system

Abstract

Provide a system. A means for acquiring voice data in real time; means for converting the voice data into text data in real time; means for summarizing the text data; means for visualizing said summary; means for accepting user input to modify the summarized or visualized information; and means for updating the summarized or visualized information based on the user input.

Description

[Technical Field]

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

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

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

[0004] In today's business environment, the frequency of remote meetings is increasing, creating a growing need for efficient recording of meeting content in real time. However, traditional methods of recording meetings make it difficult to summarize meeting content and accurately grasp key points. This can lead to issues such as meeting participants overlooking important agenda items and requiring time and effort to review the content after the meeting. Furthermore, there is no system that can quickly respond to requests for revisions during the meeting. To solve these issues, a new method is needed to summarize meeting content in real time and visualize key points. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means: a system including means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data using natural language processing technology, means for visualizing the summary, means for accepting user input for modifying the summary or visualized information as voice input, and means for converting the voice input into text data and updating the summary or visualized information based on the text data. This system efficiently summarizes and visualizes meeting content in real time, allowing participants to immediately grasp important points. Furthermore, by allowing users to voice-input correction requests during the meeting, the summary content and visualized information are quickly updated, allowing the meeting to proceed smoothly.

[0006] "Audio data" means data for recording or transmitting audio in digital form.

[0007] "Text data" refers to data for recording or transmitting character information in digital form.

[0008] A "summary" is information that shortens the original content and extracts only the important points.

[0009] "Visualization" is the process of displaying information in graphical or textual form to make it easier to understand.

[0010] "User input" refers to operations or data input performed by a user to give instructions to the system.

[0011] "Natural language processing technology" is a technology that enables computers to understand and process human language.

[0012] "Real-time" refers to processing and display occurring immediately without delay.

[0013] "Voice input" is a method by which a user inputs data or instructions into a system using voice.

[0014] "Update" is the process of replacing existing data or information with new data or information. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0020] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[0037] The server has a means for acquiring voice data in real time. Specifically, it receives voice data captured during a remote conference and processes the data. The server converts the voice data into text data in real time using voice recognition technology.

[0038] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[0039] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[0040] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[0041] Furthermore, if the user wishes to modify the summary or visualized information during the process, the device is also provided with a means for accepting user input. This user input can be performed as voice input. When the user issues a command for modification by voice, the device transmits the voice to the server. The server converts the voice input into text data and updates the summary or visualized information based on that information.

[0042] Specific examples

[0043] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[0044] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0045] The audio data of the conference is sent to the server and converted into text data in real time. This text data is summarized as follows:

[0046] The new product will be released next month.

[0047] Enhanced marketing campaigns.

[0048] Focus on online advertising.

[0049] The budget has been approved.

[0050] This summary is visually formatted as a bulleted list and displayed on the terminal, as shown below:

[0051] The new product will be released next month.

[0052] Enhanced marketing campaigns.

[0053] · Focus on online advertising.

[0054] · Budget approved.

[0055] If the user responds to this summary by voice, saying "Please include a breakdown of the marketing budget," the server converts the voice into text and updates the summary.

[0056] The system provides a concise and accurate real-time summary and visualization of meeting content, allowing participants to immediately grasp key points. It also updates the summary based on user instructions, ensuring smooth meeting progress.

[0057] Thus, the present invention provides an embodiment for supporting efficient information sharing and decision-making in remote meetings.

[0058] The processing flow will be explained below.

[0059] Specific flow of program processing

[0060] Server-side processing

[0061] Step 1:

[0062] Setting authentication information

[0063] The server loads the credentials for the Google® Cloud Speech-to-Text service, which allows the server to securely access the API.

[0064] Step 2:

[0065] Real-time speech recognition settings

[0066] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[0067] Step 3:

[0068] Creating an Audio Stream

[0069] The server sends a request to the Google Cloud Speech-to-Text service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them to text in real time.

[0070] Step 4:

[0071] Text summary

[0072] The server uses natural language processing techniques to summarize the recognized speech text, extracting key points from long text data and creating a concise summary.

[0073] Step 5:

[0074] Summary visualization

[0075] The server formats the summarized text into a visually easy-to-understand format, specifically by turning the text into bullet points and formatting it in a way that makes it easier to read.

[0076] Terminal side processing

[0077] Step 1:

[0078] Audio Stream Input

[0079] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[0080] Step 2:

[0081] Viewing summaries and visualizations

[0082] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[0083] User processing

[0084] Step 1:

[0085] Check and correct statements

[0086] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[0087] Step 2:

[0088] Entering correction instructions

[0089] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[0090] Step 3:

[0091] Processing and updating correction instructions

[0092] The server converts the voice input into text and uses that information to update summaries and visualizations.

[0093] Specific processing examples

[0094] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[0095] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0096] The server receives the audio data and converts it to text in real time:

[0097] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0098] The server summarizes this text:

[0099] The new product will be released next month.

[0100] Enhanced marketing campaigns.

[0101] Focus on online advertising.

[0102] The budget has been approved.

[0103] The device receives this summary and displays it to the user:

[0104] The new product will be released next month.

[0105] Enhanced marketing campaigns.

[0106] · Focus on online advertising.

[0107] · Budget approved.

[0108] User says "Please include a breakdown of your marketing budget":

[0109] "Please include a breakdown of your marketing budget."

[0110] The server receives this voice prompt and updates the summary:

[0111] The new product will be released next month.

[0112] Marketing campaign enhancements and budget breakdown.

[0113] Focus on online advertising.

[0114] The budget has been approved.

[0115] In this way, a system is realized in which the server, terminals, and users work together to efficiently summarize and visualize the contents of remote meetings and update the information as needed.

[0116] Example 1

[0117] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0118] Conventional remote conferencing systems make it difficult for participants to grasp the content of the meeting in real time. Furthermore, it takes a great deal of time and effort for participants to review important points after the meeting. Furthermore, it is not easy to revise summaries and visualized information in a timely manner during the meeting. To solve these problems, a system is needed that can summarize and visualize the content of the meeting in real time and quickly update it based on user instructions.

[0119] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0120] In this invention, the server includes means for acquiring voice information in real time, means for converting the voice information into document information in real time, means for summarizing the document information, means for representing the document information in graphics, means for accepting user input, and means for updating the summary or the graphic representation based on the user input. By converting voice information into document information in real time, generating a summary using natural language processing technology, and visualizing the summary, the server enables users to quickly grasp the content of the meeting and further enables information to be updated based on user instructions.

[0121] "Audio information" refers to data recorded in real time of participants' speech during a remote conference.

[0122] "Real-time" means that data is acquired, processed, or displayed with minimal real-world time delay.

[0123] "Document information" is text data generated by analyzing voice information.

[0124] A "summary" is a short text that extracts important points from the original document information and expresses them concisely and clearly.

[0125] "Graphical representation" means displaying summarized information in a visually easy-to-understand format, such as bullet points or graphs.

[0126] "User input" refers to an operation performed by a user participating in a remote conference to provide instructions or correction information to the system, and includes voice input.

[0127] "Natural language processing technology" is a technology that uses computers to understand, interpret, and generate human language.

[0128] The "update means" is a function that changes the currently displayed information to new content based on user input.

[0129] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[0130] The server includes a means for acquiring audio information captured during a conference from an online conference platform in real time, the audio information being acquired directly from a remote conference platform (e.g., a web conference system) using an API.

[0131] The acquired voice information is converted into text information in real time using voice recognition technology (for example, a voice conversion service). To convert the voice data into text information, the API of the voice conversion service is used. This text conversion process analyzes the voice information in real time and converts it into text data.

[0132] The converted document information is summarized by the server. This process uses a generative AI model (e.g., generative AI technology) to extract key points using natural language processing technology and generate a concise summary. The generative AI model receives prompts for generating a summary from text data and outputs the summary based on those prompts.

[0133] The generated summary is then visualized using HTML and CSS to format the summary text in bullet points and other formats for easy viewing. The visualized information is designed to help meeting participants grasp the key points at a glance.

[0134] The terminal has a means for displaying the summary and visualization information sent from the server, including the ability to display the summary information using a web browser or a dedicated application.

[0135] Furthermore, if a user wishes to revise the summary or visualized information during a meeting, they can issue instructions through voice input. The device then sends the speech to the server, which then uses speech recognition technology to convert the speech into text. The summary is then updated again using the generative AI model, and the latest information is visualized and displayed on the device.

[0136] Specific examples

[0137] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[0138] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0139] The audio information of the conference is sent to the server and converted into document information in real time as follows:

[0140] > Our new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved.

[0141] This document information is summarized by the generative AI model as follows:

[0142] The new product will be released next month.

[0143] > Enhanced marketing campaigns.

[0144] > Focus on online advertising.

[0145] > The budget has been approved.

[0146] This summary is visually formatted as bullet points and displayed on the terminal as follows:

[0147] The new product will be released next month.

[0148] > Enhanced marketing campaigns.

[0149] > · Focus on online advertising.

[0150] > · Budget approved.

[0151] If the user responds to this summary by voice, saying something like, "Please also include a breakdown of the marketing budget," the server converts the speech into text and updates the summary. This allows the meeting content to be summarized and visualized concisely and accurately in real time, allowing meeting participants to immediately grasp the important points. Furthermore, by updating the summary based on the user's instructions, the meeting proceeds smoothly.

[0152] Prompt sentence input example

[0153] As an example of a prompt sentence for summarizing the contents of a meeting, enter the following:

[0154] > Please summarize the meeting:

[0155] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0156] This system summarizes the contents of remote meetings in real time and provides them visually to participants, making meetings more efficient and enabling immediate understanding of important points. It is also a dynamic system that can quickly reflect corrections from users.

[0157] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0158] Step 1:

[0159] The server acquires audio information from the remote conference platform in real time. Specifically, it uses an API to capture the audio stream during the conference. This audio stream is sent as data in real time from the remote conference platform (e.g., a web conference system).

[0160] Input: Audio stream from remote conferencing platform

[0161] Output: Audio data stored on the server

[0162] Specific working example:

[0163] The server connects to the API of the online conference platform and obtains the audio data.

[0164] The acquired audio data is temporarily stored in storage.

[0165] Step 2:

[0166] The server converts the voice information into text data in real time using voice recognition technology. This process uses a voice conversion API to convert the voice data into text data.

[0167] Input: Saved audio data

[0168] Output: Converted text data

[0169] Specific working example:

[0170] The server sends the audio data to the speech conversion API.

[0171] The text data is returned from the conversion API and temporarily stored in the server's memory.

[0172] Step 3:

[0173] The server uses a generative AI model to summarize the document information. Specifically, it uses natural language processing technology to extract key points and generate a concise summary. In this process, a prompt sentence is input into the generative AI model to generate the summary.

[0174] Input: Converted text data

[0175] Output: Summarized text data

[0176] Specific working example:

[0177] The server sends text data along with a prompt to the generative AI model.

[0178] The summarized text data is received from the generative AI model and stored on the server.

[0179] Step 4:

[0180] The server uses HTML and CSS to format the generated summary visually, allowing the summarized information to be displayed in an easy-to-read format such as bullet points.

[0181] Input: Summarized text data

[0182] Output: Visualized HTML page

[0183] Specific working example:

[0184] The server embeds the summary text into an HTML template.

[0185] Apply CSS to format the summary into bullet points.

[0186] Generates the visualization HTML page and prepares it for display.

[0187] Step 5:

[0188] The terminal receives the visualized HTML page sent from the server and displays it to the user, displaying the summary information through a web browser or dedicated application.

[0189] Input: HTML page sent by the server

[0190] Output: Summary information displayed on the terminal

[0191] Specific working example:

[0192] The terminal periodically sends a request to the server to obtain the latest summary data.

[0193] The HTML page is displayed in a web browser.

[0194] Step 6:

[0195] When the user issues corrections via voice input, the device sends the speech to the server, which again uses speech recognition technology to convert the speech into text information and updates the summary using a generative AI model.

[0196] Input: User voice input

[0197] Output: Updated summary text, visualized information

[0198] Specific working example:

[0199] The user utters, "Please include a breakdown of your marketing budget."

[0200] The device sends voice data to the server, which converts the data into text.

[0201] The server inputs text data into the generative AI model and updates the summary.

[0202] The updated summary text is embedded in the HTML page again and sent to the terminal.

[0203] (Application example 1)

[0204] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0205] There is a demand for more efficient work and accurate information sharing in the field. In particular, in workplaces such as factories, it is important to grasp the situation in real time and summarize and share important information. However, currently, workers manually collect and report information, which creates issues with the accuracy and timeliness of the information. Furthermore, it is time-consuming for workers to correct and update information, which can easily lead to work delays and misunderstandings. To solve these problems, a system is needed that can grasp the situation in real time and summarize and share important information.

[0206] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0207] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for accepting user input for modifying the summarized or visualized information, means for updating the summarized or visualized information based on the user input, means for displaying the summarized or visualized information as target information on a display of a machine, and means for the machine to collect the situation in the workplace as voice data and extract and summarize important information. This makes it possible to grasp the work situation in the field in real time and efficiently communicate and modify important information.

[0208] "Audio data" refers to digital information recorded from human speech, environmental sounds, etc.

[0209] "Real-time" means that information processing, data acquisition, and transmission are carried out immediately.

[0210] "Text data" is data expressed as a string of characters.

[0211] To "summarize" means to extract the important parts from the original information and summarize them concisely.

[0212] "Visualization" is the visual display of information to make it easier to see and understand.

[0213] "User input" refers to information or instructions given to the system by a user.

[0214] A "machine" is a device that performs a specific task automatically.

[0215] A "display" is a screen for displaying information.

[0216] A "workplace" is a place where a specific task is performed.

[0217] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[0218] The present invention relates to a system for efficiently grasping the work status at a site and summarizing and sharing important information. The following describes in detail an embodiment of the present invention.

[0219] The server is equipped with a means for acquiring voice data in real time. Specifically, it uses microphones placed in the workplace to collect voices emitted by workers and machinery. The collected voice data is sent to the server, where it is processed.

[0220] The server then converts the acquired voice data into text data in real time using the SpeechRecognition library. Once the text data is generated, the server uses natural language processing technology (such as OpenAI's GPT-3.5) to extract key points and generate a summary. The summarized information is then visualized for easy viewing.

[0221] Furthermore, the summarized information is displayed on the machine's display. The display is placed in a location where workers can easily see it and is updated in real time. For example, the display may show a specific summary such as, "Part A is low in stock. Replenishment is required."

[0222] If the user wishes to modify the summary or visualization, they can give instructions using voice input, which is then sent back to the server and converted into text data, and the summary or visualization is updated based on the converted text data.

[0223] This system makes it possible to grasp the work status on site in real time, allowing workers to immediately check important information and make corrections as necessary, thereby achieving efficient and accurate information sharing.

[0224] Specific examples

[0225] When a robot worker in a factory says, "Part A is running low on stock. We need to replenish it before the next shift," the voice data is sent to the server. The server converts it into text data and generates a summary that reads, "Part A is running low on stock. We need to replenish it." This summary is then displayed on a screen so that the worker can check it.

[0226] Prompt Sentence Examples

[0227] Please summarize the following:

[0228] "Part A is low in stock. It needs to be replenished before the next shift."

[0229] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0230] Step 1:

[0231] The server acquires voice data in real time using microphones placed in the workplace. The input is the voice from workers and machinery, and the output is to send this voice to the server as digital voice data. Specifically, the microphones capture the voice in the workplace and send it to the server via a data network.

[0232] Step 2:

[0233] The server converts the voice data acquired in real time into text data using the SpeechRecognition library. The input is digital voice data, and the output is text data of character strings extracted from the voice. Specifically, the server calls the SpeechRecognition library, processes the voice data, and generates the corresponding text.

[0234] Step 3:

[0235] The server inputs the generated text data into OpenAI's generative AI model (such as GPT-3.5) to generate a summary. The input is the text data obtained in the previous step, and the output is the summarized text data. Specific operations include the process in which the server sends the text data to the generative AI model and receives the summary.

[0236] Step 4:

[0237] The server performs visualization to display the summarized text data in a visually easy-to-understand manner. The input is the generated summary text, and the output is visually formatted summary information. Specifically, the server formats the summary text into a visual format such as bullet points or color coding, and generates data for display.

[0238] Step 5:

[0239] The terminal displays the visually formatted summary information. The input is the visually formatted summary data sent from the server, and the output is the summary information displayed on the display. Specifically, the terminal receives data from the server and reflects it on the display.

[0240] Step 6:

[0241] The user can use voice input to modify the summary or visualized information. The input is the user's verbal instruction, and the output is the voice data. Specifically, the user gives instructions into the microphone of the device, and the voice is sent to the server.

[0242] Step 7:

[0243] The server converts the voice-input instructions back into text data and updates the summary information. The input is the user's voice data, and the output is the updated summary text data. Specifically, the server calls the SpeechRecognition library again to convert the speech to text and generate a new summary.

[0244] Step 8:

[0245] The terminal redisplays the updated summary information. The input is the updated summary data sent from the server, and the output is the new summary information displayed on the display. Specifically, the terminal receives the new data and reflects it on the display.

[0246] This makes it possible to efficiently grasp the work situation on site in real time and instantly communicate and correct important information.

[0247] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0248] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[0249] Basic System Configuration

[0250] The server has a means for acquiring voice data in real time. The server receives voice data captured during a remote conference and processes it in real time. The server converts the voice data into text data in real time. Voice recognition technology is used here.

[0251] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[0252] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[0253] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[0254] Adding an Emotion Engine

[0255] In the present invention, an emotion engine is further added, which detects the user's emotions in real time from what is said during the meeting and adjusts the summarized and visualized information based on the emotion information.

[0256] The emotion engine analyzes the user's emotions from the voice data and reflects the results in the summary of the text data. Based on this analysis, particularly important topics and points of strong reactions are highlighted, enabling more accurate summaries.

[0257] Furthermore, the emotion engine has the ability to display users' emotional states as visual elements such as graphs and color coding, allowing all meeting participants to grasp the emotional states of other participants in real time, making discussions proceed more smoothly.

[0258] Specific processing examples

[0259] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[0260] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0261] The server receives the audio data and converts it to text in real time:

[0262] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0263] The emotion engine analyzes the voice data and detects different emotional states:

[0264] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[0265] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[0266] 3. Detect "relief" sentiment for "budget already approved."

[0267] The server generates a summary based on this text and emotion information and visualizes it:

[0268] The new product will be released next month. (Positive)

[0269] Strengthening marketing campaigns. (Excitement)

[0270] Focus on online advertising. (Excited)

[0271] The budget has been approved. (Relax)

[0272] The device receives this summary and displays it to the user:

[0273] The new product will be released next month. (Positive)

[0274] Strengthening marketing campaigns. (Excitement)

[0275] Focus on online advertising. (Excited)

[0276] · Budget approved. (Relax)

[0277] User says "Please include a breakdown of your marketing budget":

[0278] "Please include a breakdown of your marketing budget."

