Information Processing System, Information Processing Method, and Program

Abstract

Provided are an information processing system, an information processing method, and a program that improve convenience in a minutes creation tool. 【Solution means】The information processing system 1 includes one or more server devices 2 and one or more client devices 3-1, client devices 3-2, ···, and these are connected through a network NW. The server device and the client device transmit or receive various information to and from each other via the network. The information transmission / reception unit of the server device receives, acquires, and accepts information via the network and the communication unit. The recording unit of the server device stores voice information in the storage unit. The generation control unit of the server device acquires conference text information indicating text generated based on the voice information and the term information.

Description

Technical Field

[0001] The present invention relates to an information processing system, an information processing method, and a program.

Background Art

[0002] Conventionally, a system that performs speech recognition from the speech of a person uttered during a meeting has been known. Also, a system that supports the creation of meeting minutes by speech recognition has been known.

[0003] Patent Document 1 discloses a prior art that recognizes speech and supports the creation of meeting minutes.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the present invention, a technique capable of further improving convenience in a meeting minutes creation tool is provided.

Means for Solving the Problems

[0006] According to one aspect of the present invention, it includes one or more processors capable of executing the following steps. In a first acquisition step, speech information indicating the speech of a meeting is acquired. In a second acquisition step, term information from which terms related to the meeting can be acquired is acquired. In a generation control step, meeting text information in which at least a part of the speech of the meeting is texturized based on the speech information and the term information is acquired. An information processing system.

[0007] According to the present disclosure, convenience can be further improved in a meeting minutes creation tool.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

[0009] [Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic matters shown in the following embodiments can be combined with each other.

[0010] Incidentally, the program for realizing the software appearing in this embodiment may be provided as a non-transitory computer-readable medium that can be read by a computer, may be provided so as to be downloadable from an external server, or may be provided so that the program is started on an external computer to realize its function on a client device (so-called cloud computing).

[0011] In addition, in this embodiment, the “section” may include, for example, a combination of hardware resources implemented by a circuit in a broad sense and information processing of software that can be specifically realized by these hardware resources. Also, in this embodiment, various types of information are handled, and these types of information are represented, for example, by physical values of signal values representing voltage and current, the high and low of signal values as a binary bit aggregate composed of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculation can be executed on a circuit in a broad sense.

[0012] Also, a circuit in a broad sense is a circuit realized by appropriately combining at least a circuit, circuitry, a processor, a memory, etc. That is, it includes an application specific integrated circuit (ASIC), programmable logic devices (for example, a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0013] In this specification, "A or B" includes A, B, and the combination of A and B. "At least one of A, B, and C" includes A, B, C, the combination of A and B, the combination of A and C, the combination of B and C, and the combination of A, B, and C.

[0014] In this specification, "A based on B" means that B is a factor affecting the determination of A, but it does not exclude the possibility that other elements, such as C, also affect the determination of A in addition to B.

[0015] In this specification, terms such as "first", "second", and "third" are used as labels for the preceding nouns and do not indicate any kind of ordering (such as the order displayed to the user, the order of information processing, etc.).

[0016] 1. Overall Configuration FIG. 1 is an example of a configuration diagram showing an information processing system 1. The information processing system 1 includes a server device 2 and client devices 3 (for example, client devices 3-1, 3-2,...), and these are connected through a network NW. The server device 2 and the client devices 3 can transmit or receive various information from each other via the network NW. Note that the server device 2 and the client devices 3 are examples of information processing devices and are not limited to this embodiment. Each of the client devices 3 is a computer and may be configured by a PC (Personal Computer), a tablet computer, a smartphone, or the like. Note that there is one or more of the server device 2 and the client devices 3. Here, the system exemplified by the information processing system 1 is composed of one or more information processing devices or components. Therefore, even a single server device 2 or a single client device 3 is included in the system exemplified by the information processing system 1.

[0017] 2. Hardware Configuration FIG. 2 is an example of a block diagram showing the hardware configuration of the server device 2. The server device 2 includes a processor 21, a storage unit 22, and a communication unit 23, and these components are electrically connected via a communication bus inside the server device 2. Each component will be further described below.

