Real-time" means that data processing and information provision occur immediately, resulting in a state where results are obtained without delay.

[0875] The system for carrying out this invention is centered around a server that includes speech recognition means, generative AI means, and emotion recognition means. The server receives the speech of workers as audio data through microphones installed in the work environment, such as a factory or office. The speech recognition means converts this audio data into text data.

[0876] The generative AI system creates meeting minutes in real time based on the converted text data, extracting key points and action items. Furthermore, the emotion recognition system analyzes the emotions of the participants from the audio data and provides this information to the generative AI system. This enables the generative AI system to create meeting minutes and suggest action items that take emotional information into account.

[0877] As a concrete example, during a morning meeting in a factory, when a worker expresses their opinion on a new work procedure, the server records their statement and, if the worker is feeling anxious, suggests an action item such as "Provide additional explanation to alleviate the worker's anxiety."

[0878] An example of a prompt message would be: "Analyze the worker's feelings based on this statement and reflect it in the meeting minutes. If the worker is feeling anxious, suggest a follow-up."

[0879] This system combines speech recognition using the Google Cloud Speech-to-Text API, generative AI using OpenAI's GPT model, and emotion recognition using Microsoft Azure's Emotion API. This enables more efficient communication in the work environment and allows for the creation of meeting minutes that take into account the emotions of the workers.

[0880] The flow of a specific process in Application Example 2 will be explained using Figure 18.

[0881] Step 1:

[0882] The server receives the worker's speech as audio data through microphones installed in the work environment. This audio data becomes the input. The server collects the audio data and prepares it for the next processing step.

[0883] Step 2:

[0884] The server uses speech recognition to convert the received audio data into text data. This process uses the Google Cloud Speech-to-Text API to analyze the audio data and output text data. The converted text data becomes the input for the next step.

[0885] Step 3:

[0886] The server uses generative AI methods to create meeting minutes in real time based on text data. This process utilizes OpenAI's GPT model to extract important points and action items from the text data. The generated meeting minutes are then output.

[0887] Step 4:

[0888] The server uses emotion recognition to analyze the worker's emotions from the audio data. This process uses the Microsoft Azure Emotion API to analyze the audio data and output emotion information. This emotion information then serves as input for the next step.

[0889] Step 5:

[0890] The server provides emotional information to a generative AI system, which then reflects this emotional information in the meeting minutes. This process adds emotion-based comments and action items to the minutes based on the emotional information. The final meeting minutes are then output.

[0891] Step 6:

[0892] Users review the final meeting minutes provided by the server and perform action items as needed. They then refer to the minutes to improve their work environment and follow up.

[0893] (Other examples)

[0894] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[0895] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0896] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0897] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[0898] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0899] [Fourth Embodiment]

[0900] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0901] As shown in Figure 7, the 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.

[0902] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0903] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0904] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0905] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0906] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0907] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0908] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0909] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0910] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0911] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0912] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0913] "Example of form 1"

[0914] In one embodiment of the present invention, a speech recognition module converts audio data input from a microphone installed in a conference room into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages. Next, a generative AI module receives the text data output from the speech recognition module as input and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' statements. "Embodiment Example 2"

[0915] As a concrete example of its use, when a meeting begins, microphones installed in the meeting room capture participants' speech, and a speech recognition module converts this into text data. A generative AI module analyzes this text data in real time and generates meeting minutes. When the meeting ends, the generative AI module creates a final version of the minutes, highlighting key points, summaries of discussions, and suggested action items. This makes it easier to organize information after the meeting and improves meeting productivity.

[0916] The following describes the processing flow for each example of the form.

[0917] "Example of form 1"

[0918] Step 1: Acquire audio data from the microphone installed in the conference room.

[0919] Step 2: The acquired audio data is input into the speech recognition module and converted into text data. This speech recognition module uses deep learning-based speech recognition technology and is capable of recognizing multiple languages.

[0920] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0921] Step 4: The generative AI module analyzes the text data and automatically generates meeting minutes. This generative AI module uses natural language processing technology and can not only create meeting minutes but also manage the progress of the meeting and summarize participants' remarks.

[0922] "Example of form 2"

[0923] Step 1: Once the meeting begins, microphones installed in the meeting room capture the participants' speech.

[0924] Step 2: Input the captured audio data into the speech recognition module and convert it into text data.

[0925] Step 3: Input the text data output from the speech recognition module into the generative AI module.

[0926] Step 4: The generative AI module analyzes the text data in real time and generates meeting minutes.

[0927] Step 5: Once the meeting concludes, the generative AI module creates a final version of the meeting minutes, highlighting key points, summaries of discussions, and suggested action items. This facilitates post-meeting information organization and improves meeting productivity.

[0928] (Example 1)

[0929] Next, we will describe Embodiment 1 of Example Form 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0930] There is a need to efficiently record meeting content and provide an environment where participants can focus on the discussion. However, traditional methods are time-consuming to create meeting minutes, and there is a risk of overlooking important points and action items. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder minute-taking.

[0931] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0932] In this invention, the server includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, and a generative information processing means for analyzing the text data and automatically generating meeting minutes. This makes it possible to quickly and accurately record the contents of a meeting, create meeting minutes in multiple languages ​​without overlooking important points or action items.

[0933] "Audio data" refers to information that represents the speech of meeting participants in a digital format.

[0934] An "input device" is a device used to acquire audio data, and includes microphones and the like.

[0935] "Speech recognition means" refers to a technology or device for converting speech data into text data.

[0936] "Text data" refers to character information converted by speech recognition technology.

[0937] "Generative information processing means" refers to a technology or device for analyzing text data and automatically generating meeting minutes.