[0279] The server receives this voice instruction and updates the summary and emotion information:

[0280] The new product will be released next month. (Positive)

[0281] Marketing campaign enhancements and budget breakdown. (Excitement)

[0282] Focus on online advertising. (Excited)

[0283] The budget has been approved. (Relax)

[0284] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[0285] The processing flow will be explained below.

[0286] Specific flow of program processing

[0287] Server-side processing

[0288] Step 1:

[0289] Setting authentication information

[0290] The server loads the credentials for the emotion engine and speech recognition service, which allow the server to securely access the API.

[0291] Step 2:

[0292] Real-time speech recognition settings

[0293] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[0294] Step 3:

[0295] Creating an Audio Stream

[0296] The server sends a request to the speech recognition service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them into text in real time.

[0297] Step 4:

[0298] Sending data for sentiment analysis

[0299] The server sends the text data and voice data to the emotion engine to analyze the user's emotions, and the emotion engine analyzes the voice data and returns emotion information.

[0300] Step 5:

[0301] Text summary

[0302] The server uses natural language processing techniques to summarize the recognized speech text and emotional information, extracting key points that reflect the emotional state from the long text data and creating a concise summary.

[0303] Step 6:

[0304] Summary visualization

[0305] The server then formats the summarized text and emotional information into a visually understandable format, using bullet points to indicate the text and colors and icons to indicate the emotional state.

[0306] Terminal side processing

[0307] Step 1:

[0308] Audio Stream Input

[0309] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[0310] Step 2:

[0311] Viewing summaries and visualizations

[0312] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[0313] User processing

[0314] Step 1:

[0315] Check and correct statements

[0316] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[0317] Step 2:

[0318] Entering correction instructions

[0319] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[0320] Step 3:

[0321] Processing and updating correction instructions

[0322] The server converts the voice input into text and uses that information to update summaries and visualizations.

[0323] Specific processing examples

[0324] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[0325] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0326] Acquiring voice data and converting it to text

[0327] Step 1:

[0328] The device captures the audio and sends it to the server.

[0329] Step 2:

[0330] The server receives the voice data and converts it into text in real time.

[0331] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0332] sentiment analysis

[0333] Step 3:

[0334] The server sends the text data and voice data to the emotion engine.

[0335] The sentiment engine analyzes sentiment as follows:

[0336] "The new product is scheduled to be released next month." -> Positive

[0337] "We need to step up our marketing campaigns and focus on online advertising." -> Excitement

[0338] "The budget has already been approved." -> Peace of mind

[0339] Summarization and visualization

[0340] Step 4:

[0341] The server generates a summary based on the text and emotion information.

[0342] The new product will be released next month. (Positive)

[0343] Strengthening marketing campaigns. (Excitement)

[0344] Focus on online advertising. (Excited)

[0345] The budget has been approved. (Relax)

[0346] Step 5:

[0347] The server visualizes the summary and sends it to the terminal.

[0348] Step 6:

[0349] The terminal will display a summary and visualization of the information:

[0350] The new product will be released next month. (Positive)

[0351] Strengthening marketing campaigns. (Excitement)

[0352] Focus on online advertising. (Excited)

[0353] · Budget approved. (Relax)

[0354] User-input corrections

[0355] Step 7:

[0356] The user utters, "Please include a breakdown of your marketing budget."

[0357] Step 8:

[0358] The terminal sends a voice instruction to the server.

[0359] Step 9:

[0360] The server converts the voice instructions into text and updates the summary and sentiment information:

[0361] The new product will be released next month. (Positive)

[0362] Marketing campaign enhancements and budget breakdown. (Excitement)

[0363] Focus on online advertising. (Excited)

[0364] The budget has been approved. (Relax)

[0365] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[0366] Example 2

[0367] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0368] Current remote conferencing systems lack a means to summarize the contents of a meeting in real time and grasp key points. Furthermore, there are no systems that provide visual information that accurately reflects the emotional state of users during a meeting. This can result in reduced meeting efficiency and the risk of important points being overlooked. Another problem is that it can be difficult for users to appropriately lead discussions based on their emotions.

[0369] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0370] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for analyzing a user's emotional state based on the summary, means for visualizing the summary and the user's emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input. This makes it easier to grasp important points of the meeting content and the user's emotional state in real time, enabling efficient and smooth discussions to proceed.

[0371] "Audio data" refers to data including speech and conversations of participants during a conference, captured as acoustic signals.

[0372] "Real-time" refers to a situation in which voice and text data is processed instantly with almost no delay.

[0373] "Text data" is character information converted from voice data, and is data that expresses the contents of the conference as text.

[0374] The "summarization means" is a system component that uses natural language processing technology to extract important points from long text data and summarize them in a concise form.

[0375] "Means for analyzing the user's emotional state" refers to technologies and algorithms for detecting the user's emotions from voice-recognized text or speech and analyzing that state.

[0376] A "visualization means" is a system component that has the function of displaying summarized text and emotional information in the form of graphs, color coding, etc., and providing information in a form that is easy for users to understand.

[0377] "Means for accepting user input" refers to an interface that receives voice or text instructions or requests from a user and allows the system to operate in response to those instructions or requests.

[0378] A "means for updating information" is a system component that has the functionality to recalculate and rearrange the summarized and visualized information based on input received from the user, and keep it up to date.

[0379] "Natural language processing technology" is a general term for technologies and algorithms that enable computers to understand and process human language.

[0380] "Voice input" refers to an input method in which a user gives instructions or requests to a system by voice.

[0381] "Meeting content" refers to the overall information, including the topics, comments, proposals, etc. discussed during the remote meeting.

[0382] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[0383] Basic System Configuration

[0384] The server is equipped with a means to acquire voice data in real time. Voice data during remote meetings is acquired through an API and processed in real time. The voice data is converted into text data in real time by the server. Specifically, voice recognition technology is used. For example, an API that provides voice recognition services as a general-purpose voice recognition API is used.

[0385] The converted text data is summarized by the server using natural language processing technology. Natural language processing technology extracts important points from long text data and creates a concise summary. The NLP model used for this summarization is a general-purpose generative AI model. An appropriate summary can be obtained by inputting a prompt sentence into the generative AI model.

[0386] The summarized text is then visualized. This can be done by turning the text into bullet points or formatting it for easier viewing. Furthermore, the visualization combines the output of the emotion engine to reflect the user's emotional state. For example, if a specific emotion is detected, it can be displayed in a different color.

[0387] The terminal has a means for displaying the summary and visualized information sent from the server. Specifically, this information is displayed on a web browser. Conference participants can refer to this information to proceed with the conference. If the user input is voice, the voice is converted into text data, and the summary and visualized information are updated. By performing this process in real time, it is possible to effectively summarize and visualize the contents of the conference.

[0388] Adding an Emotion Engine

[0389] One of the features of the present invention is the addition of an emotion engine. The emotion engine detects users' emotions in real time from audio data during a meeting and adjusts the summaries and visualizations based on this emotional information. The emotion engine incorporates algorithms and techniques for analyzing emotions from audio data. This highlights particularly important topics and points of strong reactions, enabling more accurate summaries. Furthermore, the results of the emotion engine can be displayed as visual elements. For example, the user's emotional state can be displayed in graphs or color-coded text format.

[0390] Specific examples

[0391] For example, suppose a remote meeting is taking place and a discussion is taking place about a new product launch strategy. If the following audio stream exists:

[0392] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0393] The server receives the audio data and converts it to text in real time:

[0394] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0395] The emotion engine analyzes the voice data and detects different emotional states:

[0396] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[0397] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[0398] 3. Detect "relief" sentiment for "budget already approved."

[0399] The server generates a summary based on this text and emotion information and visualizes it:

[0400] The new product will be released next month. (Positive)

[0401] Strengthening marketing campaigns. (Excitement)

[0402] Focus on online advertising. (Excited)

[0403] The budget has been approved. (Relax)

[0404] The terminal receives this summary and displays it to the user:

[0405] The new product will be released next month. (Positive)

[0406] Strengthening marketing campaigns. (Excitement)

[0407] Focus on online advertising. (Excited)

[0408] · Budget approved. (Relax)

[0409] User says "Please include a breakdown of your marketing budget":

[0410] "Please include a breakdown of your marketing budget."

[0411] The server receives this voice instruction and updates the summary and emotion information:

[0412] The new product will be released next month. (Positive)

[0413] Marketing campaign enhancements and budget breakdown. (Excitement)

[0414] Focus on online advertising. (Excited)

[0415] The budget has been approved. (Relax)

[0416] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that reflects the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[0417] Examples of key prompts include:

[0418] "Please summarize the meeting based on the following:

[0419] “We have a new product launch planned for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved.”

[0420] Based on these methods, a system is built that efficiently summarizes and visualizes remote meetings, as well as performs sentiment analysis.

[0421] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0422] Step 1:

[0423] The server acquires the audio data in real time.

[0424] Specific operation: The server obtains audio data from a remote conference platform (e.g., a conference application) in real time through an API. The input is the audio data stream during the conference, and the output is the buffered audio data. The server captures this audio data at regular intervals and stores it in a buffer for the next processing step. For example, the audio stream is sent as regular data packets in real time.

[0425] Step 2:

[0426] The server converts the audio data into text data

[0427] Specific operation: The server sends the acquired voice data to the voice recognition API and converts it into text data in real time. The input is the voice data stored in the buffer, and the output is the text data returned from the voice recognition API. The voice recognition API analyzes the voice signal and generates the corresponding text using a language model. For example, the speech "The new product launch is scheduled for next month" is converted into text "The new product launch is scheduled for next month."

[0428] Step 3:

[0429] The server summarizes the text data

[0430] Specific operation: The server summarizes text data using a generative AI model (e.g., an NLP model). The input is text data obtained from a speech recognition API, and the output is the summarized text. The generative AI model is prompted with "Please summarize the following text" to extract key points. For example, from the text "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved," a short summary is generated: "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved."

[0431] Step 4:

[0432] The server analyzes the user's emotional state

[0433] Specific operation: The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's emotional state based on the text data. The input is summarized text data, and the output is the emotion analysis result. The emotion analysis API detects emotions such as positive, negative, excited, and relieved for each sentence. For example, "The new product launch is scheduled for next month" may detect "positive" emotion, and "excited" emotion may be detected for "We need to strengthen our marketing campaign and focus on online advertising."

[0434] Step 5:

[0435] Server visualizes summary and emotional state

[0436] Specific operation: The server uses a visualization API (e.g., a chart drawing library) to visually display the summarized text and sentiment information. The input is the summarized text data and sentiment analysis results, and the output is the visualized information. The visualization API is used to format the text into bullet points and color-code the sentiment corresponding to each item. For example, the visualization is in the form of "New product launch next month (positive)" or "Marketing campaign intensification (excited)."

[0437] Step 6:

[0438] Device displays visualized information

[0439] Specific operation: The terminal receives visualized information sent from the server and displays it to the user. The input is the visualized text and emotion information sent from the server, and the output is the information displayed on the terminal screen. The terminal provides this information to the user in real time using a web browser or a conference display screen. For example, important points and emotional states during a conference are displayed in color.

[0440] Step 7:

[0441] The user gives additional instructions by voice input, and the server updates the information.

[0442] Specific operation: The user voice-inputs additional information or instructions during a meeting. The input is the user's voice data, and the output is an updated summary and emotional information. The server again sends this voice data to the speech recognition API and converts it into text data. The NLP model and emotional engine are then re-run to update the summary and emotional information. For example, receiving a voice instruction such as "Please also include a breakdown of the marketing budget," the summary and visualization information are reshaped.

[0443] Through this series of processing steps, the contents of remote meetings are summarized and visualized in real time, and information is provided that reflects the user's emotional state. As a result, meeting participants can more easily understand important points and the direction of the discussion, allowing the meeting to proceed smoothly.

[0444] (Application example 2)

[0445] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0446] Conventional remote conferencing systems have limited functionality for grasping and summarizing meeting content, resulting in reduced meeting efficiency. Furthermore, because they do not take into account the emotional state of meeting participants, it can be difficult to accurately understand the flow of discussion, which can lead to meetings not progressing smoothly. Remote meetings, particularly at production sites, require understanding important agenda items and making quick decisions, but there has been a lack of appropriate systems for this purpose.

[0447] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0448] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for recognizing a user's emotional state, means for adjusting the summary and visualized information based on the emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input, thereby enabling efficient understanding of the content of a remote conference and accurate summarization and visualization that take into account the emotional states of participants.

[0449] 1. "Audio Data" means audio information collected during remote meetings or communications.

[0450] 2. "Text data" means data that has been converted from audio data into text information.

[0451] 3. A "summary" is information that extracts important points from long text data and summarizes them concisely.

[0452] 4. "Visualization" means summarizing or presenting information in an easy-to-read format.

[0453] 5. "User Input" means additional information or instructions for correction provided by a conference participant.

[0454] 6. "Emotional state" refers to the emotional state detected based on the content of a user's speech.

[0455] 7. "Natural language processing technology" refers to technology that enables computers to understand, interpret, and manipulate human language.

[0456] 8. "Real-time acquisition means" refers to technologies or devices that allow for instantaneous collection of audio data.

[0457] 9. "Means for converting into text data" refers to technology or devices that automatically convert voice data into text information.

[0458] 10. "Summarizing tools" refers to techniques or devices that extract important information from long texts and summarize it concisely.

[0459] 11. "Visualization tools" refers to techniques and devices that display summaries and information in an easy-to-understand manner using graphs, color coding, etc.

[0460] 12. "Adjusting means" refers to techniques or devices that change summaries or visualizations based on emotional state.

[0461] 13. "Means for accepting user input for modification" refers to technology or devices that accept additional information or modification instructions provided by conference participants.

[0462] 14. "Means for updating" refers to techniques or devices that transform the summary or visualization into new information based on user input.

[0463] 15. "Remote meeting" means a meeting conducted online by multiple participants who are geographically separated.

[0464] This invention is a system for efficiently understanding and summarizing the contents of remote meetings. This system converts the contents of meetings into text in real time, summarizes important points, and visualizes emotional states, providing meeting participants with easy-to-understand information.

[0465] The system is configured as follows:

[0466] Acquiring and converting audio data

[0467] The server captures the voice data generated during the remote conference in real time. Specifically, it converts the voice data into text data using voice recognition technology (e.g., Google Speech-to-Text API). This allows the content of what is said during the conference to be instantly captured as text information.

[0468] Summarizing text data

[0469] The acquired text data is summarized using natural language processing techniques (e.g., the TENSORFLOW® NLP model), which extracts key points from long text data and generates concise summaries.

[0470] User Emotion Recognition

[0471] The server uses an emotion engine (e.g., IBM Watson (registered trademark) Tone Analyzer) to recognize the user's emotional state from the voice data in real time. This detects the emotional state of the speech (positive, excited, relieved, etc.) and reflects the emotional state in the summary information.

[0472] Summarization and emotional state visualization

[0473] The summarized text data and emotional states are displayed in an easy-to-read format using visualization techniques (e.g., D3.js). The text is formatted as bullet points, and the emotional states are visually represented using color coding and icons. For example, positive statements are color-coded green, and excited statements are color-coded red.

[0474] User-input corrections and updates

[0475] The device provides an interface for users to provide corrections and additional information through voice or text input, such as "Please also include a breakdown of the marketing budget." Based on this, the server dynamically updates the summarized and visualized information.

[0476] Specific examples

[0477] For example, in a remote meeting at a production site, if someone says, "We'd like to discuss the introduction of a new production line. To do this, we need to consider how to improve production efficiency and the amount of investment required. What are your thoughts?", the system will process it as follows:

[0478] 1. Acquire audio data and convert it into text data.

[0479] 2. Summarize the text data and extract important topics such as "introduction of a new production line," "improvement of production efficiency," and "amount of investment required."

[0480] 3. The emotion engine detects the emotional state of each comment and assigns a label such as positive, excited, or relieved.

[0481] 4. These summaries and emotional states are visualized and displayed on the screen of a smartphone or head-mounted display.

[0482] Prompt Sentence Examples

[0483] "We would like to discuss the introduction of a new production line. To do so, we need to consider ways to improve production efficiency and the amount of investment required. We would appreciate your opinions."

[0484] This system allows for efficient understanding of the content of remote meetings and accurate summaries and visualizations that take into account the emotional states of participants, which is expected to lead to smoother meeting progress and faster decision-making.

[0485] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0486] Step 1: Acquire audio data

[0487] The server acquires the audio data of the remote conference in real time. As input, it collects the user's conversational voice from a microphone and obtains the data as an audio signal. Specifically, the server captures the audio stream of the conference and saves it as audio data. This data is passed to subsequent processing steps.

[0488] Step 2: Convert audio data to text

[0489] The server converts the acquired voice data into text data in real time. The voice data collected in step 1 is used as input. Specifically, it uses voice recognition technology (e.g., Google Speech-to-Text API) to convert the voice data into text data and output it as text data.

[0490] Step 3: Summarizing the text data

[0491] The server summarizes the converted text data. It uses the text data generated in step 2 as input. Specifically, it uses natural language processing techniques (e.g., TensorFlow NLP models) to extract key points from the text data and generate a concise summary. This summary is used in the next step.

[0492] Step 4: Recognizing your emotional state

[0493] The server recognizes the user's emotional state from the summarized text data. The summary generated in step 3 is used as input. Specifically, it uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from each part of the summary and assigns an emotion tag (positive, excited, relieved, etc.). This generates a summary that reflects the emotional state.

[0494] Step 5: Summarize and visualize emotional states

[0495] The server visualizes the summary and the emotional state. It uses the summary with the emotional state generated in step 4 as input. Specifically, it uses visualization technology (e.g., D3.js) to display the summary in bulleted form and visually represent the emotional state using color coding and icons. This visualization data is sent to the device.

[0496] Step 6: Accepting User Input

[0497] The device accepts input for corrections or additional information provided by the user. The user can provide instructions by voice or text as input. Specifically, in the case of voice input, the device uses voice recognition technology again, and in the case of text input, the device receives the input as text data.

[0498] Step 7: Update the summary and visualization

[0499] The server updates the summarized and visualized information based on the received user input. The input is the user instruction obtained in step 6. Specifically, the server performs a series of processes to re-summarize the updated information, recognize the emotional state, and visualize it. This generates updated summarized information that meets the user's requirements and retransmits it to the device.