[0018] The processor 21 performs processing or control of the overall operations related to the server device 2. The processor 21 is, for example, a central processing unit (CPU) (not shown). The processor 21 reads a predetermined program stored in the storage unit 22 to realize various functions related to the server device 2. That is, information processing by software stored in the storage unit 22 is specifically realized by the processor 21, which is an example of hardware, and can be executed as each functional unit included in the processor 21. These will be described in more detail in the next section. Note that the processor 21 is not limited to being single, and may be implemented to have a plurality of processors 21 for each function, or a combination thereof.

[0019] The storage unit 22 stores various information defined as described above. This can be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the server device 2 executed by the processor 21, or as a memory such as a random access memory (RAM) that stores temporarily necessary information (arguments, arrays, etc.) related to program operations. A combination thereof may also be used. In addition to this, the storage unit 22 stores various programs and the like related to the server device 2 executed by the processor 21.

[0020] Although a wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), or wired LAN network communication is preferable for the communication unit 23, wireless LAN network communication, mobile communication such as LTE / 3G, Bluetooth (registered trademark) communication, etc. may be included as necessary. That is, it is more preferable to implement as a collection of these plural communication means. That is, the server device 2 communicates various information via the communication unit 23 and the network NW. For example, the server device 2 is configured to receive voice information acquired by the microphone from the client device 3. Details of these will be described later.

[0021] FIG. 3 is an example of a block diagram showing the hardware configuration of the client device 3. The client device 3 includes a processor 31, a storage unit 32, a communication unit 33, an output unit 34, and an input unit 35, and these components are electrically connected via a communication bus inside the client device 3. Explanation of the processor 31, the storage unit 32, and the communication unit 33 is omitted because it is substantially the same as that of the processor 21, the storage unit 22, and the communication unit 23 in the server device 2.

[0022] For example, the client device 3-1 is described as being used by the recorder user who is the creator of the minutes, and the client device 3-2 (client device 3 other than the client device 3-1) is described as being used by the participant user who is the participant in the meeting, but it is not limited to this. There may be a plurality of participant users. In other embodiments, without distinguishing between the recorder user and the participant user, any user may be able to edit and create the minutes.

[0023] The output unit 34 is configured such that the user can confirm the result of the information processing. The output unit 34 may be included in the housing of the client device 3, or may be externally attached. The output unit 34 is, for example, a display unit, and displays a screen of a graphical user interface (GUI) operable by the user. For this, it is preferable to selectively use display devices such as a CRT display, a liquid crystal display, an organic EL display, and a plasma display according to the type of the client device 3. Further, the output unit 34 is configured to be able to output sound, and specific examples thereof may include devices such as a speaker, headphones, and earphones.

[0024] The input unit 35 is configured to be able to receive an operation on the client device 3 from the user. The input unit 35 may be included in the housing of the client device 3, or may be externally attached. For example, the input unit 35 may be integrated with the output unit 34 and implemented as a touch panel. With a touch panel, the user can input a tap operation, a swipe operation, etc. Of course, instead of the touch panel, a switch button, a mouse, a QWERTY keyboard, etc. may be adopted. That is, the input unit 35 receives the operation input made by the user. The input is transferred as a command signal to the processor 31 via the communication bus, and the processor 31 can execute predetermined control or calculation as necessary. Further, the input unit 35 includes a so-called microphone configured to be able to convert external sound into a signal. The microphone may be directly connected to the server device 2 via the network NW, but is provided, for example, in or connected to the client device 3. The input unit 35 may further include a camera.