[0938] An "output device" is a device used to display or print the generated meeting minutes.

[0939] Meeting minutes are documents that record the content of a meeting, including important points and action items.

[0940] "Multiple languages" refers to a group of languages ​​with different linguistic systems, and means that speech recognition means are capable of recognizing them.

[0941] A description of embodiments for carrying out this invention will be given.

[0942] System Overview

[0943] This system is designed to efficiently record meeting content and provide an environment where participants can concentrate on the discussion. The system includes an input device for acquiring audio data, a speech recognition means for converting the audio data into text data, a generative information processing means for analyzing the text data and automatically generating meeting minutes, and an output device for outputting the generated meeting minutes.

[0944] Hardware and software to be used

[0945] The microphone installed in the conference room will be used as the input device.

[0946] The speech recognition method will utilize deep learning-based speech recognition technology. Specifically, speech recognition software such as the Google Speech-to-Text API can be used.

[0947] The generative information processing system employs natural language processing technology. Specifically, generative AI models such as OpenAI's GPT-3 can be used.

[0948] The terminal's display or printer is used as the output device to display or print the generated meeting minutes.

[0949] Specific example

[0950] When a user says "Let's move on to the next agenda item" during a meeting, the microphone connected to the device picks up this audio.

[0951] The server uses speech recognition to convert the speech into text data that says, "Let's move on to the next topic."

[0952] The generative information processing system automatically generates meeting minutes, such as "The meeting has moved on to the next agenda item," based on this text data.

[0953] The server sends the generated meeting minutes to the terminal, and the user can view the minutes on the terminal.

[0954] Example of a prompt

[0955] "Please summarize the meeting content and create meeting minutes."

[0956] This system enables the rapid and accurate recording of meeting content, ensuring that important points and action items are not overlooked, and allows for the creation of meeting minutes in multiple languages.

[0957] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0958] Step 1:

[0959] When a user speaks in the conference room, a microphone connected to the terminal captures their voice. The input is the user's voice, and the output is digital audio data. The microphone converts the audio into an electrical signal, and the terminal prepares this to be sent to the server as digital data.

[0960] Step 2:

[0961] The server receives audio data transmitted from the terminal. The input is digital audio data, and the output is text data. The server activates speech recognition and converts the audio data into text data. In this process, speech recognition technology is used to analyze the waveform of the audio and generate the corresponding string of characters.

[0962] Step 3:

[0963] The server inputs text data obtained from the speech recognition means into the generative information processing means. The input is text data, and the output is the text of the meeting minutes. The generative information processing means analyzes the text data using natural language processing technology and automatically generates the meeting minutes. In this process, the key points of the text are extracted and the meeting minutes are constructed in a format that follows the progress of the meeting.

[0964] Step 4:

[0965] The server sends the generated meeting minutes to the terminal. The input is the text of the meeting minutes, and the output is the meeting minutes in a format viewable by the user. The terminal displays the received meeting minutes on its screen and shares them as needed, such as by printing or emailing. Users can review the meeting minutes on their terminal and reflect on the content of the meeting.

[0966] This series of processes makes it possible to record the contents of a meeting quickly and accurately, and to provide an environment where participants can concentrate on the discussion.

[0967] (Application Example 1)

[0968] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0969] Manual record-keeping by workers for work reporting and progress management within a factory is time-consuming and labor-intensive, and presents challenges in terms of accuracy and efficiency. Furthermore, real-time monitoring of work status is difficult, making it challenging for managers to respond quickly.

[0970] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0971] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, using that as input for the generative AI to create a work record, and automatically generating logs for work progress management and quality control. This enables the automation of work reporting and real-time monitoring of work status.

[0972] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or software for accurately acquiring a worker's speech as textual information.

[0973] "Generative AI methods" refer to artificial intelligence technologies that use natural language processing techniques to automatically generate work records and progress management information based on input text data.

[0974] "Work records" are data that document the activities and progress of workers within a factory, and are useful information for improving work efficiency and quality control.

[0975] "Progress management" is a management method used to understand the progress of work and to confirm whether the work is proceeding according to plan.

[0976] "Quality control" refers to management activities carried out to maintain a consistent quality of products and services, and is a process for ensuring the accuracy and efficiency of work.

[0977] A "log" is data that shows the history and records of work, and is information used to review and analyze the work content later.

[0978] The system for carrying out this invention includes speech recognition means, generative AI means, and automatic work record generation means. The server uses speech recognition technology such as the Google Speech-to-Text API as the speech recognition means to convert the worker's speech into text data in real time. The converted text data is analyzed using natural language processing technology such as OpenAI GPT-3 as the generative AI means to automatically generate work records and progress management information.

[0979] The terminal allows workers to submit voice reports via an application installed on devices such as smartphones and tablets. Users report their work details by voice through the terminal, and this voice data is sent to a server. The server converts the voice data into text and generates work records using generative AI. The generated work records are stored in a cloud-based database and can be accessed by administrators in real time.

[0980] For example, if a worker reports "Line 1 maintenance complete, no abnormalities," the server will generate a work log such as "October 10, 2023, 14:30 Line 1 maintenance complete, no abnormalities." An example of a prompt to input to the generation AI model in this case would be, "Convert the work report into a log in the following format: Date and time, Line number, Work details, Result."

[0981] This system enables the automation of work reporting and real-time monitoring of work status, thereby improving work efficiency within the factory.

[0982] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0983] Step 1:

[0984] The user uses a terminal to report their work details verbally. The terminal acquires the user's voice data through a microphone and sends that data to a server. The input is the user's voice data, and the output is the transmission of voice data to the server.