[0500] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0501] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0502] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0503] [Second embodiment]

[0504] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0505] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0506] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0507] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0508] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0509] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0510] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0511] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0512] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0513] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0514] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0515] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0516] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[0517] The server has a means for acquiring voice data in real time. Specifically, it receives voice data captured during a remote conference and processes the data. The server converts the voice data into text data in real time using voice recognition technology.

[0518] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[0519] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[0520] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[0521] Furthermore, if the user wishes to modify the summary or visualized information during the process, the device is also provided with a means for accepting user input. This user input can be performed as voice input. When the user issues a command for modification by voice, the device transmits the voice to the server. The server converts the voice input into text data and updates the summary or visualized information based on that information.

[0522] Specific examples

[0523] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[0524] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0525] The audio data of the conference is sent to the server and converted into text data in real time. This text data is summarized as follows:

[0526] The new product will be released next month.

[0527] Enhanced marketing campaigns.

[0528] Focus on online advertising.

[0529] The budget has been approved.

[0530] This summary is visually formatted as a bulleted list and displayed on the terminal, as shown below:

[0531] The new product will be released next month.

[0532] Enhanced marketing campaigns.

[0533] · Focus on online advertising.

[0534] · Budget approved.

[0535] If the user responds to this summary by voice, saying "Please include a breakdown of the marketing budget," the server converts the voice into text and updates the summary.

[0536] The system provides a concise and accurate real-time summary and visualization of meeting content, allowing participants to immediately grasp key points. It also updates the summary based on user instructions, ensuring smooth meeting progress.

[0537] Thus, the present invention provides an embodiment for supporting efficient information sharing and decision-making in remote meetings.

[0538] The processing flow will be explained below.

[0539] Specific flow of program processing

[0540] Server-side processing

[0541] Step 1:

[0542] Setting authentication information

[0543] The server loads the credentials for the Google Cloud Speech-to-Text service, which allows the server to securely access the API.

[0544] Step 2:

[0545] Real-time speech recognition settings

[0546] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[0547] Step 3:

[0548] Creating an Audio Stream

[0549] The server sends a request to the Google Cloud Speech-to-Text service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them to text in real time.

[0550] Step 4:

[0551] Text summary

[0552] The server uses natural language processing techniques to summarize the recognized speech text, extracting key points from long text data and creating a concise summary.

[0553] Step 5:

[0554] Summary visualization

[0555] The server formats the summarized text into a visually easy-to-understand format, specifically by turning the text into bullet points and formatting it in a way that makes it easier to read.

[0556] Terminal side processing

[0557] Step 1:

[0558] Audio Stream Input

[0559] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[0560] Step 2:

[0561] Viewing summaries and visualizations

[0562] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[0563] User processing

[0564] Step 1:

[0565] Check and correct statements

[0566] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[0567] Step 2:

[0568] Entering correction instructions

[0569] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[0570] Step 3:

[0571] Processing and updating correction instructions

[0572] The server converts the voice input into text and uses that information to update summaries and visualizations.

[0573] Specific processing examples

[0574] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[0575] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0576] The server receives the audio data and converts it to text in real time:

[0577] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0578] The server summarizes this text:

[0579] The new product will be released next month.

[0580] Enhanced marketing campaigns.

[0581] Focus on online advertising.

[0582] The budget has been approved.

[0583] The device receives this summary and displays it to the user:

[0584] The new product will be released next month.

[0585] Enhanced marketing campaigns.

[0586] · Focus on online advertising.

[0587] · Budget approved.

[0588] User says "Please include a breakdown of your marketing budget":

[0589] "Please include a breakdown of your marketing budget."

[0590] The server receives this voice prompt and updates the summary:

[0591] The new product will be released next month.

[0592] Marketing campaign enhancements and budget breakdown.

[0593] Focus on online advertising.

[0594] The budget has been approved.

[0595] In this way, a system is realized in which the server, terminals, and users work together to efficiently summarize and visualize the contents of remote meetings and update the information as needed.

[0596] Example 1

[0597] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0598] Conventional remote conferencing systems make it difficult for participants to grasp the content of the meeting in real time. Furthermore, it takes a great deal of time and effort for participants to review important points after the meeting. Furthermore, it is not easy to revise summaries and visualized information in a timely manner during the meeting. To solve these problems, a system is needed that can summarize and visualize the content of the meeting in real time and quickly update it based on user instructions.

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

[0600] In this invention, the server includes means for acquiring voice information in real time, means for converting the voice information into document information in real time, means for summarizing the document information, means for representing the document information in graphics, means for accepting user input, and means for updating the summary or the graphic representation based on the user input. By converting voice information into document information in real time, generating a summary using natural language processing technology, and visualizing the summary, the server enables users to quickly grasp the content of the meeting and further enables information to be updated based on user instructions.

[0601] "Audio information" refers to data recorded in real time of participants' speech during a remote conference.

[0602] "Real-time" means that data is acquired, processed, or displayed with minimal real-world time delay.

[0603] "Document information" is text data generated by analyzing voice information.

[0604] A "summary" is a short text that extracts important points from the original document information and expresses them concisely and clearly.

[0605] "Graphical representation" means displaying summarized information in a visually easy-to-understand format, such as bullet points or graphs.

[0606] "User input" refers to an operation performed by a user participating in a remote conference to provide instructions or correction information to the system, and includes voice input.

[0607] "Natural language processing technology" is a technology that uses computers to understand, interpret, and generate human language.

[0608] The "update means" is a function that changes the currently displayed information to new content based on user input.

[0609] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[0610] The server includes a means for acquiring audio information captured during a conference from an online conference platform in real time, the audio information being acquired directly from a remote conference platform (e.g., a web conference system) using an API.

[0611] The acquired voice information is converted into text information in real time using voice recognition technology (for example, a voice conversion service). To convert the voice data into text information, the API of the voice conversion service is used. This text conversion process analyzes the voice information in real time and converts it into text data.

[0612] The converted document information is summarized by the server. This process uses a generative AI model (e.g., generative AI technology) to extract key points using natural language processing technology and generate a concise summary. The generative AI model receives prompts for generating a summary from text data and outputs the summary based on those prompts.

[0613] The generated summary is then visualized using HTML and CSS to format the summary text in bullet points and other formats for easy viewing. The visualized information is designed to help meeting participants grasp the key points at a glance.

[0614] The terminal has a means for displaying the summary and visualization information sent from the server, including the ability to display the summary information using a web browser or a dedicated application.

[0615] Furthermore, if a user wishes to revise the summary or visualized information during a meeting, they can issue instructions through voice input. The device then sends the speech to the server, which then uses speech recognition technology to convert the speech into text. The summary is then updated again using the generative AI model, and the latest information is visualized and displayed on the device.

[0616] Specific examples

[0617] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[0618] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0619] The audio information of the conference is sent to the server and converted into document information in real time as follows:

[0620] > Our new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved.

[0621] This document information is summarized by the generative AI model as follows:

[0622] The new product will be released next month.

[0623] > Enhanced marketing campaigns.

[0624] > Focus on online advertising.

[0625] > The budget has been approved.

[0626] This summary is visually formatted as bullet points and displayed on the terminal as follows:

[0627] The new product will be released next month.

[0628] > Enhanced marketing campaigns.

[0629] > · Focus on online advertising.

[0630] > · Budget approved.

[0631] If the user responds to this summary by voice, saying something like, "Please also include a breakdown of the marketing budget," the server converts the speech into text and updates the summary. This allows the meeting content to be summarized and visualized concisely and accurately in real time, allowing meeting participants to immediately grasp the important points. Furthermore, by updating the summary based on the user's instructions, the meeting proceeds smoothly.

[0632] Prompt sentence input example

[0633] As an example of a prompt sentence for summarizing the contents of a meeting, enter the following:

[0634] > Please summarize the meeting:

[0635] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0636] This system summarizes the contents of remote meetings in real time and provides them visually to participants, making meetings more efficient and enabling immediate understanding of important points. It is also a dynamic system that can quickly reflect corrections from users.

[0637] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0638] Step 1:

[0639] The server acquires audio information from the remote conference platform in real time. Specifically, it uses an API to capture the audio stream during the conference. This audio stream is sent as data in real time from the remote conference platform (e.g., a web conference system).

[0640] Input: Audio stream from remote conferencing platform

[0641] Output: Audio data stored on the server

[0642] Specific working example:

[0643] The server connects to the API of the online conference platform and obtains the audio data.

[0644] The acquired audio data is temporarily stored in storage.

[0645] Step 2:

[0646] The server converts the voice information into text data in real time using voice recognition technology. This process uses a voice conversion API to convert the voice data into text data.

[0647] Input: Saved audio data

[0648] Output: Converted text data

[0649] Specific working example:

[0650] The server sends the audio data to the speech conversion API.

[0651] The text data is returned from the conversion API and temporarily stored in the server's memory.

[0652] Step 3:

[0653] The server uses a generative AI model to summarize the document information. Specifically, it uses natural language processing technology to extract key points and generate a concise summary. In this process, a prompt sentence is input into the generative AI model to generate the summary.

[0654] Input: Converted text data

[0655] Output: Summarized text data

[0656] Specific working example:

[0657] The server sends text data along with a prompt to the generative AI model.

[0658] The summarized text data is received from the generative AI model and stored on the server.

[0659] Step 4:

[0660] The server uses HTML and CSS to format the generated summary visually, allowing the summarized information to be displayed in an easy-to-read format such as bullet points.

[0661] Input: Summarized text data

[0662] Output: Visualized HTML page

[0663] Specific working example:

[0664] The server embeds the summary text into an HTML template.

[0665] Apply CSS to format the summary into bullet points.

[0666] Generates the visualization HTML page and prepares it for display.

[0667] Step 5:

[0668] The terminal receives the visualized HTML page sent from the server and displays it to the user, displaying the summary information through a web browser or dedicated application.

[0669] Input: HTML page sent by the server

[0670] Output: Summary information displayed on the terminal

[0671] Specific working example:

[0672] The terminal periodically sends a request to the server to obtain the latest summary data.

[0673] The HTML page is displayed in a web browser.

[0674] Step 6:

[0675] When the user issues corrections via voice input, the device sends the speech to the server, which again uses speech recognition technology to convert the speech into text information and updates the summary using a generative AI model.

[0676] Input: User voice input

[0677] Output: Updated summary text, visualized information

[0678] Specific working example:

[0679] The user utters, "Please include a breakdown of your marketing budget."

[0680] The device sends voice data to the server, which converts the data into text.

[0681] The server inputs text data into the generative AI model and updates the summary.

[0682] The updated summary text is embedded in the HTML page again and sent to the terminal.

[0683] (Application example 1)

[0684] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0685] There is a demand for more efficient work and accurate information sharing in the field. In particular, in workplaces such as factories, it is important to grasp the situation in real time and summarize and share important information. However, currently, workers manually collect and report information, which creates issues with the accuracy and timeliness of the information. Furthermore, it is time-consuming for workers to correct and update information, which can easily lead to work delays and misunderstandings. To solve these problems, a system is needed that can grasp the situation in real time and summarize and share important information.

[0686] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0687] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for accepting user input for modifying the summarized or visualized information, means for updating the summarized or visualized information based on the user input, means for displaying the summarized or visualized information as target information on a display of a machine, and means for the machine to collect the situation in the workplace as voice data and extract and summarize important information. This makes it possible to grasp the work situation in the field in real time and efficiently communicate and modify important information.

[0688] "Audio data" refers to digital information recorded from human speech, environmental sounds, etc.

[0689] "Real-time" means that information processing, data acquisition, and transmission are carried out immediately.

[0690] "Text data" is data expressed as a string of characters.

[0691] To "summarize" means to extract the important parts from the original information and summarize them concisely.

[0692] "Visualization" is the visual display of information to make it easier to see and understand.

[0693] "User input" refers to information or instructions given to the system by a user.

[0694] A "machine" is a device that performs a specific task automatically.

[0695] A "display" is a screen for displaying information.

[0696] A "workplace" is a place where a specific task is performed.

[0697] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[0698] The present invention relates to a system for efficiently grasping the work status at a site and summarizing and sharing important information. The following describes in detail an embodiment of the present invention.

[0699] The server is equipped with a means for acquiring voice data in real time. Specifically, it uses microphones placed in the workplace to collect voices emitted by workers and machinery. The collected voice data is sent to the server, where it is processed.

[0700] The server then converts the captured audio data into text data in real time using the SpeechRecognition library. Once the text data is generated, the server uses natural language processing technology (such as OpenAI's GPT-3.5) to extract key points and generate a summary. The summarized information is then visualized for easy viewing.

[0701] Furthermore, the summarized information is displayed on the machine's display. The display is placed in a location where workers can easily see it and is updated in real time. For example, the display may show a specific summary such as, "Part A is low in stock. Replenishment is required."

[0702] If the user wishes to modify the summary or visualization, they can give instructions using voice input, which is then sent back to the server and converted into text data, and the summary or visualization is updated based on the converted text data.

[0703] This system makes it possible to grasp the work status on site in real time, allowing workers to immediately check important information and make corrections as necessary, thereby achieving efficient and accurate information sharing.

[0704] Specific examples

[0705] When a robot worker in a factory says, "Part A is running low on stock. We need to replenish it before the next shift," the voice data is sent to the server. The server converts it into text data and generates a summary that reads, "Part A is running low on stock. We need to replenish it." This summary is then displayed on a screen so that the worker can check it.

[0706] Prompt Sentence Examples

[0707] Please summarize the following:

[0708] "Part A is low in stock. It needs to be replenished before the next shift."

[0709] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0710] Step 1:

[0711] The server acquires voice data in real time using microphones placed in the workplace. The input is the voice from workers and machinery, and the output is to send this voice to the server as digital voice data. Specifically, the microphones capture the voice in the workplace and send it to the server via a data network.

[0712] Step 2:

[0713] The server converts the voice data acquired in real time into text data using the SpeechRecognition library. The input is digital voice data, and the output is text data of character strings extracted from the voice. Specifically, the server calls the SpeechRecognition library, processes the voice data, and generates the corresponding text.

[0714] Step 3:

[0715] The server inputs the generated text data into OpenAI's generative AI model (such as GPT-3.5) to generate a summary. The input is the text data obtained in the previous step, and the output is the summarized text data. Specific operations include the process in which the server sends the text data to the generative AI model and receives the summary.

[0716] Step 4:

[0717] The server performs visualization to display the summarized text data in a visually easy-to-understand manner. The input is the generated summary text, and the output is visually formatted summary information. Specifically, the server formats the summary text into a visual format such as bullet points or color coding, and generates data for display.

[0718] Step 5:

[0719] The terminal displays the visually formatted summary information. The input is the visually formatted summary data sent from the server, and the output is the summary information displayed on the display. Specifically, the terminal receives data from the server and reflects it on the display.

[0720] Step 6:

[0721] The user can use voice input to modify the summary or visualized information. The input is the user's verbal instruction, and the output is the voice data. Specifically, the user gives instructions into the microphone of the device, and the voice is sent to the server.

[0722] Step 7:

[0723] The server converts the voice-input instructions back into text data and updates the summary information. The input is the user's voice data, and the output is the updated summary text data. Specifically, the server calls the SpeechRecognition library again to convert the speech to text and generate a new summary.

[0724] Step 8:

[0725] The terminal redisplays the updated summary information. The input is the updated summary data sent from the server, and the output is the new summary information displayed on the display. Specifically, the terminal receives the new data and reflects it on the display.

[0726] This makes it possible to efficiently grasp the work situation on site in real time and instantly communicate and correct important information.

[0727] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0728] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[0729] Basic System Configuration

[0730] The server has a means for acquiring voice data in real time. The server receives voice data captured during a remote conference and processes it in real time. The server converts the voice data into text data in real time. Voice recognition technology is used here.

[0731] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[0732] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[0733] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[0734] Adding an Emotion Engine

[0735] In the present invention, an emotion engine is further added, which detects the user's emotions in real time from what is said during the meeting and adjusts the summarized and visualized information based on the emotion information.

[0736] The emotion engine analyzes the user's emotions from the voice data and reflects the results in the summary of the text data. Based on this analysis, particularly important topics and points of strong reactions are highlighted, enabling more accurate summaries.

[0737] Furthermore, the emotion engine has the ability to display users' emotional states as visual elements such as graphs and color coding, allowing all meeting participants to grasp the emotional states of other participants in real time, making discussions proceed more smoothly.

[0738] Specific processing examples

[0739] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[0740] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0741] The server receives the audio data and converts it to text in real time:

[0742] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0743] The emotion engine analyzes the voice data and detects different emotional states:

[0744] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[0745] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[0746] 3. Detect "relief" sentiment for "budget already approved."

[0747] The server generates a summary based on this text and emotion information and visualizes it:

[0748] The new product will be released next month. (Positive)

[0749] Strengthening marketing campaigns. (Excitement)

[0750] Focus on online advertising. (Excited)

[0751] The budget has been approved. (Relax)

[0752] The device receives this summary and displays it to the user:

[0753] The new product will be released next month. (Positive)

[0754] Strengthening marketing campaigns. (Excitement)

[0755] Focus on online advertising. (Excited)

[0756] · Budget approved. (Relax)

[0757] User says "Please include a breakdown of your marketing budget":

[0758] "Please include a breakdown of your marketing budget."

[0759] The server receives this voice instruction and updates the summary and emotion information:

[0760] The new product will be released next month. (Positive)

[0761] Marketing campaign enhancements and budget breakdown. (Excitement)

[0762] Focus on online advertising. (Excited)

[0763] The budget has been approved. (Relax)

[0764] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[0765] The processing flow will be explained below.

[0766] Specific flow of program processing

[0767] Server-side processing

[0768] Step 1:

[0769] Setting authentication information

[0770] The server loads the credentials for the emotion engine and speech recognition service, which allow the server to securely access the API.

[0771] Step 2:

[0772] Real-time speech recognition settings

[0773] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[0774] Step 3:

[0775] Creating an Audio Stream

[0776] The server sends a request to the speech recognition service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them into text in real time.

[0777] Step 4:

[0778] Sending data for sentiment analysis

[0779] The server sends the text data and voice data to the emotion engine to analyze the user's emotions, and the emotion engine analyzes the voice data and returns emotion information.

[0780] Step 5:

[0781] Text summary

[0782] The server uses natural language processing techniques to summarize the recognized speech text and emotional information, extracting key points that reflect the emotional state from the long text data and creating a concise summary.

[0783] Step 6:

[0784] Summary visualization

[0785] The server then formats the summarized text and emotional information into a visually understandable format, using bullet points to indicate the text and colors and icons to indicate the emotional state.