[0025] The microphone is configured to generate voice information by collecting the user's speech. Note that the voice information may be temporarily stored in the memory within the client device 3 and does not necessarily need to be stored non-volatilely in the storage unit 32. The voice information generated by the microphone is configured to be transferable to the server device 2 via the network NW. The microphone is not particularly limited, but at least collects voice in the audible range of humans, voice with a frequency between 20 Hz and 20,000 Hz, and converts it into an electrical signal. The voice may be monaural or stereo recording. The sampling rate in the case of digital processing of voice information is, for example, 48,000 Hz, 44,100 Hz, 32,000 Hz, 22,050 Hz, 16,000 Hz, 11,025 Hz, 11,000 Hz, 8,000 Hz, etc. It may also be within any range of the numerical values exemplified here. By increasing the sampling rate, the discretization of the temporal timing of the voice can be performed precisely, and the accuracy of voice recognition can be improved. Also, the data collected by the microphone may be appropriately compressed by the processor 31 of the client device 3, and the compression format at this time may be any of MP3, AAC, WMA, Vorbis, AC3, MP2, FLAC, TAK, etc. By compression, for example, the communication traffic due to data transfer from the client device 3 to the server device 2 can be reduced.

[0026] 3. Functional Configuration In this section, the functional configuration of the embodiment will be described. As described above, the information processing by the software stored in the storage unit 22 is specifically realized by the processor 21 which is an example of hardware, and can be executed as each functional unit included in the processor 21 as shown in FIG. 4. FIG. 4 is an example of a block diagram showing the functions realized by the processor 21 of the server device 2. As shown in FIG. 4, the processor 21 includes, as functional units, an information transmission / reception unit 210, a display control unit 211, an acquisition unit 212, a generation control unit 213, a recording unit 214, a playback unit 215, and a storage control unit 216. Each functional unit executes the steps of its respective name. For example, the display control unit 211 executes the display control step. For example, the acquisition unit 212 executes the first acquisition step, the second acquisition step, and the like.

[0027] The information transmission / reception unit 210 receives, acquires, or accepts information via the network NW and the communication unit 23. Further, the information transmission / reception unit 210 transmits information via the communication unit 23 and the network NW.

[0028] The display control unit 211 controls the screen information to be displayed on the output unit 34 of the client device 3. Note that the software may be the visual information itself generated in a mode visible to the user, such as a screen, an image, an icon, text, etc., or may be, for example, rendering information for causing various terminals to display visual information such as a screen, an image, an icon, text, etc.

[0029] The acquisition unit 212 acquires information for use by the generation control unit 213. Acquisition is a concept including both a mode of receiving data etc. from other devices and a mode of generating data etc. by itself.

[0030] The generation control unit 213 generates various information using the learned model.

[0031] The recording unit 214 stores the audio information in the storage unit 22. In a meeting, video information may be acquired in addition to or instead of the audio information. The video information is information indicating the video of the meeting.

[0032] The playback unit 215 plays back the voice information recorded by the recording unit 214.

[0033] The memory control unit 216 stores various information in the memory unit 22.

[0034] 4. Information Processing Method Next, with reference to FIG. 5, an example in which a minutes creation tool creates minutes using the information processing system 1 of the present embodiment will be described. FIG. 5 is an example of an activity diagram showing an outline of information processing executed by the information processing system 1. The meeting is assumed to be held using a meeting tool, but is not limited thereto. For example, the meeting may be held with all participants gathering in a meeting room, and the voice may be acquired by a microphone arranged in the meeting room. In that case, it is assumed that there is one client device 3. Also, meetings are assumed to be for various purposes such as business negotiations, personnel interviews, and employment interviews. When the minutes screens 5 to 6 in FIGS. 6 to 7 are displayed before and during the meeting, it is assumed that the meeting tool and the minutes creation tool are linked or integrated.

[0035] The minutes creation tool of the present embodiment is a tool for creating minutes. The minutes creation tool may be defined as, for example, at least one of a tool in which an information input area 71 and a meeting text area 73 described later can be used, a tool capable of recording the elapsed time of the meeting when characters are input, and a tool capable of reflecting the result of text generation from the voice of the meeting in the minutes.