[0985] Step 2:

[0986] The server converts received audio data into text data using speech recognition. Specifically, it analyzes the audio using the Google Speech-to-Text API and generates the corresponding text. The input is audio data, and the output is text data.

[0987] Step 3:

[0988] The server analyzes text data using generative AI methods and automatically generates work records. Using natural language processing technologies such as OpenAI GPT-3, it creates records that include work progress and results based on the text data. The input is text data, and the output is a work record.

[0989] Step 4:

[0990] The server saves the generated work records to a database in the cloud. This allows administrators to monitor the work status in real time. The input is the work records, and the output is the saving of the records to the database.

[0991] Step 5:

[0992] Administrators access a database in the cloud to review work records. This allows for efficient progress and quality control of work. Input is a request to access the database, and output is a display of work records.

[0993] (Example 2)

[0994] Next, we will describe Example 2 of the morphological example. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0995] In meetings, accurately recording participants' remarks and efficiently organizing important information is crucial for improving meeting productivity. However, manual minute-taking is time-consuming and labor-intensive, and can lead to omissions and errors. Furthermore, in meetings where multiple languages ​​are used, language barriers can hinder the accurate transmission of information. To address these challenges, a system is needed that automatically transcribes audio data into text and extracts and organizes key information.

[0996] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0997] In this invention, the server includes means for acquiring audio data, means for converting the acquired audio data into text data, and means for analyzing the text data and generating meeting minutes. This makes it possible to record what is said during a meeting in real time and to efficiently organize important information.

[0998] "Means for acquiring audio data" refers to a device or method for capturing the speech of participants during a meeting in real time and saving it in digital format.

[0999] "Means for converting acquired audio data into text data" refers to an apparatus or method that performs the process of converting audio data into text information using speech recognition technology.

[1000] "Means for analyzing text data and generating meeting minutes" refers to a device or method that uses natural language processing technology to extract important information from text data, organize the content of a meeting, and output it as meeting minutes.

[1001] "Means for formatting generated meeting minutes into a final version and highlighting important information" refers to a device or method for organizing generated meeting minutes in an easy-to-read format and making particularly important points or action items stand out.

[1002] "Means for users to review and modify meeting minutes" refers to an interface or method that allows users to view the generated meeting minutes and modify or add to their content as needed.

[1003] This invention is a system that automatically records speeches in meetings and efficiently organizes important information. A specific embodiment of this system is described below.

[1004] The server uses audio input devices installed in the conference room to capture participants' speech in real time. Specifically, it uses multiple microphones as a typical audio input device to capture the audio signals as digital data.

[1005] Next, the server converts the acquired audio data into text data using speech recognition software. A general-purpose platform providing speech recognition technology can be used as this speech recognition software. This converts the audio data into text information.

[1006] Subsequently, the server analyzes the text data using a generative AI model and generates meeting minutes. This generative AI model can be a general-purpose generative AI platform utilizing natural language processing technology. The generated meeting minutes are formatted to highlight important points and action items.

[1007] Users can review the meeting minutes generated through their device and make corrections or additions as needed. This interface is designed to allow users to easily edit the meeting minutes.

[1008] For example, if the "new product launch plan" is discussed during a meeting, the server captures the statement, "The new product launch is scheduled for next month. We will decide on the marketing strategy in detail at the next meeting," and speech recognition software converts this to text. A generative AI model analyzes this text and generates meeting minutes like the following:

[1009] Key point: The new product is scheduled to be released next month.

[1010] Summary of discussion: Details regarding the marketing strategy will be decided at the next meeting.

[1011] Action item: Finalize the marketing strategy details at the next meeting.

[1012] An example of a prompt to input into a generative AI model might be, "Generate meeting minutes. Include key points, a summary of the discussion, and action items."

[1013] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1014] Step 1:

[1015] The server uses audio input devices installed in the conference room to capture participants' speech in real time. The input is the audio signal during the meeting, and the output is digital audio data. Specifically, it uses multiple microphones to capture the audio signal as digital data.

[1016] Step 2:

[1017] The server inputs the acquired audio data into speech recognition software and converts it into text data. The input is digital audio data, and the output is text data. Specifically, it performs a process of converting audio data into text information using speech recognition technology.

[1018] Step 3:

[1019] The server inputs text data into a generative AI model and generates meeting minutes. The input is text data, and the output is the generated meeting minutes. Specifically, it uses natural language processing techniques to analyze the text data, extract important information, and create the meeting minutes.

[1020] Step 4:

[1021] The server formats the generated meeting minutes as the final version and highlights important information. The input is the generated meeting minutes, and the output is the formatted, final version of the meeting minutes. Specifically, it organizes the content of the meeting minutes for easier reading and highlights particularly important points and action items.

[1022] Step 5:

[1023] The user reviews the meeting minutes generated through the terminal and makes corrections or additions as needed. The input is the formatted final version of the meeting minutes, and the output is the meeting minutes modified by the user. Specifically, the user views the meeting minutes and edits the content through the interface.

[1024] (Application Example 2)

[1025] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1026] In factory and other work environments, there is a need to efficiently record worker instructions and reports and to grasp the progress of work in real time. However, traditional methods require manual recording and management, which leads to decreased work efficiency. Furthermore, it is difficult to immediately grasp important points and progress of work, increasing the burden on managers.

[1027] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1028] In this invention, the server includes a speech recognition means, a generative AI means, and means for converting the worker's speech into text using speech recognition, and for the generative AI to create a work record using that as input, highlighting the progress of the work and important work items. This enables real-time work recording and progress management at the work site.

[1029] "Speech recognition means" refers to a technology that converts speech into text data, and is a device or software for recording a worker's speech in digital format.