[0786] Terminal side processing

[0787] Step 1:

[0788] Audio Stream Input

[0789] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[0790] Step 2:

[0791] Viewing summaries and visualizations

[0792] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[0793] User processing

[0794] Step 1:

[0795] Check and correct statements

[0796] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[0797] Step 2:

[0798] Entering correction instructions

[0799] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[0800] Step 3:

[0801] Processing and updating correction instructions

[0802] The server converts the voice input into text and uses that information to update summaries and visualizations.

[0803] Specific processing examples

[0804] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[0805] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0806] Acquiring voice data and converting it to text

[0807] Step 1:

[0808] The device captures the audio and sends it to the server.

[0809] Step 2:

[0810] The server receives the voice data and converts it into text in real time.

[0811] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0812] sentiment analysis

[0813] Step 3:

[0814] The server sends the text data and voice data to the emotion engine.

[0815] The sentiment engine analyzes sentiment as follows:

[0816] "The new product is scheduled to be released next month." -> Positive

[0817] "We need to step up our marketing campaigns and focus on online advertising." -> Excitement

[0818] "The budget has already been approved." -> Peace of mind

[0819] Summarization and visualization

[0820] Step 4:

[0821] The server generates a summary based on the text and emotion information.

[0822] The new product will be released next month. (Positive)

[0823] Strengthening marketing campaigns. (Excitement)

[0824] Focus on online advertising. (Excited)

[0825] The budget has been approved. (Relax)

[0826] Step 5:

[0827] The server visualizes the summary and sends it to the terminal.

[0828] Step 6:

[0829] The terminal will display a summary and visualization of the information:

[0830] The new product will be released next month. (Positive)

[0831] Strengthening marketing campaigns. (Excitement)

[0832] Focus on online advertising. (Excited)

[0833] · Budget approved. (Relax)

[0834] User-input corrections

[0835] Step 7:

[0836] The user utters, "Please include a breakdown of your marketing budget."

[0837] Step 8:

[0838] The terminal sends a voice instruction to the server.

[0839] Step 9:

[0840] The server converts the voice instructions into text and updates the summary and sentiment information:

[0841] The new product will be released next month. (Positive)

[0842] Marketing campaign enhancements and budget breakdown. (Excitement)

[0843] Focus on online advertising. (Excited)

[0844] The budget has been approved. (Relax)

[0845] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[0846] Example 2

[0847] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0848] Current remote conferencing systems lack a means to summarize the contents of a meeting in real time and grasp key points. Furthermore, there are no systems that provide visual information that accurately reflects the emotional state of users during a meeting. This can result in reduced meeting efficiency and the risk of important points being overlooked. Another problem is that it can be difficult for users to appropriately lead discussions based on their emotions.

[0849] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0850] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for analyzing a user's emotional state based on the summary, means for visualizing the summary and the user's emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input. This makes it easier to grasp important points of the meeting content and the user's emotional state in real time, enabling efficient and smooth discussions to proceed.

[0851] "Audio data" refers to data including speech and conversations of participants during a conference, captured as acoustic signals.

[0852] "Real-time" refers to a situation in which voice and text data is processed instantly with almost no delay.

[0853] "Text data" is character information converted from voice data, and is data that expresses the contents of the conference as text.

[0854] The "summarization means" is a system component that uses natural language processing technology to extract important points from long text data and summarize them in a concise form.

[0855] "Means for analyzing the user's emotional state" refers to technologies and algorithms for detecting the user's emotions from voice-recognized text or speech and analyzing that state.

[0856] A "visualization means" is a system component that has the function of displaying summarized text and emotional information in the form of graphs, color coding, etc., and providing information in a form that is easy for users to understand.

[0857] "Means for accepting user input" refers to an interface that receives voice or text instructions or requests from a user and allows the system to operate in response to those instructions or requests.

[0858] A "means for updating information" is a system component that has the functionality to recalculate and rearrange the summarized and visualized information based on input received from the user, and keep it up to date.

[0859] "Natural language processing technology" is a general term for technologies and algorithms that enable computers to understand and process human language.

[0860] "Voice input" refers to an input method in which a user gives instructions or requests to a system by voice.

[0861] "Meeting content" refers to the overall information, including the topics, comments, proposals, etc. discussed during the remote meeting.

[0862] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[0863] Basic System Configuration

[0864] The server is equipped with a means to acquire voice data in real time. Voice data during remote meetings is acquired through an API and processed in real time. The voice data is converted into text data in real time by the server. Specifically, voice recognition technology is used. For example, an API that provides voice recognition services as a general-purpose voice recognition API is used.

[0865] The converted text data is summarized by the server using natural language processing technology. Natural language processing technology extracts important points from long text data and creates a concise summary. The NLP model used for this summarization is a general-purpose generative AI model. An appropriate summary can be obtained by inputting a prompt sentence into the generative AI model.

[0866] The summarized text is then visualized. This can be done by turning the text into bullet points or formatting it for easier viewing. Furthermore, the visualization combines the output of the emotion engine to reflect the user's emotional state. For example, if a specific emotion is detected, it can be displayed in a different color.

[0867] The terminal has a means for displaying the summary and visualized information sent from the server. Specifically, this information is displayed on a web browser. Conference participants can refer to this information to proceed with the conference. If the user input is voice, the voice is converted into text data, and the summary and visualized information are updated. By performing this process in real time, it is possible to effectively summarize and visualize the contents of the conference.

[0868] Adding an Emotion Engine

[0869] One of the features of the present invention is the addition of an emotion engine. The emotion engine detects users' emotions in real time from audio data during a meeting and adjusts the summaries and visualizations based on this emotional information. The emotion engine incorporates algorithms and techniques for analyzing emotions from audio data. This highlights particularly important topics and points of strong reactions, enabling more accurate summaries. Furthermore, the results of the emotion engine can be displayed as visual elements. For example, the user's emotional state can be displayed in graphs or color-coded text format.

[0870] Specific examples

[0871] For example, suppose a remote meeting is taking place and a discussion is taking place about a new product launch strategy. If the following audio stream exists:

[0872] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0873] The server receives the audio data and converts it to text in real time:

[0874] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0875] The emotion engine analyzes the voice data and detects different emotional states:

[0876] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[0877] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[0878] 3. Detect "relief" sentiment for "budget already approved."

[0879] The server generates a summary based on this text and emotion information and visualizes it:

[0880] The new product will be released next month. (Positive)

[0881] Strengthening marketing campaigns. (Excitement)

[0882] Focus on online advertising. (Excited)

[0883] The budget has been approved. (Relax)

[0884] The terminal receives this summary and displays it to the user:

[0885] The new product will be released next month. (Positive)

[0886] Strengthening marketing campaigns. (Excitement)

[0887] Focus on online advertising. (Excited)

[0888] · Budget approved. (Relax)

[0889] User says "Please include a breakdown of your marketing budget":

[0890] "Please include a breakdown of your marketing budget."

[0891] The server receives this voice instruction and updates the summary and emotion information:

[0892] The new product will be released next month. (Positive)

[0893] Marketing campaign enhancements and budget breakdown. (Excitement)

[0894] Focus on online advertising. (Excited)

[0895] The budget has been approved. (Relax)

[0896] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that reflects the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[0897] Examples of key prompts include:

[0898] "Please summarize the meeting based on the following:

[0899] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[0900] Based on these methods, a system is built that efficiently summarizes and visualizes remote meetings, as well as performs sentiment analysis.

[0901] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0902] Step 1:

[0903] The server acquires the audio data in real time.

[0904] Specific operation: The server obtains audio data from a remote conference platform (e.g., a conference application) in real time through an API. The input is the audio data stream during the conference, and the output is the buffered audio data. The server captures this audio data at regular intervals and stores it in a buffer for the next processing step. For example, the audio stream is sent as regular data packets in real time.

[0905] Step 2:

[0906] The server converts the audio data into text data

[0907] Specific operation: The server sends the acquired voice data to the voice recognition API and converts it into text data in real time. The input is the voice data stored in the buffer, and the output is the text data returned from the voice recognition API. The voice recognition API analyzes the voice signal and generates the corresponding text using a language model. For example, the speech "The new product launch is scheduled for next month" is converted into text "The new product launch is scheduled for next month."

[0908] Step 3:

[0909] The server summarizes the text data

[0910] Specific operation: The server summarizes text data using a generative AI model (e.g., an NLP model). The input is text data obtained from a speech recognition API, and the output is the summarized text. The generative AI model is prompted with "Please summarize the following text" to extract key points. For example, from the text "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved," a short summary is generated: "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved."

[0911] Step 4:

[0912] The server analyzes the user's emotional state

[0913] Specific operation: The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's emotional state based on the text data. The input is summarized text data, and the output is the emotion analysis result. The emotion analysis API detects emotions such as positive, negative, excited, and relieved for each sentence. For example, "The new product launch is scheduled for next month" may detect "positive" emotion, and "excited" emotion may be detected for "We need to strengthen our marketing campaign and focus on online advertising."

[0914] Step 5:

[0915] Server visualizes summary and emotional state

[0916] Specific operation: The server uses a visualization API (e.g., a chart drawing library) to visually display the summarized text and sentiment information. The input is the summarized text data and sentiment analysis results, and the output is the visualized information. The visualization API is used to format the text into bullet points and color-code the sentiment corresponding to each item. For example, the visualization is in the form of "New product launch next month (positive)" or "Marketing campaign intensification (excited)."

[0917] Step 6:

[0918] Device displays visualized information

[0919] Specific operation: The terminal receives visualized information sent from the server and displays it to the user. The input is the visualized text and emotion information sent from the server, and the output is the information displayed on the terminal screen. The terminal provides this information to the user in real time using a web browser or a conference display screen. For example, important points and emotional states during a conference are displayed in color.

[0920] Step 7:

[0921] The user gives additional instructions by voice input, and the server updates the information.

[0922] Specific operation: The user voice-inputs additional information or instructions during a meeting. The input is the user's voice data, and the output is an updated summary and emotional information. The server again sends this voice data to the speech recognition API and converts it into text data. The NLP model and emotional engine are then re-run to update the summary and emotional information. For example, receiving a voice instruction such as "Please also include a breakdown of the marketing budget," the summary and visualization information are reshaped.

[0923] Through this series of processing steps, the contents of remote meetings are summarized and visualized in real time, and information is provided that reflects the user's emotional state. As a result, meeting participants can more easily understand important points and the direction of the discussion, allowing the meeting to proceed smoothly.

[0924] (Application example 2)

[0925] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0926] Conventional remote conferencing systems have limited functionality for grasping and summarizing meeting content, resulting in reduced meeting efficiency. Furthermore, because they do not take into account the emotional state of meeting participants, it can be difficult to accurately understand the flow of discussion, which can lead to meetings not progressing smoothly. Remote meetings, particularly at production sites, require understanding important agenda items and making quick decisions, but there has been a lack of appropriate systems for this purpose.

[0927] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0928] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for recognizing a user's emotional state, means for adjusting the summary and visualized information based on the emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input, thereby enabling efficient understanding of the content of a remote conference and accurate summarization and visualization that take into account the emotional states of participants.

[0929] 1. "Audio Data" means audio information collected during remote meetings or communications.

[0930] 2. "Text data" means data that has been converted from audio data into text information.

[0931] 3. A "summary" is information that extracts important points from long text data and summarizes them concisely.

[0932] 4. "Visualization" means summarizing or presenting information in an easy-to-read format.

[0933] 5. "User Input" means additional information or instructions for correction provided by a conference participant.

[0934] 6. "Emotional state" refers to the emotional state detected based on the content of a user's speech.

[0935] 7. "Natural language processing technology" refers to technology that enables computers to understand, interpret, and manipulate human language.

[0936] 8. "Real-time acquisition means" refers to technologies or devices that allow for instantaneous collection of audio data.

[0937] 9. "Means for converting into text data" refers to technology or devices that automatically convert voice data into text information.

[0938] 10. "Summarizing tools" refers to techniques or devices that extract important information from long texts and summarize it concisely.

[0939] 11. "Visualization tools" refers to techniques and devices that display summaries and information in an easy-to-understand manner using graphs, color coding, etc.

[0940] 12. "Adjusting means" refers to techniques or devices that change summaries or visualizations based on emotional state.

[0941] 13. "Means for accepting user input for modification" refers to technology or devices that accept additional information or modification instructions provided by conference participants.

[0942] 14. "Means for updating" refers to techniques or devices that transform the summary or visualization into new information based on user input.

[0943] 15. "Remote meeting" means a meeting conducted online by multiple participants who are geographically separated.

[0944] This invention is a system for efficiently understanding and summarizing the contents of remote meetings. This system converts the contents of meetings into text in real time, summarizes important points, and visualizes emotional states, providing meeting participants with easy-to-understand information.

[0945] The system is configured as follows:

[0946] Acquiring and converting audio data

[0947] The server captures the voice data generated during the remote conference in real time. Specifically, it converts the voice data into text data using voice recognition technology (e.g., Google Speech-to-Text API). This allows the content of what is said during the conference to be instantly captured as text information.

[0948] Summarizing text data

[0949] The acquired text data is summarized using natural language processing techniques (e.g., TensorFlow NLP models), which extract key points from long text data and generate concise summaries.

[0950] User Emotion Recognition

[0951] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotional state from the speech data in real time. This detects the emotional state of the speech (positive, excited, relieved, etc.) and reflects the emotional state in the summary information.

[0952] Summarization and emotional state visualization

[0953] The summarized text data and emotional states are displayed in an easy-to-read format using visualization techniques (e.g., D3.js). The text is formatted as bullet points, and the emotional states are visually represented using color coding and icons. For example, positive statements are color-coded green, and excited statements are color-coded red.

[0954] User-input corrections and updates

[0955] The device provides an interface for users to provide corrections and additional information through voice or text input, such as "Please also include a breakdown of the marketing budget." Based on this, the server dynamically updates the summarized and visualized information.

[0956] Specific examples

[0957] For example, in a remote meeting at a production site, if someone says, "We'd like to discuss the introduction of a new production line. To do this, we need to consider how to improve production efficiency and the amount of investment required. What are your thoughts?", the system will process it as follows:

[0958] 1. Acquire audio data and convert it into text data.

[0959] 2. Summarize the text data and extract important topics such as "introduction of a new production line," "improvement of production efficiency," and "amount of investment required."

[0960] 3. The emotion engine detects the emotional state of each comment and assigns a label such as positive, excited, or relieved.

[0961] 4. These summaries and emotional states are visualized and displayed on the screen of a smartphone or head-mounted display.

[0962] Prompt Sentence Examples

[0963] "We would like to discuss the introduction of a new production line. To do so, we need to consider ways to improve production efficiency and the amount of investment required. We would appreciate your opinions."

[0964] This system allows for efficient understanding of the content of remote meetings and accurate summaries and visualizations that take into account the emotional states of participants, which is expected to lead to smoother meeting progress and faster decision-making.

[0965] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0966] Step 1: Acquire audio data

[0967] The server acquires the audio data of the remote conference in real time. As input, it collects the user's conversational voice from a microphone and obtains the data as an audio signal. Specifically, the server captures the audio stream of the conference and saves it as audio data. This data is passed to subsequent processing steps.

[0968] Step 2: Convert audio data to text

[0969] The server converts the acquired voice data into text data in real time. The voice data collected in step 1 is used as input. Specifically, it uses voice recognition technology (e.g., Google Speech-to-Text API) to convert the voice data into text data and output it as text data.

[0970] Step 3: Summarizing the text data

[0971] The server summarizes the converted text data. It uses the text data generated in step 2 as input. Specifically, it uses natural language processing techniques (e.g., TensorFlow NLP models) to extract key points from the text data and generate a concise summary. This summary is used in the next step.

[0972] Step 4: Recognizing your emotional state

[0973] The server recognizes the user's emotional state from the summarized text data. The summary generated in step 3 is used as input. Specifically, it uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from each part of the summary and assigns an emotion tag (positive, excited, relieved, etc.). This generates a summary that reflects the emotional state.

[0974] Step 5: Summarize and visualize emotional states

[0975] The server visualizes the summary and the emotional state. It uses the summary with the emotional state generated in step 4 as input. Specifically, it uses visualization technology (e.g., D3.js) to display the summary in bulleted form and visually represent the emotional state using color coding and icons. This visualization data is sent to the device.

[0976] Step 6: Accepting User Input

[0977] The device accepts input for corrections or additional information provided by the user. The user can provide instructions by voice or text as input. Specifically, in the case of voice input, the device uses voice recognition technology again, and in the case of text input, the device receives the input as text data.

[0978] Step 7: Update the summary and visualization

[0979] The server updates the summarized and visualized information based on the received user input. The input is the user instruction obtained in step 6. Specifically, the server performs a series of processes to re-summarize the updated information, recognize the emotional state, and visualize it. This generates updated summarized information that meets the user's requirements and retransmits it to the device.

[0980] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0981] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0982] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0983] [Third embodiment]

[0984] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0985] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0986] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0987] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0988] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0989] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0990] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0991] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0992] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0993] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0994] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0995] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0996] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[0997] The server has a means for acquiring voice data in real time. Specifically, it receives voice data captured during a remote conference and processes the data. The server converts the voice data into text data in real time using voice recognition technology.

[0998] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[0999] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[1000] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[1001] Furthermore, if the user wishes to modify the summary or visualized information during the process, the device is also provided with a means for accepting user input. This user input can be performed as voice input. When the user issues a command for modification by voice, the device transmits the voice to the server. The server converts the voice input into text data and updates the summary or visualized information based on that information.

[1002] Specific examples

[1003] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[1004] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1005] The audio data of the conference is sent to the server and converted into text data in real time. This text data is summarized as follows:

[1006] The new product will be released next month.

[1007] Enhanced marketing campaigns.

[1008] Focus on online advertising.

[1009] The budget has been approved.

[1010] This summary is visually formatted as a bulleted list and displayed on the terminal, as shown below:

[1011] The new product will be released next month.

[1012] Enhanced marketing campaigns.

[1013] · Focus on online advertising.

[1014] · Budget approved.

[1015] If the user responds to this summary by voice, saying "Please include a breakdown of the marketing budget," the server converts the voice into text and updates the summary.

[1016] The system provides a concise and accurate real-time summary and visualization of meeting content, allowing participants to immediately grasp key points. It also updates the summary based on user instructions, ensuring smooth meeting progress.

[1017] Thus, the present invention provides an embodiment for supporting efficient information sharing and decision-making in remote meetings.

[1018] The processing flow will be explained below.

[1019] Specific flow of program processing

[1020] Server-side processing

[1021] Step 1:

[1022] Setting authentication information

[1023] The server loads the credentials for the Google Cloud Speech-to-Text service, which allows the server to securely access the API.

[1024] Step 2:

[1025] Real-time speech recognition settings

[1026] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[1027] Step 3:

[1028] Creating an Audio Stream

[1029] The server sends a request to the Google Cloud Speech-to-Text service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them to text in real time.