[0036] (Activity A1) First, generation information may be generated prior to the start of the meeting. That is, the processor 31 receives an instruction to input input information into the learned model via an operation on the input unit 35 by the user. The processor 31 transmits the instruction to the server device 2 via the communication unit 33 and the network NW. The information transmission / reception unit 210 receives the instruction from the client device 3 via the network NW and the communication unit 23. The acquisition unit 212 acquires input information based on the instruction. The generation control unit 213 acquires generation information generated by inputting the acquired input information into the learned model. More specifically, for example, the generation control unit 213 acquires generation information generated by inputting the input information into the learned model in a manner that the input information is included as part of a prompt. The display control unit 211 generates screen information in which the generation information is included in the screen. The information transmission / reception unit 210 transmits the generation information to the client device 3 via the communication unit 23 and the network NW. The processor 31 receives the generation information from the server device 2 via the network NW and the communication unit 33. The processor 31 displays a screen including the generation information. Note that the generation information is assumed to be displayed, for example, in the generation area 55 of the minutes screen 5 described later in FIG. 6, but the minutes screen 5 does not necessarily have to be displayed before the meeting and may be displayed on a web browser or on other sharing tools. Note that the activity A1 may be omitted.

[0037] The input information is information input into the learned model and may include at least one of related information, participant information, past speech information, material information, purpose information, insight information, and sharing tool information. The input information is assumed to be mainly composed of text or images, but may include information in any other form such as voice and video. The related information is information related to the meeting. The acquisition unit 212 may acquire, as related information, information obtained from at least one of emails related to the meeting, the history of inquiries sent to the organization to which the persons related to the meeting belong, and the shared tool information described later. The acquisition unit 212 may access the mailer used by the user to acquire emails. Further, the acquisition unit 212 may acquire the history of inquiries sent to the organization to which the persons related to the meeting belong from a mailbox, a database, or the like. The participant information is a list of the persons participating in the meeting. The participant information may identify attendees, absentees, latecomers, and early leavers. For example, the acquisition unit 212 may acquire the participant information from the users invited to the meeting via a meeting tool or a minutes creation tool. Further, the acquisition unit 212 may acquire the participant information from the users participating in the meeting via the meeting tool. The participant information acquired from the meeting tool may be included in the shared tool information. The past speech information is information indicating the content of the speech during the past meeting associated with the participants of the past meeting. The acquisition unit 212 may acquire the past speech information based on the past minutes information via the minutes creation tool. The past minutes information is information indicating the minutes of the past meeting and may be managed, for example, by a minutes creation tool or may be any file (such as a text file) used as a memo of the MTG. The material information is information obtained from related materials. More specifically, it is information obtained from the materials used in the related meeting or the materials planned to be used in the meeting, and it is assumed that the materials include text, images, audio, or video. The acquisition unit 212 may acquire information such as text and images included in the materials as the material information. The material information may include the resumes, portfolios, etc. of the candidates. The purpose information is information indicating the purpose of the meeting, and may include content related to the theme, title, topic, etc. of the meeting. The acquisition unit 212 may acquire the purpose information based on the text included in the notice of the meeting such as emails, meeting tools, and schedule adjustment tools, or may acquire the purpose information based on the text input by the user such as "Management Meeting", "Regular Meeting of Mr. ABC", "Budget for this year", "Discuss the personnel change of Mr. A". In the former case, the purpose information may be included in the shared tool information. The insight information is information indicating the degree of interest, emotion, sense of warmth, inference, etc. of the persons related to the meeting. The persons related to the meeting may include at least one of the participants of the meeting, the past participants of the related meeting, the customers of the participants of the meeting, and the supervisors of the participants of the meeting. The inference is the user's thinking that it would be better in this way based on the content of the speech, purpose, emotion, etc. The insight information may be related to at least one of the topic, the topic outside the scope of the topic, the content of the speech, and the news. The acquisition unit 212 may use the learned model to generate the insight information from at least one of the text included in the transcription information, the text included in the meeting text information, the tone of the voice included in the voice information, and the expression or tone of the voice of the participants included in the video information. The shared tool information is information that can be obtained from the shared tool. For example, the acquisition unit 212 acquires the information managed by the shared tool as shared tool information through API (Application Programming Interface) cooperation, CSV (Comma Separated Values) cooperation, scraping, etc. The shared tool information may include information generated by a learned model functioning within the shared tool. The shared tool information may include information indicating a contract draft, the latest legal amendments related to the terms to be negotiated, case laws, the latest research papers, industry news, students' test results, assignment submission status, online learning records, students' career aspirations, learning attitudes, the resumes, claims, papers, articles, SNS posts of the experts to be interviewed, the past defective product occurrence rate, the number of delivery delays, inspection results, etc. Further, the shared tool information may include information indicating improvement requirements for suppliers such as the required quality level, target cost, delivery compliance rate, etc. Further, the shared tool information may include information indicating the event scale, theme, budget, venue type, etc. desired by the client, past emails · RFP (Request for Proposal), success examples of similar events, cost-effectiveness analysis, the latest laws related to new policy areas (education policy, environmental regulations, etc.), statistical data, expert opinions, the past statements and / or positions of each stakeholder (citizen groups, industry groups, NPOs), the client's owned assets, investment history, risk tolerance, the latest economic indicators, stock price trends, bond yields, exchange rate fluctuations, past media exposure data related to the company, SNS mentions, news reporting trends, usage logs, feedback forms, known defect lists, improvement request lists.In addition, the shared tool information may include information indicating functional requirements submitted by various groups (development team, sales team, customer success team, etc.), analysis results of the success or / and failure of past releases, similar cases in the customer industry, contract details with clients, past policy proposals, client needs analysis results, promotion requirements based on personnel policies (work history, achievements, leadership indicators), repositories, current contracts, various compliance reports, legal amendment information, concept proposals submitted by agents, measurement data on the effectiveness of past campaigns, brand guidelines, customer survey results, influence (number of followers, engagement rate), past collaboration achievements, brand compliance, profiles of test subjects, devices used, success rate of test tasks, frequently occurring UI issues in the problem log, points of insufficient understanding of functions, corporate risk matrix, compliance reports, internal control evaluation results, measurement data on the effectiveness of past countermeasures, past bug occurrence patterns, performance data for the previous quarter, income statement, balance sheet, cash flow statement, server operation log, error log, response time statistics, past incident reports, NPS questionnaires, customer reviews, SNS mentions, user dissatisfaction points, learning test results of new employees, progress of online courses, feedback comments, lead scores in CRM, email opening rates, web behavior analysis, past successful campaigns for similar leads, latest research papers, patent application trends, in-house test results, budget constraints, financial statements of target companies, industry average variations, past successful M&A cases, prices, non-financial factors, management retention conditions, internal audit reports, past violation notices, compliance score, past corrective measures, measurement of the effectiveness of past events by sponsor (brand exposure, number of leads obtained, SNS engagement), value required by sponsors (target audience, exposure frequency, in-venue branding locations, etc.), customer segment analysis, competitive comparison, message test results, medical guidelines, latest research papers, etc. The shared tool information may also include information indicating the utilization status, operation rate, inquiry history, etc. of the products and / or services used by customers, and information indicating multiple KPIs such as the past performance evaluation sheets, skill test results, 360-degree feedback comments, etc. of the evaluated person, as well as sales, costs, customer satisfaction, market indicators, etc.