[1030] "Generative AI methods" are artificial intelligence technologies that analyze input text data and generate information tailored to specific purposes, such as systems for automatically creating work records and progress highlights.

[1031] A "work record" is a text-based record of instructions and reports from workers at a work site, and serves as data for understanding the progress of work and important work items.

[1032] "Progress status" refers to information indicating how far a task has progressed, and is an essential indicator for efficient task management and planning.

[1033] "Important work items" refer to points in a task that require particular attention, or factors that significantly influence the success or failure of the task, and are information that managers should prioritize understanding.

[1034] The system for carrying out this invention is for recording workers' speech in real time and managing the progress of work in a workplace such as a factory. The server includes speech recognition means, generative AI means, and means for creating work records.

[1035] The speech recognition system captures the worker's speech and converts the audio data into text data. Specifically, speech recognition software such as the Google Speech-to-Text API can be used. The converted text data is sent to the server.

[1036] The generative AI system analyzes received text data and generates work records. Using generative AI models such as OpenAI GPT-3, it is possible to highlight work progress and important work items. The generated work records are displayed in real time on the administrator's terminal.

[1037] For example, if a worker says, "I will proceed to the next step," the speech recognition system converts this into text, and the generative AI system records "Step in progress: Proceeding to the next step" in the work log. This information is displayed on the terminal so that the administrator can check it immediately.

[1038] Examples of prompt messages include the following:

[1039] "Voice input: 'Proceed to the next step.'"

[1040] "Prompt: 'Update the work log and record that you are proceeding to the next step.'"

[1041] In this way, real-time work recording and progress management are achieved at the work site.

[1042] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1043] Step 1:

[1044] The user speaks at the work site. The voice input device captures the user's speech and sends it to the server as audio data. The input is the user's voice, and the output is audio data.

[1045] Step 2:

[1046] The server converts received audio data into text data using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to analyze the audio data and generate the corresponding text. The input is audio data, and the output is text data.

[1047] Step 3:

[1048] The server analyzes text data using generative AI methods and generates work records. Using generative AI models such as OpenAI GPT-3, it extracts work progress and important work items from the text data and creates a record. The input is text data, and the output is a work record.

[1049] Step 4:

[1050] The server sends the generated work log to the administrator's terminal. The administrator's terminal displays the received work log in real time, allowing them to check the progress of the work. The input is the work log, and the output is the display on the administrator's terminal.

[1051] Step 5:

[1052] The administrator monitors the progress of work based on the work records displayed on the terminal and issues instructions as needed. This enables efficient work management. The input is the work records displayed on the terminal, and the output is the administrator's judgment and instructions.

[1053] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1054] "Example of form 1"

[1055] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[1056] "Example of form 2"

[1057] One embodiment of the present invention provides a system that includes a speech recognition means, a generative AI means, and an emotion engine that recognizes the user's emotions. This system converts the speech of meeting participants into text using speech recognition, and the generative AI uses this as input to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items. Furthermore, the emotion engine recognizes emotions from the user's voice and provides this emotion information to the generative AI. Based on the provided emotion information, the generative AI creates meeting minutes and summarizes discussions, and suggests action items based on the emotion information. As a specific example, when a participant speaks during a meeting, the emotion engine recognizes emotions not only from the content of the statement but also from the tone, volume, and speed of the voice during the statement. For example, if the system senses that the speaker is feeling anger or dissatisfaction, it provides this information to the generative AI. Based on this emotion information, the generative AI adds information to the meeting minutes stating that "the speaker may be feeling dissatisfied." It also suggests "follow-up on the speaker's dissatisfaction" as an action item based on this emotion information. This means that meeting minutes will not only be a record of what was said, but will also reflect the emotional state of the participants, enabling more effective meeting management.

[1058] The following describes the processing flow for each example of the form.

[1059] "Example of form 1"

[1060] Step 1: Meeting participants speak. These speeches are entered into the system as audio data.

[1061] Step 2: The speech recognition system converts the spoken audio data into text data.

[1062] Step 3: The emotion engine recognizes emotions from the audio data of the speech. This emotion is input into the system as emotion information.

[1063] Step 4: The generative AI receives text data and emotional information as input.

[1064] Step 5: The generative AI creates meeting minutes from the text data, highlighting key points and summaries of the discussion.

[1065] Step 6: The generative AI suggests action items based on emotional information.

[1066] (Example 1)

[1067] Next, we will describe Embodiment 1 of Example Form 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1068] In meetings, it is important to accurately record what participants say and create meeting minutes. However, traditional methods only record the content of what was said, making it difficult to create minutes that reflect the emotions of the participants or the progress of the meeting. Furthermore, it was difficult to appropriately grasp the emotional reactions that arose during the meeting and propose actions based on them.

[1069] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1070] In this invention, the server includes means for acquiring audio data, speech recognition means for converting the audio data into text data, generative information processing means for receiving the text data as input and automatically generating meeting minutes using natural language processing technology, and emotion recognition means for recognizing emotions from the audio data and providing that information to the generative information processing means. This makes it possible to create meeting minutes that reflect not only the content of what the participants said, but also their emotions and the progress of the meeting, thereby enabling the effective management of meetings.

[1071] "Audio data" refers to sound information acquired to record what participants say in a meeting.

[1072] "Speech recognition means" refers to technology that analyzes speech data and converts it into corresponding text data.

[1073] A "generative information processing system" is a system that receives text data as input and has the function of automatically generating meeting minutes using natural language processing technology.

[1074] "Emotion recognition means" refers to a technology that analyzes participants' emotions from audio data and provides that information to generative information processing means.

[1075] Meeting minutes are documents that record what was said, the progress of the meeting, and the feelings of the participants.