[1030] Step 4:

[1031] Text summary

[1032] The server uses natural language processing techniques to summarize the recognized speech text, extracting key points from long text data and creating a concise summary.

[1033] Step 5:

[1034] Summary visualization

[1035] The server formats the summarized text into a visually easy-to-understand format, specifically by turning the text into bullet points and formatting it in a way that makes it easier to read.

[1036] Terminal side processing

[1037] Step 1:

[1038] Audio Stream Input

[1039] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[1040] Step 2:

[1041] Viewing summaries and visualizations

[1042] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[1043] User processing

[1044] Step 1:

[1045] Check and correct statements

[1046] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[1047] Step 2:

[1048] Entering correction instructions

[1049] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[1050] Step 3:

[1051] Processing and updating correction instructions

[1052] The server converts the voice input into text and uses that information to update summaries and visualizations.

[1053] Specific processing examples

[1054] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[1055] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1056] The server receives the audio data and converts it to text in real time:

[1057] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1058] The server summarizes this text:

[1059] The new product will be released next month.

[1060] Enhanced marketing campaigns.

[1061] Focus on online advertising.

[1062] The budget has been approved.

[1063] The device receives this summary and displays it to the user:

[1064] The new product will be released next month.

[1065] Enhanced marketing campaigns.

[1066] · Focus on online advertising.

[1067] · Budget approved.

[1068] User says "Please include a breakdown of your marketing budget":

[1069] "Please include a breakdown of your marketing budget."

[1070] The server receives this voice prompt and updates the summary:

[1071] The new product will be released next month.

[1072] Marketing campaign enhancements and budget breakdown.

[1073] Focus on online advertising.

[1074] The budget has been approved.

[1075] In this way, a system is realized in which the server, terminals, and users work together to efficiently summarize and visualize the contents of remote meetings and update the information as needed.

[1076] Example 1

[1077] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1078] Conventional remote conferencing systems make it difficult for participants to grasp the content of the meeting in real time. Furthermore, it takes a great deal of time and effort for participants to review important points after the meeting. Furthermore, it is not easy to revise summaries and visualized information in a timely manner during the meeting. To solve these problems, a system is needed that can summarize and visualize the content of the meeting in real time and quickly update it based on user instructions.

[1079] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1080] In this invention, the server includes means for acquiring voice information in real time, means for converting the voice information into document information in real time, means for summarizing the document information, means for representing the document information in graphics, means for accepting user input, and means for updating the summary or the graphic representation based on the user input. By converting voice information into document information in real time, generating a summary using natural language processing technology, and visualizing the summary, the server enables users to quickly grasp the content of the meeting and further enables information to be updated based on user instructions.

[1081] "Audio information" refers to data recorded in real time of participants' speech during a remote conference.

[1082] "Real-time" means that data is acquired, processed, or displayed with minimal real-world time delay.

[1083] "Document information" is text data generated by analyzing voice information.

[1084] A "summary" is a short text that extracts important points from the original document information and expresses them concisely and clearly.

[1085] "Graphical representation" means displaying summarized information in a visually easy-to-understand format, such as bullet points or graphs.

[1086] "User input" refers to an operation performed by a user participating in a remote conference to provide instructions or correction information to the system, and includes voice input.

[1087] "Natural language processing technology" is a technology that uses computers to understand, interpret, and generate human language.

[1088] The "update means" is a function that changes the currently displayed information to new content based on user input.

[1089] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[1090] The server includes a means for acquiring audio information captured during a conference from an online conference platform in real time, the audio information being acquired directly from a remote conference platform (e.g., a web conference system) using an API.

[1091] The acquired voice information is converted into text information in real time using voice recognition technology (for example, a voice conversion service). To convert the voice data into text information, the API of the voice conversion service is used. This text conversion process analyzes the voice information in real time and converts it into text data.

[1092] The converted document information is summarized by the server. This process uses a generative AI model (e.g., generative AI technology) to extract key points using natural language processing technology and generate a concise summary. The generative AI model receives prompts for generating a summary from text data and outputs the summary based on those prompts.

[1093] The generated summary is then visualized using HTML and CSS to format the summary text in bullet points and other formats for easy viewing. The visualized information is designed to help meeting participants grasp the key points at a glance.

[1094] The terminal has a means for displaying the summary and visualization information sent from the server, including the ability to display the summary information using a web browser or a dedicated application.

[1095] Furthermore, if a user wishes to revise the summary or visualized information during a meeting, they can issue instructions through voice input. The device then sends the speech to the server, which then uses speech recognition technology to convert the speech into text. The summary is then updated again using the generative AI model, and the latest information is visualized and displayed on the device.

[1096] Specific examples

[1097] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[1098] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1099] The audio information of the conference is sent to the server and converted into document information in real time as follows:

[1100] > Our new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved.

[1101] This document information is summarized by the generative AI model as follows:

[1102] The new product will be released next month.

[1103] > Enhanced marketing campaigns.

[1104] > Focus on online advertising.

[1105] > The budget has been approved.

[1106] This summary is visually formatted as bullet points and displayed on the terminal as follows:

[1107] The new product will be released next month.

[1108] > Enhanced marketing campaigns.

[1109] > · Focus on online advertising.

[1110] > · Budget approved.

[1111] If the user responds to this summary by voice, saying something like, "Please also include a breakdown of the marketing budget," the server converts the speech into text and updates the summary. This allows the meeting content to be summarized and visualized concisely and accurately in real time, allowing meeting participants to immediately grasp the important points. Furthermore, by updating the summary based on the user's instructions, the meeting proceeds smoothly.

[1112] Prompt sentence input example

[1113] As an example of a prompt sentence for summarizing the contents of a meeting, enter the following:

[1114] > Please summarize the meeting:

[1115] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1116] This system summarizes the contents of remote meetings in real time and provides them visually to participants, making meetings more efficient and enabling immediate understanding of important points. It is also a dynamic system that can quickly reflect corrections from users.

[1117] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1118] Step 1:

[1119] The server acquires audio information from the remote conference platform in real time. Specifically, it uses an API to capture the audio stream during the conference. This audio stream is sent as data in real time from the remote conference platform (e.g., a web conference system).

[1120] Input: Audio stream from remote conferencing platform

[1121] Output: Audio data stored on the server

[1122] Specific working example:

[1123] The server connects to the API of the online conference platform and obtains the audio data.

[1124] The acquired audio data is temporarily stored in storage.

[1125] Step 2:

[1126] The server converts the voice information into text data in real time using voice recognition technology. This process uses a voice conversion API to convert the voice data into text data.

[1127] Input: Saved audio data

[1128] Output: Converted text data

[1129] Specific working example:

[1130] The server sends the audio data to the speech conversion API.

[1131] The text data is returned from the conversion API and temporarily stored in the server's memory.

[1132] Step 3:

[1133] The server uses a generative AI model to summarize the document information. Specifically, it uses natural language processing technology to extract key points and generate a concise summary. In this process, a prompt sentence is input into the generative AI model to generate the summary.

[1134] Input: Converted text data

[1135] Output: Summarized text data

[1136] Specific working example:

[1137] The server sends text data along with a prompt to the generative AI model.

[1138] The summarized text data is received from the generative AI model and stored on the server.

[1139] Step 4:

[1140] The server uses HTML and CSS to format the generated summary visually, allowing the summarized information to be displayed in an easy-to-read format such as bullet points.

[1141] Input: Summarized text data

[1142] Output: Visualized HTML page

[1143] Specific working example:

[1144] The server embeds the summary text into an HTML template.

[1145] Apply CSS to format the summary into bullet points.

[1146] Generates the visualization HTML page and prepares it for display.

[1147] Step 5:

[1148] The terminal receives the visualized HTML page sent from the server and displays it to the user, displaying the summary information through a web browser or dedicated application.

[1149] Input: HTML page sent by the server

[1150] Output: Summary information displayed on the terminal

[1151] Specific working example:

[1152] The terminal periodically sends a request to the server to obtain the latest summary data.

[1153] The HTML page is displayed in a web browser.

[1154] Step 6:

[1155] When the user issues corrections via voice input, the device sends the speech to the server, which again uses speech recognition technology to convert the speech into text information and updates the summary using a generative AI model.

[1156] Input: User voice input

[1157] Output: Updated summary text, visualized information

[1158] Specific working example:

[1159] The user utters, "Please include a breakdown of your marketing budget."

[1160] The device sends voice data to the server, which converts the data into text.

[1161] The server inputs text data into the generative AI model and updates the summary.

[1162] The updated summary text is embedded in the HTML page again and sent to the terminal.

[1163] (Application example 1)

[1164] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1165] There is a demand for more efficient work and accurate information sharing in the field. In particular, in workplaces such as factories, it is important to grasp the situation in real time and summarize and share important information. However, currently, workers manually collect and report information, which creates issues with the accuracy and timeliness of the information. Furthermore, it is time-consuming for workers to correct and update information, which can easily lead to work delays and misunderstandings. To solve these problems, a system is needed that can grasp the situation in real time and summarize and share important information.

[1166] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1167] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for accepting user input for modifying the summarized or visualized information, means for updating the summarized or visualized information based on the user input, means for displaying the summarized or visualized information as target information on a display of a machine, and means for the machine to collect the situation in the workplace as voice data and extract and summarize important information. This makes it possible to grasp the work situation in the field in real time and efficiently communicate and modify important information.

[1168] "Audio data" refers to digital information recorded from human speech, environmental sounds, etc.

[1169] "Real-time" means that information processing, data acquisition, and transmission are carried out immediately.

[1170] "Text data" is data expressed as a string of characters.

[1171] To "summarize" means to extract the important parts from the original information and summarize them concisely.

[1172] "Visualization" is the visual display of information to make it easier to see and understand.

[1173] "User input" refers to information or instructions given to the system by a user.

[1174] A "machine" is a device that performs a specific task automatically.

[1175] A "display" is a screen for displaying information.

[1176] A "workplace" is a place where a specific task is performed.

[1177] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[1178] The present invention relates to a system for efficiently grasping the work status at a site and summarizing and sharing important information. The following describes in detail an embodiment of the present invention.

[1179] The server is equipped with a means for acquiring voice data in real time. Specifically, it uses microphones placed in the workplace to collect voices emitted by workers and machinery. The collected voice data is sent to the server, where it is processed.

[1180] The server then converts the captured audio data into text data in real time using the SpeechRecognition library. Once the text data is generated, the server uses natural language processing technology (such as OpenAI's GPT-3.5) to extract key points and generate a summary. The summarized information is then visualized for easy viewing.

[1181] Furthermore, the summarized information is displayed on the machine's display. The display is placed in a location where workers can easily see it and is updated in real time. For example, the display may show a specific summary such as, "Part A is low in stock. Replenishment is required."

[1182] If the user wishes to modify the summary or visualization, they can give instructions using voice input, which is then sent back to the server and converted into text data, and the summary or visualization is updated based on the converted text data.

[1183] This system makes it possible to grasp the work status on site in real time, allowing workers to immediately check important information and make corrections as necessary, thereby achieving efficient and accurate information sharing.

[1184] Specific examples

[1185] When a robot worker in a factory says, "Part A is running low on stock. We need to replenish it before the next shift," the voice data is sent to the server. The server converts it into text data and generates a summary that reads, "Part A is running low on stock. We need to replenish it." This summary is then displayed on a screen so that the worker can check it.

[1186] Prompt Sentence Examples

[1187] Please summarize the following:

[1188] "Part A is low in stock. It needs to be replenished before the next shift."

[1189] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1190] Step 1:

[1191] The server acquires voice data in real time using microphones placed in the workplace. The input is the voice from workers and machinery, and the output is to send this voice to the server as digital voice data. Specifically, the microphones capture the voice in the workplace and send it to the server via a data network.

[1192] Step 2:

[1193] The server converts the voice data acquired in real time into text data using the SpeechRecognition library. The input is digital voice data, and the output is text data of character strings extracted from the voice. Specifically, the server calls the SpeechRecognition library, processes the voice data, and generates the corresponding text.

[1194] Step 3:

[1195] The server inputs the generated text data into OpenAI's generative AI model (such as GPT-3.5) to generate a summary. The input is the text data obtained in the previous step, and the output is the summarized text data. Specific operations include the process in which the server sends the text data to the generative AI model and receives the summary.

[1196] Step 4:

[1197] The server performs visualization to display the summarized text data in a visually easy-to-understand manner. The input is the generated summary text, and the output is visually formatted summary information. Specifically, the server formats the summary text into a visual format such as bullet points or color coding, and generates data for display.

[1198] Step 5:

[1199] The terminal displays the visually formatted summary information. The input is the visually formatted summary data sent from the server, and the output is the summary information displayed on the display. Specifically, the terminal receives data from the server and reflects it on the display.

[1200] Step 6:

[1201] The user can use voice input to modify the summary or visualized information. The input is the user's verbal instruction, and the output is the voice data. Specifically, the user gives instructions into the microphone of the device, and the voice is sent to the server.

[1202] Step 7:

[1203] The server converts the voice-input instructions back into text data and updates the summary information. The input is the user's voice data, and the output is the updated summary text data. Specifically, the server calls the SpeechRecognition library again to convert the speech to text and generate a new summary.

[1204] Step 8:

[1205] The terminal redisplays the updated summary information. The input is the updated summary data sent from the server, and the output is the new summary information displayed on the display. Specifically, the terminal receives the new data and reflects it on the display.

[1206] This makes it possible to efficiently grasp the work situation on site in real time and instantly communicate and correct important information.

[1207] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1208] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[1209] Basic System Configuration

[1210] The server has a means for acquiring voice data in real time. The server receives voice data captured during a remote conference and processes it in real time. The server converts the voice data into text data in real time. Voice recognition technology is used here.

[1211] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[1212] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[1213] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[1214] Adding an Emotion Engine

[1215] In the present invention, an emotion engine is further added, which detects the user's emotions in real time from what is said during the meeting and adjusts the summarized and visualized information based on the emotion information.

[1216] The emotion engine analyzes the user's emotions from the voice data and reflects the results in the summary of the text data. Based on this analysis, particularly important topics and points of strong reactions are highlighted, enabling more accurate summaries.

[1217] Furthermore, the emotion engine has the ability to display users' emotional states as visual elements such as graphs and color coding, allowing all meeting participants to grasp the emotional states of other participants in real time, making discussions proceed more smoothly.

[1218] Specific processing examples

[1219] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[1220] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1221] The server receives the audio data and converts it to text in real time:

[1222] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1223] The emotion engine analyzes the voice data and detects different emotional states:

[1224] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[1225] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[1226] 3. Detect "relief" sentiment for "budget already approved."

[1227] The server generates a summary based on this text and emotion information and visualizes it:

[1228] The new product will be released next month. (Positive)

[1229] Strengthening marketing campaigns. (Excitement)

[1230] Focus on online advertising. (Excited)

[1231] The budget has been approved. (Relax)

[1232] The device receives this summary and displays it to the user:

[1233] The new product will be released next month. (Positive)

[1234] Strengthening marketing campaigns. (Excitement)

[1235] Focus on online advertising. (Excited)

[1236] · Budget approved. (Relax)

[1237] User says "Please include a breakdown of your marketing budget":

[1238] "Please include a breakdown of your marketing budget."

[1239] The server receives this voice instruction and updates the summary and emotion information:

[1240] The new product will be released next month. (Positive)

[1241] Marketing campaign enhancements and budget breakdown. (Excitement)

[1242] Focus on online advertising. (Excited)

[1243] The budget has been approved. (Relax)

[1244] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[1245] The processing flow will be explained below.

[1246] Specific flow of program processing

[1247] Server-side processing

[1248] Step 1:

[1249] Setting authentication information

[1250] The server loads the credentials for the emotion engine and speech recognition service, which allow the server to securely access the API.

[1251] Step 2:

[1252] Real-time speech recognition settings

[1253] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[1254] Step 3:

[1255] Creating an Audio Stream

[1256] The server sends a request to the speech recognition service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them into text in real time.

[1257] Step 4:

[1258] Sending data for sentiment analysis

[1259] The server sends the text data and voice data to the emotion engine to analyze the user's emotions, and the emotion engine analyzes the voice data and returns emotion information.

[1260] Step 5:

[1261] Text summary

[1262] The server uses natural language processing techniques to summarize the recognized speech text and emotional information, extracting key points that reflect the emotional state from the long text data and creating a concise summary.

[1263] Step 6:

[1264] Summary visualization

[1265] The server then formats the summarized text and emotional information into a visually understandable format, using bullet points to indicate the text and colors and icons to indicate the emotional state.

[1266] Terminal side processing

[1267] Step 1:

[1268] Audio Stream Input

[1269] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[1270] Step 2:

[1271] Viewing summaries and visualizations

[1272] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[1273] User processing

[1274] Step 1:

[1275] Check and correct statements

[1276] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[1277] Step 2:

[1278] Entering correction instructions

[1279] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[1280] Step 3:

[1281] Processing and updating correction instructions

[1282] The server converts the voice input into text and uses that information to update summaries and visualizations.

[1283] Specific processing examples

[1284] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[1285] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1286] Acquiring voice data and converting it to text

[1287] Step 1:

[1288] The device captures the audio and sends it to the server.

[1289] Step 2:

[1290] The server receives the voice data and converts it into text in real time.

[1291] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1292] sentiment analysis

[1293] Step 3:

[1294] The server sends the text data and voice data to the emotion engine.

[1295] The sentiment engine analyzes sentiment as follows:

[1296] "The new product is scheduled to be released next month." -> Positive

[1297] "We need to step up our marketing campaigns and focus on online advertising." -> Excitement

[1298] "The budget has already been approved." -> Peace of mind

[1299] Summarization and visualization

[1300] Step 4:

[1301] The server generates a summary based on the text and emotion information.

[1302] The new product will be released next month. (Positive)

[1303] Strengthening marketing campaigns. (Excitement)

[1304] Focus on online advertising. (Excited)

[1305] The budget has been approved. (Relax)

[1306] Step 5:

[1307] The server visualizes the summary and sends it to the terminal.

[1308] Step 6:

[1309] The terminal will display a summary and visualization of the information:

[1310] The new product will be released next month. (Positive)

[1311] Strengthening marketing campaigns. (Excitement)

[1312] Focus on online advertising. (Excited)

[1313] · Budget approved. (Relax)

[1314] User-input corrections

[1315] Step 7:

[1316] The user utters, "Please include a breakdown of your marketing budget."

[1317] Step 8:

[1318] The terminal sends a voice instruction to the server.