[0038] The sharing tool is a tool capable of sharing data with the minutes creation tool and may include any tool. The sharing tool may include, for example, CRM (Customer Relationship Management) tools, document creation tools, presentation tools, spreadsheet tools, project management tools, communication tools, design tools, code repository & version control tools, database management tools, CMS (Content Management System) tools, email marketing tools, recruitment management systems, storage tools, workflow tools, issue management tools, schedule adjustment tools, meeting tools, MA (Marketing Automation) tools, SNS (Social Networking Service) tools, intellectual property databases, legal databases, etc. The sharing tool may include those provided by SaaS (Software as a Service). Also, the sharing tool may include any website that provides an API and / or is capable of scraping. The CRM tool is a tool that supports customer relationship management. The CRM tool is used for centralized management of customer information, customer information analysis, promotion management, etc. The CRM tool may be, for example, Salesforce Sales Cloud, Mazrica Sales, Zoho CRM, GENIEE SFA / CRM, Kintone, etc. The document creation tool is a tool that can centrally manage the creation, management, storage, sharing, utilization, disposal, etc. of documents. Also, the document creation tool is used, for example, to improve the efficiency and convenience of document handling. The document creation tool may be Notion, NotePM, Google Docs, Evernote, Zoho Learn, Cacoo, astah, Enterprise Architect, Gridraw, draw.io, etc. The presentation tool is used to support presentations. The spreadsheet tool is used to create spreadsheets. The project management tool is used to manage projects. The communication tool is used to conduct text or voice communication among multiple users, and it may be LINE, Chatwork, Slack, WeChat, etc. The communication tool may include groupware, etc. The design tool is used to assist in creating designs. The code repository & version control tool is used to assist in managing source code. The database management tool is used to manage databases. The CMS (Content Management System) tool is used to assist in creating websites. The email marketing tool is used to assist in delivering emails. The recruitment management system may be an ATS (Applicant Tracking System), etc. The storage tool is a tool that can save data in a disk space for file storage on the Internet, and it may be Google Drive, OneDrive, etc. The workflow tool is a tool for streamlining in-house business procedures and may include Jobcan, SmartHR, etc. The issue management tool is a tool for managing issues such as projects and tasks, and it may be Jira, etc. The schedule adjustment tool is a tool for registering schedules, tasks, goals, etc., visually organizing and displaying them, and sharing them with other employees, and it may be Calendly, Google Calendar, Outlook Calendar, etc. The meeting tool is a tool for holding meetings with parties in remote locations using terminals such as personal computers and smartphones connected to the network NW, and it may be Zoom, Google Meet, Teams, etc. The MA tool is a tool for managing, automating, or streamlining marketing activities in customer acquisition. The SNS tool is a tool that enables people to connect and share information on the Internet, and may include X, Instagram, YouTube, Facebook, LinkedIn (registered trademark), etc. The intellectual property database is a database related to intellectual property rights, and may include J-Plat Pat, PATENTSCOPE, etc. The legal database is a database related to information such as laws and case precedents, and may include e-Gov, etc.