[1076] "Action items" refer to specific actions or follow-up items proposed during a meeting.

[1077] This invention is a system that efficiently records the content of discussions in meetings and automatically generates meeting minutes that reflect the emotions of the participants. The system acquires audio data, converts it into text data using speech recognition technology, and then creates meeting minutes using generative information processing technology. In addition, it analyzes the emotions of the participants using emotion recognition technology and reflects that information in the meeting minutes.

[1078] The server captures user speech as audio data through microphones installed in the conference room. This audio data is converted into text data using speech recognition technologies such as Google Speech-to-Text API or IBM Watson Speech to Text. The speech recognition technology is capable of recognizing multiple languages.

[1079] Next, the server uses natural language processing technologies such as OpenAI's GPT-3 and Google's BERT as generative information processing tools. This allows for the automatic generation of meeting minutes based on text data, as well as management of meeting progress and summarization of participants' statements.

[1080] Furthermore, the server uses emotion recognition means to analyze the participants' emotions from the audio data. The emotion engine analyzes the tone, volume, and speed of the voice and provides the emotion information to the generative information processing means. Based on this emotion information, the generative information processing means adds information about emotions to the meeting minutes and proposes action items based on the emotion information.

[1081] For example, if a user says, "I am very dissatisfied with the slow progress of this project," the emotion engine recognizes anger from the tone of the statement. The generative information processing system records in the meeting minutes that "Participants are dissatisfied with the project's progress" and proposes "Schedule a follow-up meeting regarding project progress" as an action item.

[1082] An example of a prompt message is: "Convert meeting comments to text, analyze sentiment, and create meeting minutes. Also, suggest action items based on the speaker's sentiment."

[1083] The flow of the specific processing in Example 1 will be explained using Figure 15.

[1084] Step 1:

[1085] When a user speaks in a conference room, their voice is captured by a microphone installed in the room. The server receives the audio data transmitted from the microphone. The input is the user's voice, and the output is the audio data sent to the server. This audio data is used in subsequent processing steps.

[1086] Step 2:

[1087] The server passes the received audio data to the speech recognition system. The speech recognition system uses technologies such as the Google Speech-to-Text API or IBM Watson Speech to Text to convert the audio data into text data. The input is audio data, and the output is text data. This conversion allows the audio information to be treated as text information.

[1088] Step 3:

[1089] The server inputs text data output from the speech recognition system into a generative information processing system. The generative information processing system analyzes the text data using natural language processing technologies such as OpenAI's GPT-3 and Google's BERT, and automatically generates meeting minutes. The input is text data, and the output is meeting minutes. This process is used for managing the progress of meetings and summarizing participants' statements.

[1090] Step 4:

[1091] The server passes audio data to an emotion recognition system. The emotion recognition system analyzes the tone, volume, and speed of the voice to recognize the user's emotions. The input is audio data, and the output is emotion information. This information is used to understand the emotional state of the participants.

[1092] Step 5:

[1093] The server passes the emotion information provided by the emotion recognition means to the generative information processing means. The generative information processing means adds information about emotions to the meeting minutes based on the emotion information and proposes action items based on the emotion information. The input is emotion information, and the output is meeting minutes and action items that reflect the emotions. This process makes the meeting minutes more detailed and useful.

[1094] (Application Example 1)

[1095] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1096] Efficiently recording worker instructions and reports and managing the progress of work is crucial in the workplace. However, conventional methods make it difficult to respond appropriately while considering workers' emotions and stress levels, resulting in insufficient improvements in work efficiency and safety.

[1097] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1098] In this invention, the server includes speech recognition means, generative information processing means, and emotion recognition means. This makes it possible to transcribe the worker's speech into text in real time, automatically generate work records, and recognize the worker's emotions to suggest appropriate actions.

[1099] "Speech recognition means" refers to a technology that converts speech data into text data, and is a device or system that converts a worker's speech into text information in real time.

[1100] "Generative information processing means" refers to technology that processes information based on text data and automatically generates work records and action items.

[1101] "Emotion recognition means" refers to a technology that analyzes the emotions of a worker from voice data and recognizes their emotional state.

[1102] A "work record" is a document that records the progress and important points of a task, generated based on the worker's spoken content.

[1103] "Action items" are specific actions or countermeasures proposed based on the worker's emotional state and the nature of their work.

[1104] The system for carrying out this invention includes speech recognition means, generative information processing means, and emotion recognition means. The server uses the Google Cloud Speech-to-Text API as the speech recognition means to convert the worker's voice into text data in real time. The generative information processing means uses an OpenAI generative AI model to automatically generate work records based on the text data and propose action items. The emotion recognition means uses IBM Watson Tone Analyzer to analyze the worker's emotions from the voice data and provides that information to the generative information processing means.

[1105] The system uses smartphones or tablets as terminals, connecting a microphone to collect voice data. Users input instructions and reports at the work site using voice, and the content is automatically converted to text and saved as a work record.

[1106] For example, if a worker says, "This task is difficult," the emotion recognition system recognizes stress, and the generative information processing system generates the action item, "The worker may be experiencing stress. We suggest a break."

[1107] An example of a prompt message is: "Transcribe the worker's statements into text, analyze their emotions, and suggest necessary actions."

[1108] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[1109] Step 1:

[1110] The terminal collects the worker's voice through a microphone. The input is the worker's voice data, and the output is that voice data. The terminal sends this voice data to the server.

[1111] Step 2:

[1112] The server uses the Google Cloud Speech-to-Text API as a speech recognition tool to convert received audio data into text data. The input is audio data, and the output is text data. This conversion allows the operator's speech to be obtained as text information.