[1319] Step 9:

[1320] The server converts the voice instructions into text and updates the summary and sentiment information:

[1321] The new product will be released next month. (Positive)

[1322] Marketing campaign enhancements and budget breakdown. (Excitement)

[1323] Focus on online advertising. (Excited)

[1324] The budget has been approved. (Relax)

[1325] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[1326] Example 2

[1327] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1328] Current remote conferencing systems lack a means to summarize the contents of a meeting in real time and grasp key points. Furthermore, there are no systems that provide visual information that accurately reflects the emotional state of users during a meeting. This can result in reduced meeting efficiency and the risk of important points being overlooked. Another problem is that it can be difficult for users to appropriately lead discussions based on their emotions.

[1329] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1330] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for analyzing a user's emotional state based on the summary, means for visualizing the summary and the user's emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input. This makes it easier to grasp important points of the meeting content and the user's emotional state in real time, enabling efficient and smooth discussions to proceed.

[1331] "Audio data" refers to data including speech and conversations of participants during a conference, captured as acoustic signals.

[1332] "Real-time" refers to a situation in which voice and text data is processed instantly with almost no delay.

[1333] "Text data" is character information converted from voice data, and is data that expresses the contents of the conference as text.

[1334] The "summarization means" is a system component that uses natural language processing technology to extract important points from long text data and summarize them in a concise form.

[1335] "Means for analyzing the user's emotional state" refers to technologies and algorithms for detecting the user's emotions from voice-recognized text or speech and analyzing that state.

[1336] A "visualization means" is a system component that has the function of displaying summarized text and emotional information in the form of graphs, color coding, etc., and providing information in a form that is easy for users to understand.

[1337] "Means for accepting user input" refers to an interface that receives voice or text instructions or requests from a user and allows the system to operate in response to those instructions or requests.

[1338] A "means for updating information" is a system component that has the functionality to recalculate and rearrange the summarized and visualized information based on input received from the user, and keep it up to date.

[1339] "Natural language processing technology" is a general term for technologies and algorithms that enable computers to understand and process human language.

[1340] "Voice input" refers to an input method in which a user gives instructions or requests to a system by voice.

[1341] "Meeting content" refers to the overall information, including the topics, comments, proposals, etc. discussed during the remote meeting.

[1342] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[1343] Basic System Configuration

[1344] The server is equipped with a means to acquire voice data in real time. Voice data during remote meetings is acquired through an API and processed in real time. The voice data is converted into text data in real time by the server. Specifically, voice recognition technology is used. For example, an API that provides voice recognition services as a general-purpose voice recognition API is used.

[1345] The converted text data is summarized by the server using natural language processing technology. Natural language processing technology extracts important points from long text data and creates a concise summary. The NLP model used for this summarization is a general-purpose generative AI model. An appropriate summary can be obtained by inputting a prompt sentence into the generative AI model.

[1346] The summarized text is then visualized. This can be done by turning the text into bullet points or formatting it for easier viewing. Furthermore, the visualization combines the output of the emotion engine to reflect the user's emotional state. For example, if a specific emotion is detected, it can be displayed in a different color.

[1347] The terminal has a means for displaying the summary and visualized information sent from the server. Specifically, this information is displayed on a web browser. Conference participants can refer to this information to proceed with the conference. If the user input is voice, the voice is converted into text data, and the summary and visualized information are updated. By performing this process in real time, it is possible to effectively summarize and visualize the contents of the conference.

[1348] Adding an Emotion Engine

[1349] One of the features of the present invention is the addition of an emotion engine. The emotion engine detects users' emotions in real time from audio data during a meeting and adjusts the summaries and visualizations based on this emotional information. The emotion engine incorporates algorithms and techniques for analyzing emotions from audio data. This highlights particularly important topics and points of strong reactions, enabling more accurate summaries. Furthermore, the results of the emotion engine can be displayed as visual elements. For example, the user's emotional state can be displayed in graphs or color-coded text format.

[1350] Specific examples

[1351] For example, suppose a remote meeting is taking place and a discussion is taking place about a new product launch strategy. If the following audio stream exists:

[1352] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1353] The server receives the audio data and converts it to text in real time:

[1354] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1355] The emotion engine analyzes the voice data and detects different emotional states:

[1356] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[1357] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[1358] 3. Detect "relief" sentiment for "budget already approved."

[1359] The server generates a summary based on this text and emotion information and visualizes it:

[1360] The new product will be released next month. (Positive)

[1361] Strengthening marketing campaigns. (Excitement)

[1362] Focus on online advertising. (Excited)

[1363] The budget has been approved. (Relax)

[1364] The terminal receives this summary and displays it to the user:

[1365] The new product will be released next month. (Positive)

[1366] Strengthening marketing campaigns. (Excitement)

[1367] Focus on online advertising. (Excited)

[1368] · Budget approved. (Relax)

[1369] User says "Please include a breakdown of your marketing budget":

[1370] "Please include a breakdown of your marketing budget."

[1371] The server receives this voice instruction and updates the summary and emotion information:

[1372] The new product will be released next month. (Positive)

[1373] Marketing campaign enhancements and budget breakdown. (Excitement)

[1374] Focus on online advertising. (Excited)

[1375] The budget has been approved. (Relax)

[1376] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that reflects the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[1377] Examples of key prompts include:

[1378] "Please summarize the meeting based on the following:

[1379] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1380] Based on these methods, a system is built that efficiently summarizes and visualizes remote meetings, as well as performs sentiment analysis.

[1381] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1382] Step 1:

[1383] The server acquires the audio data in real time.

[1384] Specific operation: The server obtains audio data from a remote conference platform (e.g., a conference application) in real time through an API. The input is the audio data stream during the conference, and the output is the buffered audio data. The server captures this audio data at regular intervals and stores it in a buffer for the next processing step. For example, the audio stream is sent as regular data packets in real time.

[1385] Step 2:

[1386] The server converts the audio data into text data

[1387] Specific operation: The server sends the acquired voice data to the voice recognition API and converts it into text data in real time. The input is the voice data stored in the buffer, and the output is the text data returned from the voice recognition API. The voice recognition API analyzes the voice signal and generates the corresponding text using a language model. For example, the speech "The new product launch is scheduled for next month" is converted into text "The new product launch is scheduled for next month."

[1388] Step 3:

[1389] The server summarizes the text data

[1390] Specific operation: The server summarizes text data using a generative AI model (e.g., an NLP model). The input is text data obtained from a speech recognition API, and the output is the summarized text. The generative AI model is prompted with "Please summarize the following text" to extract key points. For example, from the text "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved," a short summary is generated: "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved."

[1391] Step 4:

[1392] The server analyzes the user's emotional state

[1393] Specific operation: The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's emotional state based on the text data. The input is summarized text data, and the output is the emotion analysis result. The emotion analysis API detects emotions such as positive, negative, excited, and relieved for each sentence. For example, "The new product launch is scheduled for next month" may detect "positive" emotion, and "excited" emotion may be detected for "We need to strengthen our marketing campaign and focus on online advertising."

[1394] Step 5:

[1395] Server visualizes summary and emotional state

[1396] Specific operation: The server uses a visualization API (e.g., a chart drawing library) to visually display the summarized text and sentiment information. The input is the summarized text data and sentiment analysis results, and the output is the visualized information. The visualization API is used to format the text into bullet points and color-code the sentiment corresponding to each item. For example, the visualization is in the form of "New product launch next month (positive)" or "Marketing campaign intensification (excited)."

[1397] Step 6:

[1398] Device displays visualized information

[1399] Specific operation: The terminal receives visualized information sent from the server and displays it to the user. The input is the visualized text and emotion information sent from the server, and the output is the information displayed on the terminal screen. The terminal provides this information to the user in real time using a web browser or a conference display screen. For example, important points and emotional states during a conference are displayed in color.

[1400] Step 7:

[1401] The user gives additional instructions by voice input, and the server updates the information.

[1402] Specific operation: The user voice-inputs additional information or instructions during a meeting. The input is the user's voice data, and the output is an updated summary and emotional information. The server again sends this voice data to the speech recognition API and converts it into text data. The NLP model and emotional engine are then re-run to update the summary and emotional information. For example, receiving a voice instruction such as "Please also include a breakdown of the marketing budget," the summary and visualization information are reshaped.

[1403] Through this series of processing steps, the contents of remote meetings are summarized and visualized in real time, and information is provided that reflects the user's emotional state. As a result, meeting participants can more easily understand important points and the direction of the discussion, allowing the meeting to proceed smoothly.

[1404] (Application example 2)

[1405] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1406] Conventional remote conferencing systems have limited functionality for grasping and summarizing meeting content, resulting in reduced meeting efficiency. Furthermore, because they do not take into account the emotional state of meeting participants, it can be difficult to accurately understand the flow of discussion, which can lead to meetings not progressing smoothly. Remote meetings, particularly at production sites, require understanding important agenda items and making quick decisions, but there has been a lack of appropriate systems for this purpose.

[1407] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1408] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for recognizing a user's emotional state, means for adjusting the summary and visualized information based on the emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input, thereby enabling efficient understanding of the content of a remote conference and accurate summarization and visualization that take into account the emotional states of participants.

[1409] 1. "Audio Data" means audio information collected during remote meetings or communications.

[1410] 2. "Text data" means data that has been converted from audio data into text information.

[1411] 3. A "summary" is information that extracts important points from long text data and summarizes them concisely.

[1412] 4. "Visualization" means summarizing or presenting information in an easy-to-read format.

[1413] 5. "User Input" means additional information or instructions for correction provided by a conference participant.

[1414] 6. "Emotional state" refers to the emotional state detected based on the content of a user's speech.

[1415] 7. "Natural language processing technology" refers to technology that enables computers to understand, interpret, and manipulate human language.

[1416] 8. "Real-time acquisition means" refers to technologies or devices that allow for instantaneous collection of audio data.

[1417] 9. "Means for converting into text data" refers to technology or devices that automatically convert voice data into text information.

[1418] 10. "Summarizing tools" refers to techniques or devices that extract important information from long texts and summarize it concisely.

[1419] 11. "Visualization tools" refers to techniques and devices that display summaries and information in an easy-to-understand manner using graphs, color coding, etc.

[1420] 12. "Adjusting means" refers to techniques or devices that change summaries or visualizations based on emotional state.

[1421] 13. "Means for accepting user input for modification" refers to technology or devices that accept additional information or modification instructions provided by conference participants.

[1422] 14. "Means for updating" refers to techniques or devices that transform the summary or visualization into new information based on user input.

[1423] 15. "Remote meeting" means a meeting conducted online by multiple participants who are geographically separated.

[1424] This invention is a system for efficiently understanding and summarizing the contents of remote meetings. This system converts the contents of meetings into text in real time, summarizes important points, and visualizes emotional states, providing meeting participants with easy-to-understand information.

[1425] The system is configured as follows:

[1426] Acquiring and converting audio data

[1427] The server captures the voice data generated during the remote conference in real time. Specifically, it converts the voice data into text data using voice recognition technology (e.g., Google Speech-to-Text API). This allows the content of what is said during the conference to be instantly captured as text information.

[1428] Summarizing text data

[1429] The acquired text data is summarized using natural language processing techniques (e.g., TensorFlow NLP models), which extract key points from long text data and generate concise summaries.

[1430] User Emotion Recognition

[1431] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotional state from the speech data in real time. This detects the emotional state of the speech (positive, excited, relieved, etc.) and reflects the emotional state in the summary information.

[1432] Summarization and emotional state visualization

[1433] The summarized text data and emotional states are displayed in an easy-to-read format using visualization techniques (e.g., D3.js). The text is formatted as bullet points, and the emotional states are visually represented using color coding and icons. For example, positive statements are color-coded green, and excited statements are color-coded red.

[1434] User-input corrections and updates

[1435] The device provides an interface for users to provide corrections and additional information through voice or text input, such as "Please also include a breakdown of the marketing budget." Based on this, the server dynamically updates the summarized and visualized information.

[1436] Specific examples

[1437] For example, in a remote meeting at a production site, if someone says, "We'd like to discuss the introduction of a new production line. To do this, we need to consider how to improve production efficiency and the amount of investment required. What are your thoughts?", the system will process it as follows:

[1438] 1. Acquire audio data and convert it into text data.

[1439] 2. Summarize the text data and extract important topics such as "introduction of a new production line," "improvement of production efficiency," and "amount of investment required."

[1440] 3. The emotion engine detects the emotional state of each comment and assigns a label such as positive, excited, or relieved.

[1441] 4. These summaries and emotional states are visualized and displayed on the screen of a smartphone or head-mounted display.

[1442] Prompt Sentence Examples

[1443] "We would like to discuss the introduction of a new production line. To do so, we need to consider ways to improve production efficiency and the amount of investment required. We would appreciate your opinions."

[1444] This system allows for efficient understanding of the content of remote meetings and accurate summaries and visualizations that take into account the emotional states of participants, which is expected to lead to smoother meeting progress and faster decision-making.

[1445] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1446] Step 1: Acquire audio data

[1447] The server acquires the audio data of the remote conference in real time. As input, it collects the user's conversational voice from a microphone and obtains the data as an audio signal. Specifically, the server captures the audio stream of the conference and saves it as audio data. This data is passed to subsequent processing steps.

[1448] Step 2: Convert audio data to text

[1449] The server converts the acquired voice data into text data in real time. The voice data collected in step 1 is used as input. Specifically, it uses voice recognition technology (e.g., Google Speech-to-Text API) to convert the voice data into text data and output it as text data.

[1450] Step 3: Summarizing the text data

[1451] The server summarizes the converted text data. It uses the text data generated in step 2 as input. Specifically, it uses natural language processing techniques (e.g., TensorFlow NLP models) to extract key points from the text data and generate a concise summary. This summary is used in the next step.

[1452] Step 4: Recognizing your emotional state

[1453] The server recognizes the user's emotional state from the summarized text data. The summary generated in step 3 is used as input. Specifically, it uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from each part of the summary and assigns an emotion tag (positive, excited, relieved, etc.). This generates a summary that reflects the emotional state.

[1454] Step 5: Summarize and visualize emotional states

[1455] The server visualizes the summary and the emotional state. It uses the summary with the emotional state generated in step 4 as input. Specifically, it uses visualization technology (e.g., D3.js) to display the summary in bulleted form and visually represent the emotional state using color coding and icons. This visualization data is sent to the device.

[1456] Step 6: Accepting User Input

[1457] The device accepts input for corrections or additional information provided by the user. The user can provide instructions by voice or text as input. Specifically, in the case of voice input, the device uses voice recognition technology again, and in the case of text input, the device receives the input as text data.

[1458] Step 7: Update the summary and visualization

[1459] The server updates the summarized and visualized information based on the received user input. The input is the user instruction obtained in step 6. Specifically, the server performs a series of processes to re-summarize the updated information, recognize the emotional state, and visualize it. This generates updated summarized information that meets the user's requirements and retransmits it to the device.

[1460] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1461] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1462] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1463] [Fourth embodiment]

[1464] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1465] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1466] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1467] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1468] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1469] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1470] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1471] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1472] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1473] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1474] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1475] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1476] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1477] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[1478] The server has a means for acquiring voice data in real time. Specifically, it receives voice data captured during a remote conference and processes the data. The server converts the voice data into text data in real time using voice recognition technology.

[1479] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[1480] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[1481] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[1482] Furthermore, if the user wishes to modify the summary or visualized information during the process, the device is also provided with a means for accepting user input. This user input can be performed as voice input. When the user issues a command for modification by voice, the device transmits the voice to the server. The server converts the voice input into text data and updates the summary or visualized information based on that information.

[1483] Specific examples

[1484] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[1485] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1486] The audio data of the conference is sent to the server and converted into text data in real time. This text data is summarized as follows:

[1487] The new product will be released next month.

[1488] Enhanced marketing campaigns.

[1489] Focus on online advertising.

[1490] The budget has been approved.

[1491] This summary is visually formatted as a bulleted list and displayed on the terminal, as shown below:

[1492] The new product will be released next month.

[1493] Enhanced marketing campaigns.

[1494] · Focus on online advertising.

[1495] · Budget approved.

[1496] If the user responds to this summary by voice, saying "Please include a breakdown of the marketing budget," the server converts the voice into text and updates the summary.

[1497] The system provides a concise and accurate real-time summary and visualization of meeting content, allowing participants to immediately grasp key points. It also updates the summary based on user instructions, ensuring smooth meeting progress.

[1498] Thus, the present invention provides an embodiment for supporting efficient information sharing and decision-making in remote meetings.

[1499] The processing flow will be explained below.

[1500] Specific flow of program processing

[1501] Server-side processing

[1502] Step 1:

[1503] Setting authentication information

[1504] The server loads the credentials for the Google Cloud Speech-to-Text service, which allows the server to securely access the API.

[1505] Step 2:

[1506] Real-time speech recognition settings

[1507] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[1508] Step 3:

[1509] Creating an Audio Stream

[1510] The server sends a request to the Google Cloud Speech-to-Text service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them to text in real time.

[1511] Step 4:

[1512] Text summary

[1513] The server uses natural language processing techniques to summarize the recognized speech text, extracting key points from long text data and creating a concise summary.

[1514] Step 5:

[1515] Summary visualization

[1516] The server formats the summarized text into a visually easy-to-understand format, specifically by turning the text into bullet points and formatting it in a way that makes it easier to read.

[1517] Terminal side processing

[1518] Step 1:

[1519] Audio Stream Input

[1520] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[1521] Step 2:

[1522] Viewing summaries and visualizations

[1523] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[1524] User processing

[1525] Step 1:

[1526] Check and correct statements

[1527] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[1528] Step 2:

[1529] Entering correction instructions

[1530] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[1531] Step 3:

[1532] Processing and updating correction instructions

[1533] The server converts the voice input into text and uses that information to update summaries and visualizations.

[1534] Specific processing examples

[1535] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[1536] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1537] The server receives the audio data and converts it to text in real time:

[1538] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1539] The server summarizes this text:

[1540] The new product will be released next month.

[1541] Enhanced marketing campaigns.

[1542] Focus on online advertising.

[1543] The budget has been approved.

[1544] The device receives this summary and displays it to the user:

[1545] The new product will be released next month.

[1546] Enhanced marketing campaigns.

[1547] · Focus on online advertising.

[1548] · Budget approved.

[1549] User says "Please include a breakdown of your marketing budget":

[1550] "Please include a breakdown of your marketing budget."

[1551] The server receives this voice prompt and updates the summary:

[1552] The new product will be released next month.

[1553] Marketing campaign enhancements and budget breakdown.

[1554] Focus on online advertising.

[1555] The budget has been approved.

[1556] In this way, a system is realized in which the server, terminals, and users work together to efficiently summarize and visualize the contents of remote meetings and update the information as needed.