[0039] By inputting relevant information as input information into the learned model, the generation control unit 213 may obtain summary key point information as generation information. The summary key point information is information indicating key points or summaries generated from the information included in the relevant information. Thereby, since the user can check the information organized before the meeting, the user can efficiently preview the content of the previous meeting.

[0040] By inputting at least one of participant information, past speech information, and material information as input information into the learned model, the generation control unit 213 may obtain interest information as generation information. The interest information is information indicating who has analyzed which part with high interest. For example, information indicating that the sales manager has a high interest in sales may be obtained from the content of the sales manager's past speeches. Thereby, since the user can check what each participant is interested in before the meeting, it becomes possible to smoothly conduct the meeting.

[0041] By inputting at least one of past speech information and insight information as input information into the learned model, the generation control unit 213 may acquire presentation information as generation information. The presentation information is information prepared for a meeting based on insights from past meetings, the feelings of participants, etc. Further, the presentation information may include information indicating what should be agreed upon or discussed in the meeting, such as content with which participants have positive feelings and content with which they have negative feelings. Further, the presentation information may be information indicating the content of remarks, the content of proposals, etc. that should be made during the meeting based on past reference conversations. Thereby, it is possible to support the user so that the user can grasp best practices such as what kind of dialogue should be made based on the reference information and have an appropriate dialogue.