[1113] Step 3:

[1114] The server uses IBM Watson Tone Analyzer as an emotion recognition tool to analyze the worker's emotions from text data. The input is text data, and the output is emotion information. The server provides this emotion information to a generative information processing system.

[1115] Step 4:

[1116] The server uses OpenAI's generative AI model as a generative information processing tool to automatically generate work records based on text data and sentiment information. The input is text data and sentiment information, and the output is the work record. The server saves this work record and suggests action items as needed.

[1117] Step 5:

[1118] The user reviews the generated work records and action items on their terminal. The input consists of work records and action items, while the output is the information the user reviews. Based on this, the user can consider the progress of the work and potential countermeasures.

[1119] (Example 2)

[1120] Next, we will describe Example 2 of the morphological example. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1121] In meetings, it is difficult to accurately record what participants say and to create minutes that reflect their emotional state. Furthermore, while it is necessary to quickly grasp important points and action items after a meeting, traditional methods are time-consuming and labor-intensive.

[1122] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1123] In this invention, the server includes means for acquiring audio data, means for converting audio data into text information, and means for recognizing emotional information. This makes it possible to accurately record the content of meetings and create meeting minutes that reflect the emotional state of the participants.

[1124] "Means for acquiring audio data" refers to a device or method for collecting the speech of meeting participants in real time.

[1125] "Means for converting audio data into text information" refers to a technology or device for analyzing acquired audio data and converting it into a corresponding string of characters.

[1126] "Means for analyzing textual information to generate meeting records" refers to a technology or device that uses converted textual information to organize the content of a meeting and record important points and the flow of discussion.

[1127] "Means for recognizing emotional information" refers to technologies or devices for analyzing and identifying emotional states from participants' voices.

[1128] "Methods for revising meeting records based on emotional information to emphasize important information and action items" refers to technologies or devices that utilize recognized emotional information to add emotional nuances to meeting records and clarify important information and next steps to take.

[1129] This invention is a system that accurately records the content of speeches and the emotional state of participants during meetings, and generates effective meeting minutes. When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time.

[1130] When the server receives audio data, it uses speech recognition software to convert the audio data into text. Specifically, a common speech recognition API can be used for speech recognition. This conversion records the speeches during the meeting in text format.

[1131] Next, the server uses a generative AI model to analyze the converted text information and generate a meeting record. This generative AI model can utilize natural language processing technology. The generative AI model extracts important points from the text information and organizes the flow of the discussion.

[1132] Furthermore, the server uses an emotion recognition engine to recognize emotional information from the audio data. The emotion recognition engine analyzes the tone, volume, and speed of the speech to determine the emotional state of the participants. This allows the server to grasp the emotional nuances contained in what is said during the meeting.

[1133] The server modifies the generated meeting minutes based on recognized sentiment information, highlighting important information and action items. Specifically, it uses sentiment information to add information to the minutes such as "the speaker may be feeling dissatisfied." It can also suggest action items based on sentiment information, such as "follow up on the speaker's dissatisfaction."

[1134] An example of a prompt might be, "Please create meeting minutes. Based on the following text data and sentiment information, highlight key points, summaries of discussions, and action items." This system ensures that meeting minutes are not merely a record of what was said, but also reflect the emotional state of the participants, enabling more effective meeting management.

[1135] The flow of the specific processing in Example 2 will be explained using Figure 17.

[1136] Step 1:

[1137] When a meeting begins, the user acquires participants' speech through an audio input device installed in the meeting room. The terminal transmits this audio data to the server in real time. The input is the participants' speech, and the output is the audio data sent to the server. Specifically, the microphone captures the audio and transmits the data to the server over the network.

[1138] Step 2:

[1139] The server passes the received audio data to the speech recognition software. The speech recognition software converts the audio data into text information. The input is audio data, and the output is text information. As a data processing step, the audio signal is analyzed and the corresponding string is generated. Specifically, the speech recognition API analyzes the audio data and converts it into text format.

[1140] Step 3:

[1141] The server inputs text information into a generative AI model. The generative AI model analyzes the text information and generates meeting minutes. The input is text information, and the output is meeting minutes. As a data calculation, it extracts important points from the text information and organizes the flow of the discussion. Specifically, it utilizes natural language processing techniques to summarize the text data and create meeting minutes.

[1142] Step 4:

[1143] The server passes the audio data to the emotion recognition engine. The emotion recognition engine recognizes emotional information from the audio. The input is audio data, and the output is emotional information. As part of the data processing, the tone, volume, and speed of the voice are analyzed to determine the emotional state. Specifically, an emotion analysis tool analyzes the audio data and identifies emotional nuances.

[1144] Step 5:

[1145] The server provides emotional information to the generative AI model. The generative AI model modifies the meeting minutes based on the emotional information, highlighting important information and action items. The input is emotional information and the meeting minutes, and the output is the modified meeting minutes. As a data calculation, emotional information is used to add emotional nuances to the minutes and clarify action items. Specifically, the generative AI model analyzes the emotional information and adds information to the minutes such as "the speaker may be feeling dissatisfied."

[1146] Step 6:

[1147] The server provides the user with the finalized meeting transcript. The user can view the meeting transcript through their terminal and understand the content of the meeting. The input is the revised meeting transcript, and the output is the meeting transcript provided to the user. Specifically, the server sends the meeting transcript to the user's terminal, and the user views it.

[1148] (Application Example 2)

[1149] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1150] In a work environment, it is crucial to accurately record what workers say and create meeting minutes in real time. However, conventional systems have difficulty considering workers' emotions when creating meeting minutes or suggesting action items, limiting improvements in work efficiency. Furthermore, they are unable to properly recognize workers' emotions and follow up accordingly, thus necessitating improvements to the work environment.