[1557] Example 1

[1558] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1559] Conventional remote conferencing systems make it difficult for participants to grasp the content of the meeting in real time. Furthermore, it takes a great deal of time and effort for participants to review important points after the meeting. Furthermore, it is not easy to revise summaries and visualized information in a timely manner during the meeting. To solve these problems, a system is needed that can summarize and visualize the content of the meeting in real time and quickly update it based on user instructions.

[1560] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1561] In this invention, the server includes means for acquiring voice information in real time, means for converting the voice information into document information in real time, means for summarizing the document information, means for representing the document information in graphics, means for accepting user input, and means for updating the summary or the graphic representation based on the user input. By converting voice information into document information in real time, generating a summary using natural language processing technology, and visualizing the summary, the server enables users to quickly grasp the content of the meeting and further enables information to be updated based on user instructions.

[1562] "Audio information" refers to data recorded in real time of participants' speech during a remote conference.

[1563] "Real-time" means that data is acquired, processed, or displayed with minimal real-world time delay.

[1564] "Document information" is text data generated by analyzing voice information.

[1565] A "summary" is a short text that extracts important points from the original document information and expresses them concisely and clearly.

[1566] "Graphical representation" means displaying summarized information in a visually easy-to-understand format, such as bullet points or graphs.

[1567] "User input" refers to an operation performed by a user participating in a remote conference to provide instructions or correction information to the system, and includes voice input.

[1568] "Natural language processing technology" is a technology that uses computers to understand, interpret, and generate human language.

[1569] The "update means" is a function that changes the currently displayed information to new content based on user input.

[1570] The present invention relates to a system for summarizing the contents of a remote conference in real time, grasping important points, and visualizing them. Hereinafter, an embodiment of the present invention will be described in detail.

[1571] The server includes a means for acquiring audio information captured during a conference from an online conference platform in real time, the audio information being acquired directly from a remote conference platform (e.g., a web conference system) using an API.

[1572] The acquired voice information is converted into text information in real time using voice recognition technology (for example, a voice conversion service). To convert the voice data into text information, the API of the voice conversion service is used. This text conversion process analyzes the voice information in real time and converts it into text data.

[1573] The converted document information is summarized by the server. This process uses a generative AI model (e.g., generative AI technology) to extract key points using natural language processing technology and generate a concise summary. The generative AI model receives prompts for generating a summary from text data and outputs the summary based on those prompts.

[1574] The generated summary is then visualized using HTML and CSS to format the summary text in bullet points and other formats for easy viewing. The visualized information is designed to help meeting participants grasp the key points at a glance.

[1575] The terminal has a means for displaying the summary and visualization information sent from the server, including the ability to display the summary information using a web browser or a dedicated application.

[1576] Furthermore, if a user wishes to revise the summary or visualized information during a meeting, they can issue instructions through voice input. The device then sends the speech to the server, which then uses speech recognition technology to convert the speech into text. The summary is then updated again using the generative AI model, and the latest information is visualized and displayed on the device.

[1577] Specific examples

[1578] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[1579] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1580] The audio information of the conference is sent to the server and converted into document information in real time as follows:

[1581] > Our new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved.

[1582] This document information is summarized by the generative AI model as follows:

[1583] The new product will be released next month.

[1584] > Enhanced marketing campaigns.

[1585] > Focus on online advertising.

[1586] > The budget has been approved.

[1587] This summary is visually formatted as bullet points and displayed on the terminal as follows:

[1588] The new product will be released next month.

[1589] > Enhanced marketing campaigns.

[1590] > · Focus on online advertising.

[1591] > · Budget approved.

[1592] If the user responds to this summary by voice, saying something like, "Please also include a breakdown of the marketing budget," the server converts the speech into text and updates the summary. This allows the meeting content to be summarized and visualized concisely and accurately in real time, allowing meeting participants to immediately grasp the important points. Furthermore, by updating the summary based on the user's instructions, the meeting proceeds smoothly.

[1593] Prompt sentence input example

[1594] As an example of a prompt sentence for summarizing the contents of a meeting, enter the following:

[1595] > Please summarize the meeting:

[1596] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1597] This system summarizes the contents of remote meetings in real time and provides them visually to participants, making meetings more efficient and enabling immediate understanding of important points. It is also a dynamic system that can quickly reflect corrections from users.

[1598] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1599] Step 1:

[1600] The server acquires audio information from the remote conference platform in real time. Specifically, it uses an API to capture the audio stream during the conference. This audio stream is sent as data in real time from the remote conference platform (e.g., a web conference system).

[1601] Input: Audio stream from remote conferencing platform

[1602] Output: Audio data stored on the server

[1603] Specific working example:

[1604] The server connects to the API of the online conference platform and obtains the audio data.

[1605] The acquired audio data is temporarily stored in storage.

[1606] Step 2:

[1607] The server converts the voice information into text data in real time using voice recognition technology. This process uses a voice conversion API to convert the voice data into text data.

[1608] Input: Saved audio data

[1609] Output: Converted text data

[1610] Specific working example:

[1611] The server sends the audio data to the speech conversion API.

[1612] The text data is returned from the conversion API and temporarily stored in the server's memory.

[1613] Step 3:

[1614] The server uses a generative AI model to summarize the document information. Specifically, it uses natural language processing technology to extract key points and generate a concise summary. In this process, a prompt sentence is input into the generative AI model to generate the summary.

[1615] Input: Converted text data

[1616] Output: Summarized text data

[1617] Specific working example:

[1618] The server sends text data along with a prompt to the generative AI model.

[1619] The summarized text data is received from the generative AI model and stored on the server.

[1620] Step 4:

[1621] The server uses HTML and CSS to format the generated summary visually, allowing the summarized information to be displayed in an easy-to-read format such as bullet points.

[1622] Input: Summarized text data

[1623] Output: Visualized HTML page

[1624] Specific working example:

[1625] The server embeds the summary text into an HTML template.

[1626] Apply CSS to format the summary into bullet points.

[1627] Generates the visualization HTML page and prepares it for display.

[1628] Step 5:

[1629] The terminal receives the visualized HTML page sent from the server and displays it to the user, displaying the summary information through a web browser or dedicated application.

[1630] Input: HTML page sent by the server

[1631] Output: Summary information displayed on the terminal

[1632] Specific working example:

[1633] The terminal periodically sends a request to the server to obtain the latest summary data.

[1634] The HTML page is displayed in a web browser.

[1635] Step 6:

[1636] When the user issues corrections via voice input, the device sends the speech to the server, which again uses speech recognition technology to convert the speech into text information and updates the summary using a generative AI model.

[1637] Input: User voice input

[1638] Output: Updated summary text, visualized information

[1639] Specific working example:

[1640] The user utters, "Please include a breakdown of your marketing budget."

[1641] The device sends voice data to the server, which converts the data into text.

[1642] The server inputs text data into the generative AI model and updates the summary.

[1643] The updated summary text is embedded in the HTML page again and sent to the terminal.

[1644] (Application example 1)

[1645] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1646] There is a demand for more efficient work and accurate information sharing in the field. In particular, in workplaces such as factories, it is important to grasp the situation in real time and summarize and share important information. However, currently, workers manually collect and report information, which creates issues with the accuracy and timeliness of the information. Furthermore, it is time-consuming for workers to correct and update information, which can easily lead to work delays and misunderstandings. To solve these problems, a system is needed that can grasp the situation in real time and summarize and share important information.

[1647] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1648] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for accepting user input for modifying the summarized or visualized information, means for updating the summarized or visualized information based on the user input, means for displaying the summarized or visualized information as target information on a display of a machine, and means for the machine to collect the situation in the workplace as voice data and extract and summarize important information. This makes it possible to grasp the work situation in the field in real time and efficiently communicate and modify important information.

[1649] "Audio data" refers to digital information recorded from human speech, environmental sounds, etc.

[1650] "Real-time" means that information processing, data acquisition, and transmission are carried out immediately.

[1651] "Text data" is data expressed as a string of characters.

[1652] To "summarize" means to extract the important parts from the original information and summarize them concisely.

[1653] "Visualization" is the visual display of information to make it easier to see and understand.

[1654] "User input" refers to information or instructions given to the system by a user.

[1655] A "machine" is a device that performs a specific task automatically.

[1656] A "display" is a screen for displaying information.

[1657] A "workplace" is a place where a specific task is performed.

[1658] "Natural language processing technology" is a technology that allows computers to understand, analyze, and generate human language.

[1659] The present invention relates to a system for efficiently grasping the work status at a site and summarizing and sharing important information. The following describes in detail an embodiment of the present invention.

[1660] The server is equipped with a means for acquiring voice data in real time. Specifically, it uses microphones placed in the workplace to collect voices emitted by workers and machinery. The collected voice data is sent to the server, where it is processed.

[1661] The server then converts the captured audio data into text data in real time using the SpeechRecognition library. Once the text data is generated, the server uses natural language processing technology (such as OpenAI's GPT-3.5) to extract key points and generate a summary. The summarized information is then visualized for easy viewing.

[1662] Furthermore, the summarized information is displayed on the machine's display. The display is placed in a location where workers can easily see it and is updated in real time. For example, the display may show a specific summary such as, "Part A is low in stock. Replenishment is required."

[1663] If the user wishes to modify the summary or visualization, they can give instructions using voice input, which is then sent back to the server and converted into text data, and the summary or visualization is updated based on the converted text data.

[1664] This system makes it possible to grasp the work status on site in real time, allowing workers to immediately check important information and make corrections as necessary, thereby achieving efficient and accurate information sharing.

[1665] Specific examples

[1666] When a robot worker in a factory says, "Part A is running low on stock. We need to replenish it before the next shift," the voice data is sent to the server. The server converts it into text data and generates a summary that reads, "Part A is running low on stock. We need to replenish it." This summary is then displayed on a screen so that the worker can check it.

[1667] Prompt Sentence Examples

[1668] Please summarize the following:

[1669] "Part A is low in stock. It needs to be replenished before the next shift."

[1670] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1671] Step 1:

[1672] The server acquires voice data in real time using microphones placed in the workplace. The input is the voice from workers and machinery, and the output is to send this voice to the server as digital voice data. Specifically, the microphones capture the voice in the workplace and send it to the server via a data network.

[1673] Step 2:

[1674] The server converts the voice data acquired in real time into text data using the SpeechRecognition library. The input is digital voice data, and the output is text data of character strings extracted from the voice. Specifically, the server calls the SpeechRecognition library, processes the voice data, and generates the corresponding text.

[1675] Step 3:

[1676] The server inputs the generated text data into OpenAI's generative AI model (such as GPT-3.5) to generate a summary. The input is the text data obtained in the previous step, and the output is the summarized text data. Specific operations include the process in which the server sends the text data to the generative AI model and receives the summary.

[1677] Step 4:

[1678] The server performs visualization to display the summarized text data in a visually easy-to-understand manner. The input is the generated summary text, and the output is visually formatted summary information. Specifically, the server formats the summary text into a visual format such as bullet points or color coding, and generates data for display.

[1679] Step 5:

[1680] The terminal displays the visually formatted summary information. The input is the visually formatted summary data sent from the server, and the output is the summary information displayed on the display. Specifically, the terminal receives data from the server and reflects it on the display.

[1681] Step 6:

[1682] The user can use voice input to modify the summary or visualized information. The input is the user's verbal instruction, and the output is the voice data. Specifically, the user gives instructions into the microphone of the device, and the voice is sent to the server.

[1683] Step 7:

[1684] The server converts the voice-input instructions back into text data and updates the summary information. The input is the user's voice data, and the output is the updated summary text data. Specifically, the server calls the SpeechRecognition library again to convert the speech to text and generate a new summary.

[1685] Step 8:

[1686] The terminal redisplays the updated summary information. The input is the updated summary data sent from the server, and the output is the new summary information displayed on the display. Specifically, the terminal receives the new data and reflects it on the display.

[1687] This makes it possible to efficiently grasp the work situation on site in real time and instantly communicate and correct important information.

[1688] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1689] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[1690] Basic System Configuration

[1691] The server has a means for acquiring voice data in real time. The server receives voice data captured during a remote conference and processes it in real time. The server converts the voice data into text data in real time. Voice recognition technology is used here.

[1692] The converted text data is then summarized by the server using natural language processing technology, which extracts important points from long text data and creates a concise summary.

[1693] The summarized text is then visualized, using bulleted and formatted text to allow users to grasp the key points of the meeting at a glance.

[1694] The terminal has a means for displaying the summary and visualized information sent from the server, and the conference participants can proceed with the conference while referring to this information.

[1695] Adding an Emotion Engine

[1696] In the present invention, an emotion engine is further added, which detects the user's emotions in real time from what is said during the meeting and adjusts the summarized and visualized information based on the emotion information.

[1697] The emotion engine analyzes the user's emotions from the voice data and reflects the results in the summary of the text data. Based on this analysis, particularly important topics and points of strong reactions are highlighted, enabling more accurate summaries.

[1698] Furthermore, the emotion engine has the ability to display users' emotional states as visual elements such as graphs and color coding, allowing all meeting participants to grasp the emotional states of other participants in real time, making discussions proceed more smoothly.

[1699] Specific processing examples

[1700] For example, suppose a discussion is taking place during a remote conference about a new product launch strategy. Let's consider the case where the audio stream during the conference contains the following content:

[1701] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1702] The server receives the audio data and converts it to text in real time:

[1703] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1704] The emotion engine analyzes the voice data and detects different emotional states:

[1705] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[1706] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[1707] 3. Detect "relief" sentiment for "budget already approved."

[1708] The server generates a summary based on this text and emotion information and visualizes it:

[1709] The new product will be released next month. (Positive)

[1710] Strengthening marketing campaigns. (Excitement)

[1711] Focus on online advertising. (Excited)

[1712] The budget has been approved. (Relax)

[1713] The device receives this summary and displays it to the user:

[1714] The new product will be released next month. (Positive)

[1715] Strengthening marketing campaigns. (Excitement)

[1716] Focus on online advertising. (Excited)

[1717] · Budget approved. (Relax)

[1718] User says "Please include a breakdown of your marketing budget":

[1719] "Please include a breakdown of your marketing budget."

[1720] The server receives this voice instruction and updates the summary and emotion information:

[1721] The new product will be released next month. (Positive)

[1722] Marketing campaign enhancements and budget breakdown. (Excitement)

[1723] Focus on online advertising. (Excited)

[1724] The budget has been approved. (Relax)

[1725] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[1726] The processing flow will be explained below.

[1727] Specific flow of program processing

[1728] Server-side processing

[1729] Step 1:

[1730] Setting authentication information

[1731] The server loads the credentials for the emotion engine and speech recognition service, which allow the server to securely access the API.

[1732] Step 2:

[1733] Real-time speech recognition settings

[1734] The server configures the voice recognition settings, including the audio encoding format, sample rate, and language to be used.

[1735] Step 3:

[1736] Creating an Audio Stream

[1737] The server sends a request to the speech recognition service to process audio from the audio stream in real time, processing chunks of audio sequentially and converting them into text in real time.

[1738] Step 4:

[1739] Sending data for sentiment analysis

[1740] The server sends the text data and voice data to the emotion engine to analyze the user's emotions, and the emotion engine analyzes the voice data and returns emotion information.

[1741] Step 5:

[1742] Text summary

[1743] The server uses natural language processing techniques to summarize the recognized speech text and emotional information, extracting key points that reflect the emotional state from the long text data and creating a concise summary.

[1744] Step 6:

[1745] Summary visualization

[1746] The server then formats the summarized text and emotional information into a visually understandable format, using bullet points to indicate the text and colors and icons to indicate the emotional state.

[1747] Terminal side processing

[1748] Step 1:

[1749] Audio Stream Input

[1750] The device captures the audio during the meeting and sends it as an audio stream to the server, in real time.

[1751] Step 2:

[1752] Viewing summaries and visualizations

[1753] The terminal receives the summary and visualization information sent from the server and displays it to the user, allowing meeting participants to grasp important information in real time.

[1754] User processing

[1755] Step 1:

[1756] Check and correct statements

[1757] The user checks the summary and visualized information displayed on the terminal, and if necessary, gives correction instructions by voice input.

[1758] Step 2:

[1759] Entering correction instructions

[1760] When the user issues a correction instruction by voice, the voice is transmitted again to the server via the terminal.

[1761] Step 3:

[1762] Processing and updating correction instructions

[1763] The server converts the voice input into text and uses that information to update summaries and visualizations.

[1764] Specific processing examples

[1765] For example, suppose a remote meeting is taking place and a discussion is taking place about "New Product Launch Strategy." The audio stream during the meeting includes the following:

[1766] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1767] Acquiring voice data and converting it to text

[1768] Step 1:

[1769] The device captures the audio and sends it to the server.

[1770] Step 2:

[1771] The server receives the voice data and converts it into text in real time.

[1772] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1773] sentiment analysis

[1774] Step 3:

[1775] The server sends the text data and voice data to the emotion engine.

[1776] The sentiment engine analyzes sentiment as follows:

[1777] "The new product is scheduled to be released next month." -> Positive

[1778] "We need to step up our marketing campaigns and focus on online advertising." -> Excitement

[1779] "The budget has already been approved." -> Peace of mind

[1780] Summarization and visualization

[1781] Step 4:

[1782] The server generates a summary based on the text and emotion information.

[1783] The new product will be released next month. (Positive)

[1784] Strengthening marketing campaigns. (Excitement)

[1785] Focus on online advertising. (Excited)

[1786] The budget has been approved. (Relax)

[1787] Step 5:

[1788] The server visualizes the summary and sends it to the terminal.

[1789] Step 6:

[1790] The terminal will display a summary and visualization of the information:

[1791] The new product will be released next month. (Positive)

[1792] Strengthening marketing campaigns. (Excitement)

[1793] Focus on online advertising. (Excited)

[1794] · Budget approved. (Relax)

[1795] User-input corrections

[1796] Step 7:

[1797] The user utters, "Please include a breakdown of your marketing budget."

[1798] Step 8:

[1799] The terminal sends a voice instruction to the server.

[1800] Step 9:

[1801] The server converts the voice instructions into text and updates the summary and sentiment information:

[1802] The new product will be released next month. (Positive)

[1803] Marketing campaign enhancements and budget breakdown. (Excitement)

[1804] Focus on online advertising. (Excited)

[1805] The budget has been approved. (Relax)

[1806] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that takes into account the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[1807] Example 2

[1808] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1809] Current remote conferencing systems lack a means to summarize the contents of a meeting in real time and grasp key points. Furthermore, there are no systems that provide visual information that accurately reflects the emotional state of users during a meeting. This can result in reduced meeting efficiency and the risk of important points being overlooked. Another problem is that it can be difficult for users to appropriately lead discussions based on their emotions.

[1810] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1811] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for analyzing a user's emotional state based on the summary, means for visualizing the summary and the user's emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input. This makes it easier to grasp important points of the meeting content and the user's emotional state in real time, enabling efficient and smooth discussions to proceed.