[1151] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1152] In this invention, the server includes speech recognition means, generative AI means, and emotion recognition means. This enables the conversion of a worker's speech into text via speech recognition in the work environment, the generative AI to create meeting minutes in real time, the emotion recognition means to recognize the worker's emotions, and the suggestion of action items based on that information.

[1153] "Speech recognition means" refers to a technology that converts speech data into text data, making it possible to process spoken content in a digital format.

[1154] "Generative AI methods" refer to artificial intelligence technologies that use input text data to create meeting minutes, extract key points, and suggest action items.

[1155] "Emotion recognition means" refers to technology that analyzes the speaker's emotions from audio data and recognizes their emotional state.

[1156] "Meeting minutes" are documents that record the content of conversations during meetings or work sessions and compile them in a format that can be referenced later.

[1157] An "action item" is an item that indicates an issue identified in meeting minutes or the work environment, or the next action to be taken.

[1158] "Work environment" refers to the location and conditions in which a specific task is performed, such as a factory or office.

[1159] "Real-time" means that data processing and information provision occur immediately, resulting in a state where results are obtained without delay.

[1160] The system for carrying out this invention is centered around a server that includes speech recognition means, generative AI means, and emotion recognition means. The server receives the speech of workers as audio data through microphones installed in the work environment, such as a factory or office. The speech recognition means converts this audio data into text data.

[1161] The generative AI system creates meeting minutes in real time based on the converted text data, extracting key points and action items. Furthermore, the emotion recognition system analyzes the emotions of the participants from the audio data and provides this information to the generative AI system. This enables the generative AI system to create meeting minutes and suggest action items that take emotional information into account.

[1162] As a concrete example, during a morning meeting in a factory, when a worker expresses their opinion on a new work procedure, the server records their statement and, if the worker is feeling anxious, suggests an action item such as "Provide additional explanation to alleviate the worker's anxiety."

[1163] An example of a prompt message would be: "Analyze the worker's feelings based on this statement and reflect it in the meeting minutes. If the worker is feeling anxious, suggest a follow-up."

[1164] This system combines speech recognition using the Google Cloud Speech-to-Text API, generative AI using OpenAI's GPT model, and emotion recognition using Microsoft Azure's Emotion API. This enables more efficient communication in the work environment and allows for the creation of meeting minutes that take into account the emotions of the workers.

[1165] The flow of a specific process in Application Example 2 will be explained using Figure 18.

[1166] Step 1:

[1167] The server receives the worker's speech as audio data through microphones installed in the work environment. This audio data becomes the input. The server collects the audio data and prepares it for the next processing step.

[1168] Step 2:

[1169] The server uses speech recognition to convert the received audio data into text data. This process uses the Google Cloud Speech-to-Text API to analyze the audio data and output text data. The converted text data becomes the input for the next step.

[1170] Step 3:

[1171] The server uses generative AI methods to create meeting minutes in real time based on text data. This process utilizes OpenAI's GPT model to extract important points and action items from the text data. The generated meeting minutes are then output.

[1172] Step 4:

[1173] The server uses emotion recognition to analyze the worker's emotions from the audio data. This process uses the Microsoft Azure Emotion API to analyze the audio data and output emotion information. This emotion information then serves as input for the next step.

[1174] Step 5:

[1175] The server provides emotional information to a generative AI system, which then reflects this emotional information in the meeting minutes. This process adds emotion-based comments and action items to the minutes based on the emotional information. The final meeting minutes are then output.

[1176] Step 6:

[1177] Users review the final meeting minutes provided by the server and perform action items as needed. They then refer to the minutes to improve their work environment and follow up.

[1178] (Other examples)

[1179] Since this is the same as the specific processing described in the other embodiments of the first embodiment above, the explanation will be omitted.

[1180] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1181] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1182] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[1183] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1184] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1185] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1186] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1187] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1188] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1189] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1190] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1191] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[1192] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1193] 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.

[1194] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1195] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1196] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1197] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1198] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1199] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1200] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1201] The following is further disclosed regarding the embodiments described above.

[1202] (Claim 1)

[1203] A system including a speech recognition method, a generative AI method, and a method for converting the speech of meeting participants into text using speech recognition, and for the generative AI to create meeting minutes using that as input, highlighting key points and summaries of discussions, and suggesting action items.

[1204] (Claim 2)

[1205] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1206] (Claim 3)

[1207] The system according to claim 1, wherein the generation AI means also performs tasks other than creating meeting minutes, such as managing the progress of the meeting and summarizing the statements of the participants.

[1208] (Claim 4)

[1209] The system according to claim 1, further comprising, along with the speech recognition means and generative AI means, an emotion engine that recognizes the user's emotions.

[1210] (Claim 5)

[1211] The system according to claim 4, wherein the emotion engine recognizes emotions from the user's voice and provides the emotion information to the generation AI means.

[1212] (Claim 6)

[1213] The system according to claim 5, wherein the generation AI means creates meeting minutes and summarizes discussions based on the provided emotional information, and proposes action items based on the emotional information.

[1214] "Example 1"

[1215] (Claim 1)

[1216] An input device for acquiring audio data,

[1217] A speech recognition means for converting audio data into text data,

[1218] A generative information processing system that analyzes text data and automatically generates meeting minutes,

[1219] An output device that outputs the generated meeting minutes,

[1220] A system that includes this.

[1221] (Claim 2)

[1222] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1223] (Claim 3)

[1224] The system according to claim 1, wherein the generation information processing means performs tasks other than creating meeting minutes, such as managing the progress of the meeting and summarizing the statements of the participants.

[1225] "Application Example 1"

[1226] (Claim 1)

[1227] Voice recognition means and

[1228] Generative AI methods,

[1229] A method for automatically generating logs for work progress management and quality control by converting the worker's speech into text using speech recognition, and using that as input for a generative AI to create work records, and

[1230] A system that includes this.

[1231] (Claim 2)

[1232] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1233] (Claim 3)

[1234] The system according to claim 1, wherein the generation AI means performs tasks other than creating work records, such as managing the progress of work and summarizing the worker's statements.

[1235] Example 2

[1236] (Claim 1)

[1237] Means for acquiring audio data,

[1238] A means of converting acquired audio data into text data,

[1239] A means of analyzing text data and generating meeting minutes,

[1240] The generated meeting minutes are formatted as the final version, and a means of highlighting important information is provided.

[1241] A system that includes means for users to review and correct meeting minutes.

[1242] (Claim 2)

[1243] The system according to claim 1, wherein the means for acquiring the voice data is capable of recognizing multiple languages.

[1244] (Claim 3)

[1245] The system according to claim 1, wherein the means for analyzing the text data and generating meeting minutes also manages the progress of the meeting and summarizes the statements of the participants.

[1246] "Application Example 2"

[1247] (Claim 1)

[1248] Voice recognition means and

[1249] Generative AI methods,

[1250] A system that converts the worker's speech into text using speech recognition, uses that as input for a generative AI to create a work record, and highlights the progress of the work and important work items.

[1251] A system that includes this.

[1252] (Claim 2)

[1253] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1254] (Claim 3)

[1255] The system according to claim 1, wherein the generation AI means performs tasks other than creating work records, such as managing the progress of work and summarizing the worker's statements.

[1256] "Example 1 of combining an emotion engine"

[1257] (Claim 1)

[1258] Means for acquiring audio data,

[1259] A speech recognition means for converting audio data into text data,

[1260] A generative information processing system that receives text data as input and automatically generates meeting minutes using natural language processing technology,

[1261] An emotion recognition means that recognizes emotions from audio data and provides that information to a generative information processing means,

[1262] A generative information processing means adds emotional information to meeting minutes based on emotional information and proposes action items based on emotional information,

[1263] A system that includes this.

[1264] (Claim 2)

[1265] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1266] (Claim 3)

[1267] The system according to claim 1, wherein the generation information processing means performs tasks other than creating meeting minutes, such as managing the progress of the meeting and summarizing the statements of the participants.

[1268] "Application example 1 of combining emotional engines"

[1269] (Claim 1)

[1270] Voice recognition means and

[1271] Generative information processing means,

[1272] A method that converts the worker's speech into text using speech recognition, uses that as input for generative information processing to create a work record, highlighting important points and summaries of the work, and suggesting action items.

[1273] Means of recognizing emotions,

[1274] A means of recognizing the emotions of workers and proposing action items based on that information,

[1275] A system that includes this.

[1276] (Claim 2)

[1277] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1278] (Claim 3)

[1279] The system according to claim 1, wherein the generation information processing means performs tasks other than creating work records, such as managing the progress of work and summarizing the statements of workers.

[1280] "Example 2 of combining an emotion engine"

[1281] (Claim 1)

[1282] Means for acquiring audio data,

[1283] A means of converting audio data into text information,

[1284] A means of analyzing textual information to generate meeting records,

[1285] Means of recognizing emotional information,

[1286] A means of revising meeting records based on emotional information and emphasizing important information and action items,

[1287] A system that includes this.

[1288] (Claim 2)

[1289] The system according to claim 1, wherein the means for converting the audio data into text information is capable of recognizing multiple languages.

[1290] (Claim 3)

[1291] The system according to claim 1, wherein the means for analyzing the aforementioned textual information to generate meeting records also manages the progress of the meeting and summarizes the statements of the participants.

[1292] "Application example 2 when combining with an emotional engine"

[1293] (Claim 1)

[1294] Voice recognition means and

[1295] Generative AI methods,

[1296] A method that converts the speech of meeting participants into text using speech recognition, and then uses that as input for a generative AI to create meeting minutes, highlighting key points and summaries of discussions, and suggesting action items.

[1297] Means of recognizing emotions,

[1298] A system that includes a mechanism for converting workers' speech into text using speech recognition in the work environment, a generative AI for creating meeting minutes in real time, and an emotion recognition mechanism for recognizing workers' emotions and suggesting action items based on that information.

[1299] (Claim 2)

[1300] The system according to claim 1, wherein the speech recognition means is capable of recognizing multiple languages.

[1301] (Claim 3)

[1302] The system according to claim 1, wherein the generation AI means performs tasks other than creating meeting minutes, such as managing the progress of the work environment and summarizing the statements of workers. [Explanation of Symbols]

[1303] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

[Claim 1] Equipped with a processor, The aforementioned processor, To capture the speech of meeting participants, microphones are used to collect audio data. The acquired audio data is converted into text data using acoustic and language models to support multiple languages. Audio features are extracted from the aforementioned audio data, and the emotions of the meeting participants are recognized using a machine learning model. The converted text data is analyzed to identify the meeting agenda, and prompts are generated to instruct the generating AI model to automatically generate meeting minutes that include the identified agenda and also include key points, summaries of discussions, and suggested action items. Based on the recognized emotion, adjust the prompt to instruct that information regarding the emotion be added to the minutes and that suggestions for action items based on the emotion be included. The adjusted prompts are input to the generative AI model, which then generates the minutes, including the key points and summary of the discussion, and suggestions for action items based on the sentiment, with added information about the sentiment. system.

Images