[1812] "Audio data" refers to data including speech and conversations of participants during a conference, captured as acoustic signals.

[1813] "Real-time" refers to a situation in which voice and text data is processed instantly with almost no delay.

[1814] "Text data" is character information converted from voice data, and is data that expresses the contents of the conference as text.

[1815] The "summarization means" is a system component that uses natural language processing technology to extract important points from long text data and summarize them in a concise form.

[1816] "Means for analyzing the user's emotional state" refers to technologies and algorithms for detecting the user's emotions from voice-recognized text or speech and analyzing that state.

[1817] A "visualization means" is a system component that has the function of displaying summarized text and emotional information in the form of graphs, color coding, etc., and providing information in a form that is easy for users to understand.

[1818] "Means for accepting user input" refers to an interface that receives voice or text instructions or requests from a user and allows the system to operate in response to those instructions or requests.

[1819] A "means for updating information" is a system component that has the functionality to recalculate and rearrange the summarized and visualized information based on input received from the user, and keep it up to date.

[1820] "Natural language processing technology" is a general term for technologies and algorithms that enable computers to understand and process human language.

[1821] "Voice input" refers to an input method in which a user gives instructions or requests to a system by voice.

[1822] "Meeting content" refers to the overall information, including the topics, comments, proposals, etc. discussed during the remote meeting.

[1823] This invention relates to a system that summarizes the contents of a remote conference in real time, grasps the key points, and visualizes them. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides summaries and visualizations that reflect the user's emotional state during the conference. The following describes embodiments of the invention in detail.

[1824] Basic System Configuration

[1825] The server is equipped with a means to acquire voice data in real time. Voice data during remote meetings is acquired through an API and processed in real time. The voice data is converted into text data in real time by the server. Specifically, voice recognition technology is used. For example, an API that provides voice recognition services as a general-purpose voice recognition API is used.

[1826] The converted text data is summarized by the server using natural language processing technology. Natural language processing technology extracts important points from long text data and creates a concise summary. The NLP model used for this summarization is a general-purpose generative AI model. An appropriate summary can be obtained by inputting a prompt sentence into the generative AI model.

[1827] The summarized text is then visualized. This can be done by turning the text into bullet points or formatting it for easier viewing. Furthermore, the visualization combines the output of the emotion engine to reflect the user's emotional state. For example, if a specific emotion is detected, it can be displayed in a different color.

[1828] The terminal has a means for displaying the summary and visualized information sent from the server. Specifically, this information is displayed on a web browser. Conference participants can refer to this information to proceed with the conference. If the user input is voice, the voice is converted into text data, and the summary and visualized information are updated. By performing this process in real time, it is possible to effectively summarize and visualize the contents of the conference.

[1829] Adding an Emotion Engine

[1830] One of the features of the present invention is the addition of an emotion engine. The emotion engine detects users' emotions in real time from audio data during a meeting and adjusts the summaries and visualizations based on this emotional information. The emotion engine incorporates algorithms and techniques for analyzing emotions from audio data. This highlights particularly important topics and points of strong reactions, enabling more accurate summaries. Furthermore, the results of the emotion engine can be displayed as visual elements. For example, the user's emotional state can be displayed in graphs or color-coded text format.

[1831] Specific examples

[1832] For example, suppose a remote meeting is taking place and a discussion is taking place about a new product launch strategy. If the following audio stream exists:

[1833] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1834] The server receives the audio data and converts it to text in real time:

[1835] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1836] The emotion engine analyzes the voice data and detects different emotional states:

[1837] 1. Detect "positive" sentiment when the speaker says, "The new product is scheduled to be released next month."

[1838] 2. Detect "excited" sentiment for "We need to step up our marketing campaigns and focus on online advertising."

[1839] 3. Detect "relief" sentiment for "budget already approved."

[1840] The server generates a summary based on this text and emotion information and visualizes it:

[1841] The new product will be released next month. (Positive)

[1842] Strengthening marketing campaigns. (Excitement)

[1843] Focus on online advertising. (Excited)

[1844] The budget has been approved. (Relax)

[1845] The terminal receives this summary and displays it to the user:

[1846] The new product will be released next month. (Positive)

[1847] Strengthening marketing campaigns. (Excitement)

[1848] Focus on online advertising. (Excited)

[1849] · Budget approved. (Relax)

[1850] User says "Please include a breakdown of your marketing budget":

[1851] "Please include a breakdown of your marketing budget."

[1852] The server receives this voice instruction and updates the summary and emotion information:

[1853] The new product will be released next month. (Positive)

[1854] Marketing campaign enhancements and budget breakdown. (Excitement)

[1855] Focus on online advertising. (Excited)

[1856] The budget has been approved. (Relax)

[1857] In this way, the server, device, user, and emotion engine work together to more efficiently summarize and visualize the contents of remote meetings, and provide information that reflects the user's emotional state. This makes it easier for meeting participants to understand important points and the direction of the discussion, and the meeting proceeds more smoothly.

[1858] Examples of key prompts include:

[1859] "Please summarize the meeting based on the following:

[1860] "The new product launch is scheduled for next month. We need to step up our marketing campaign and focus on online advertising. The budget has already been approved."

[1861] Based on these methods, a system is built that efficiently summarizes and visualizes remote meetings, as well as performs sentiment analysis.

[1862] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1863] Step 1:

[1864] The server acquires the audio data in real time.

[1865] Specific operation: The server obtains audio data from a remote conference platform (e.g., a conference application) in real time through an API. The input is the audio data stream during the conference, and the output is the buffered audio data. The server captures this audio data at regular intervals and stores it in a buffer for the next processing step. For example, the audio stream is sent as regular data packets in real time.

[1866] Step 2:

[1867] The server converts the audio data into text data

[1868] Specific operation: The server sends the acquired voice data to the voice recognition API and converts it into text data in real time. The input is the voice data stored in the buffer, and the output is the text data returned from the voice recognition API. The voice recognition API analyzes the voice signal and generates the corresponding text using a language model. For example, the speech "The new product launch is scheduled for next month" is converted into text "The new product launch is scheduled for next month."

[1869] Step 3:

[1870] The server summarizes the text data

[1871] Specific operation: The server summarizes text data using a generative AI model (e.g., an NLP model). The input is text data obtained from a speech recognition API, and the output is the summarized text. The generative AI model is prompted with "Please summarize the following text" to extract key points. For example, from the text "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved," a short summary is generated: "The new product launch is scheduled for next month. We need to intensify our marketing campaign and focus on online advertising. The budget has already been approved."

[1872] Step 4:

[1873] The server analyzes the user's emotional state

[1874] Specific operation: The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's emotional state based on the text data. The input is summarized text data, and the output is the emotion analysis result. The emotion analysis API detects emotions such as positive, negative, excited, and relieved for each sentence. For example, "The new product launch is scheduled for next month" may detect "positive" emotion, and "excited" emotion may be detected for "We need to strengthen our marketing campaign and focus on online advertising."

[1875] Step 5:

[1876] Server visualizes summary and emotional state

[1877] Specific operation: The server uses a visualization API (e.g., a chart drawing library) to visually display the summarized text and sentiment information. The input is the summarized text data and sentiment analysis results, and the output is the visualized information. The visualization API is used to format the text into bullet points and color-code the sentiment corresponding to each item. For example, the visualization is in the form of "New product launch next month (positive)" or "Marketing campaign intensification (excited)."

[1878] Step 6:

[1879] Device displays visualized information

[1880] Specific operation: The terminal receives visualized information sent from the server and displays it to the user. The input is the visualized text and emotion information sent from the server, and the output is the information displayed on the terminal screen. The terminal provides this information to the user in real time using a web browser or a conference display screen. For example, important points and emotional states during a conference are displayed in color.

[1881] Step 7:

[1882] The user gives additional instructions by voice input, and the server updates the information.

[1883] Specific operation: The user voice-inputs additional information or instructions during a meeting. The input is the user's voice data, and the output is an updated summary and emotional information. The server again sends this voice data to the speech recognition API and converts it into text data. The NLP model and emotional engine are then re-run to update the summary and emotional information. For example, receiving a voice instruction such as "Please also include a breakdown of the marketing budget," the summary and visualization information are reshaped.

[1884] Through this series of processing steps, the contents of remote meetings are summarized and visualized in real time, and information is provided that reflects the user's emotional state. As a result, meeting participants can more easily understand important points and the direction of the discussion, allowing the meeting to proceed smoothly.

[1885] (Application example 2)

[1886] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1887] Conventional remote conferencing systems have limited functionality for grasping and summarizing meeting content, resulting in reduced meeting efficiency. Furthermore, because they do not take into account the emotional state of meeting participants, it can be difficult to accurately understand the flow of discussion, which can lead to meetings not progressing smoothly. Remote meetings, particularly at production sites, require understanding important agenda items and making quick decisions, but there has been a lack of appropriate systems for this purpose.

[1888] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1889] In this invention, the server includes means for acquiring voice data in real time, means for converting the voice data into text data in real time, means for summarizing the text data, means for visualizing the summary, means for recognizing a user's emotional state, means for adjusting the summary and visualized information based on the emotional state, means for accepting user input for modifying the summary or visualized information, and means for updating the summary or visualized information based on the user input, thereby enabling efficient understanding of the content of a remote conference and accurate summarization and visualization that take into account the emotional states of participants.

[1890] 1. "Audio Data" means audio information collected during remote meetings or communications.

[1891] 2. "Text data" means data that has been converted from audio data into text information.

[1892] 3. A "summary" is information that extracts important points from long text data and summarizes them concisely.

[1893] 4. "Visualization" means summarizing or presenting information in an easy-to-read format.

[1894] 5. "User Input" means additional information or instructions for correction provided by a conference participant.

[1895] 6. "Emotional state" refers to the emotional state detected based on the content of a user's speech.

[1896] 7. "Natural language processing technology" refers to technology that enables computers to understand, interpret, and manipulate human language.

[1897] 8. "Real-time acquisition means" refers to technologies or devices that allow for instantaneous collection of audio data.

[1898] 9. "Means for converting into text data" refers to technology or devices that automatically convert voice data into text information.

[1899] 10. "Summarizing tools" refers to techniques or devices that extract important information from long texts and summarize it concisely.

[1900] 11. "Visualization tools" refers to techniques and devices that display summaries and information in an easy-to-understand manner using graphs, color coding, etc.

[1901] 12. "Adjusting means" refers to techniques or devices that change summaries or visualizations based on emotional state.

[1902] 13. "Means for accepting user input for modification" refers to technology or devices that accept additional information or modification instructions provided by conference participants.

[1903] 14. "Means for updating" refers to techniques or devices that transform the summary or visualization into new information based on user input.

[1904] 15. "Remote meeting" means a meeting conducted online by multiple participants who are geographically separated.

[1905] This invention is a system for efficiently understanding and summarizing the contents of remote meetings. This system converts the contents of meetings into text in real time, summarizes important points, and visualizes emotional states, providing meeting participants with easy-to-understand information.

[1906] The system is configured as follows:

[1907] Acquiring and converting audio data

[1908] The server captures the voice data generated during the remote conference in real time. Specifically, it converts the voice data into text data using voice recognition technology (e.g., Google Speech-to-Text API). This allows the content of what is said during the conference to be instantly captured as text information.

[1909] Summarizing text data

[1910] The acquired text data is summarized using natural language processing techniques (e.g., TensorFlow NLP models), which extract key points from long text data and generate concise summaries.

[1911] User Emotion Recognition

[1912] The server uses an emotion engine (e.g., IBM Watson Tone Analyzer) to recognize the user's emotional state from the speech data in real time. This detects the emotional state of the speech (positive, excited, relieved, etc.) and reflects the emotional state in the summary information.

[1913] Summarization and emotional state visualization

[1914] The summarized text data and emotional states are displayed in an easy-to-read format using visualization techniques (e.g., D3.js). The text is formatted as bullet points, and the emotional states are visually represented using color coding and icons. For example, positive statements are color-coded green, and excited statements are color-coded red.

[1915] User-input corrections and updates

[1916] The device provides an interface for users to provide corrections and additional information through voice or text input, such as "Please also include a breakdown of the marketing budget." Based on this, the server dynamically updates the summarized and visualized information.

[1917] Specific examples

[1918] For example, in a remote meeting at a production site, if someone says, "We'd like to discuss the introduction of a new production line. To do this, we need to consider how to improve production efficiency and the amount of investment required. What are your thoughts?", the system will process it as follows:

[1919] 1. Acquire audio data and convert it into text data.

[1920] 2. Summarize the text data and extract important topics such as "introduction of a new production line," "improvement of production efficiency," and "amount of investment required."

[1921] 3. The emotion engine detects the emotional state of each comment and assigns a label such as positive, excited, or relieved.

[1922] 4. These summaries and emotional states are visualized and displayed on the screen of a smartphone or head-mounted display.

[1923] Prompt Sentence Examples

[1924] "We would like to discuss the introduction of a new production line. To do so, we need to consider ways to improve production efficiency and the amount of investment required. We would appreciate your opinions."

[1925] This system allows for efficient understanding of the content of remote meetings and accurate summaries and visualizations that take into account the emotional states of participants, which is expected to lead to smoother meeting progress and faster decision-making.

[1926] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1927] Step 1: Acquire audio data

[1928] The server acquires the audio data of the remote conference in real time. As input, it collects the user's conversational voice from a microphone and obtains the data as an audio signal. Specifically, the server captures the audio stream of the conference and saves it as audio data. This data is passed to subsequent processing steps.

[1929] Step 2: Convert audio data to text

[1930] The server converts the acquired voice data into text data in real time. The voice data collected in step 1 is used as input. Specifically, it uses voice recognition technology (e.g., Google Speech-to-Text API) to convert the voice data into text data and output it as text data.

[1931] Step 3: Summarizing the text data

[1932] The server summarizes the converted text data. It uses the text data generated in step 2 as input. Specifically, it uses natural language processing techniques (e.g., TensorFlow NLP models) to extract key points from the text data and generate a concise summary. This summary is used in the next step.

[1933] Step 4: Recognizing your emotional state

[1934] The server recognizes the user's emotional state from the summarized text data. The summary generated in step 3 is used as input. Specifically, it uses an emotion engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional state from each part of the summary and assigns an emotion tag (positive, excited, relieved, etc.). This generates a summary that reflects the emotional state.

[1935] Step 5: Summarize and visualize emotional states

[1936] The server visualizes the summary and the emotional state. It uses the summary with the emotional state generated in step 4 as input. Specifically, it uses visualization technology (e.g., D3.js) to display the summary in bulleted form and visually represent the emotional state using color coding and icons. This visualization data is sent to the device.

[1937] Step 6: Accepting User Input

[1938] The device accepts input for corrections or additional information provided by the user. The user can provide instructions by voice or text as input. Specifically, in the case of voice input, the device uses voice recognition technology again, and in the case of text input, the device receives the input as text data.

[1939] Step 7: Update the summary and visualization

[1940] The server updates the summarized and visualized information based on the received user input. The input is the user instruction obtained in step 6. Specifically, the server performs a series of processes to re-summarize the updated information, recognize the emotional state, and visualize it. This generates updated summarized information that meets the user's requirements and retransmits it to the device.

[1941] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1942] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1943] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1944] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1945] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1946] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1947] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1948] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1949] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1950] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1951] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1952] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1953] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1954] 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.

[1955] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1956] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1957] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1958] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1959] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1960] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1961] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1962] The following is further disclosed regarding the above embodiment.

[1963] (Claim 1)

[1964] a means for acquiring audio data in real time;

[1965] means for converting the voice data into text data in real time;

[1966] means for summarizing the text data;

[1967] means for visualizing said summary;

[1968] means for accepting user input to modify the summarized or visualized information;

[1969] The system includes means for updating the summarized or visualized information based on the user input.

[1970] (Claim 2)

[1971] 2. The system according to claim 1, wherein the summarizing means uses natural language processing techniques to summarize the contents of the meeting.

[1972] (Claim 3)

[1973] 2. The system of claim 1, wherein the user input is a voice input, the voice input is converted into text data, and the summarized or visualized information is updated based on the text data.

[1974] "Example 1"

[1975] (Claim 1)

[1976] a means for acquiring audio information in real time;

[1977] means for converting the voice information into text information in real time;

[1978] means for summarizing the document information;

[1979] means for graphically representing said summary;

[1980] means for accepting user input to modify said abstract or graphically represented information;

[1981] means for updating said summary or graphically represented information based on said user input.

[1982] (Claim 2)

[1983] 10. The system of claim 1, wherein the summarizing means uses natural language processing techniques to summarize the content of the meeting.

[1984] (Claim 3)

[1985] 2. The system of claim 1, wherein the user input is a voice input, the voice input is converted into text information, and the summary or graphical representation information is updated based on the text information.

[1986] "Application Example 1"

[1987] (Claim 1)

[1988] a means for acquiring audio data in real time;

[1989] means for converting the voice data into text data in real time;

[1990] means for summarizing the text data;

[1991] means for visualizing said summary;

[1992] means for accepting user input to modify the summarized or visualized information;

[1993] means for updating the summarized or visualized information based on the user input;

[1994] means for displaying the summarized or visualized information as target information on a display of a machine device;

[1995] The system includes a means for the machine to collect voice data on the situation in the workplace and extract and summarize important information.

[1996] (Claim 2)

[1997] 2. The system of claim 1, wherein the summarizing means uses natural language processing techniques to summarize the contents of the meeting and the situation in the workplace.

[1998] (Claim 3)

[1999] 2. The system of claim 1, wherein the user input is a voice input, the voice input is converted into text data, and the summarized or visualized information is updated based on the text data.

[2000] "Example 2: Combining Emotion Engines"

[2001] (Claim 1)

[2002] a means for acquiring audio data in real time;

[2003] means for converting the voice data into text data in real time;

[2004] means for summarizing the text data;

[2005] means for analyzing a user's emotional state based on said summary;

[2006] means for visualizing said summary and the user's emotional state;

[2007] means for accepting user input to modify the summarized or visualized information;

[2008] The system includes means for updating the summarized or visualized information based on the user input.

[2009] (Claim 2)

[2010] 2....

Claims

1. a means for acquiring audio data in real time; means for converting the voice data into text data in real time; means for summarizing the text data; means for visualizing said summary; means for accepting user input to modify the summarized or visualized information; and means for updating the summarized or visualized information based on the user input.

2. 2. The system of claim 1, wherein the summarizing means uses natural language processing techniques to summarize the contents of the meeting.

3. The system of claim 1 , wherein the user input is a voice input, the voice input is converted into text data, and the summarized or visualized information is updated based on the text data.

Images