[0042] By inputting at least one of past speech information, voice information, video information, past minutes information, and gist information as input information into the learned model, the generation control unit 213 may acquire meeting content information as generation information. The meeting content information may be character information generated based on voice information, information in the information input areas 51, 61, 71, or the meeting text areas 53, 63, 73, 90. The meeting content information includes at least one of information on the summary of the meeting, information on the ToDo items of the meeting, information on the decisions made in the meeting, information on the key points of the meeting, information on the remarks of the speakers participating in the meeting, information on the related parties of the meeting, information on the Q&A in the meeting, information on issues, information on the cases under consideration, and information on requests. Here, the meeting content information is all related to meetings before the previous one. The summary information is a sentence that concisely summarizes the content of a specific information source or conveys the essential part. The summary may be restricted to a predetermined number of characters or may not be restricted by the number of characters. The ToDo item information is an item determined during the meeting as the next action, such as the creation of materials. The decision item information is an item on which agreement was reached during the meeting, such as the date and time of the next meeting. The key information is a summary of one or more sentences in bullet points of important points. The key points may or may not be restricted to a predetermined number of characters. The information of the statements of the speakers participating in the meeting may be information summarized for each speaker, information separated for each statement, information in which the information of each statement is associated with time, or information in which the information of each statement is associated with the speaker. As described above, the information of the relevant persons may include at least one of the participants in the meeting, the past participants in the relevant meetings, the customers of the participants in the meeting, and the supervisors of the participants in the meeting. In addition, the information of the relevant persons may include the information of the names or surnames of the persons who appear during the meeting. Also, the information of the relevant persons may be specified together with the position, title, company name, department name, etc. The information of the Q&A is the information of the combination of the questions and answers during the meeting. The information of the Q&A in the meeting may be information in which a plurality of answers are associated with one question, or information in which a plurality of questions are associated with one answer. The information of the issues is the information indicating the problems to be solved when promoting the project, and may include the information indicating the issues from specific viewpoints such as frequently mentioned issues and unsolved issues. The information of the project under consideration is the information indicating the tasks that have not been completed (in progress). The information of the requests is the information indicating the requests raised by the relevant persons. As a result, the user can check the information organized before the meeting, so that the user can efficiently preview the content of the previous meeting.

[0043] For example, before the business negotiation, by inputting the CRM, MA tool, SNS, past emails, inquiry history, etc. into the learned model as input information, the generation control unit 213 may generate information that organizes or summarizes the customer information as the generated information, or may generate information that summarizes the business content, recent news, areas of interest, etc. as the generated information. For example, before business negotiations, by inputting past meeting minutes, negotiation memos, etc. as input information into the learned model, the generation control unit 213 may generate important agreement items, unresolved issues, items for consideration, etc. as generated information. For example, before business negotiations, by inputting past dialogue logs as input information into the learned model, the generation control unit 213 may generate information summarizing issues or requests frequently mentioned by the customer as generated information.

[0044] For example, before an employment interview, by inputting a candidate's resume, SNS (such as LinkedIn (registered trademark)) profile, portfolio, etc. as input information into the learned model, the generation control unit 213 generates information summarizing the candidate's strengths, skill set, past achievements, etc. as generated information. For example, before an employment interview, by inputting past questions for applicants as input information into the learned model, the generation control unit 213 may generate a set of questions that match the requirements of the position (such as a technical position) as generated information.

[0045] For example, before a meeting, by inputting the status of tasks collected from team members as input information into the learned model, the generation control unit 213 may generate information in a state where the progress of the tasks determined at the previous meeting is visible as generated information. For example, before a meeting, by inputting past meeting logs, emails, etc. as input information into the learned model, the generation control unit 213 may extract themes such as unresolved issues and items that must be considered this time, and generate a list of topics as generated information.

[0046] For example, before a phone call, by inputting past inquiry records, purchase histories, chat logs, etc. as input information into the learned model, the generation control unit 213 may generate frequently occurring problems, improvement requests, etc. as generated information. For example, before making a phone call, by inputting information indicating what the customer is seeking as input information into the learned model, the generation control unit 213 may generate, as generation information, an optimal solution, a link to a FAQ document, etc.

[0047] For example, before negotiation, by inputting a contract draft as input information into the learned model, the generation control unit 213 may generate, as generation information, information indicating issues, risks, negotiation points, etc. of important clauses. For example, before negotiation, by inputting the latest legal amendments, case laws, etc. related to the clauses of the negotiation target as input information into the learned model, the generation control unit 213 may generate, as generation information, information summarizing the input information.

[0048] For example, before an interview, by inputting the purchasing behavior, website usage history, questionnaire answers, etc. of the interview target as input information into the learned model, the generation control unit 213 may generate, as generation information, information indicating tendencies that the target should pay attention to. For example, before an interview, by inputting a goal (such as user usability confirmation of a new function) as input information into the learned model, the generation control unit 213 may generate, as generation information, a question list for user usability testing.

[0049] For example, before brainstorming, by inputting the latest research papers, patent databases, industry news, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, information summarizing the trends of related technologies. For example, before brainstorming, by inputting ideas shelved in past brainstorms, unsolved technical problems, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, information listing the information included in the input information.

[0050] For example, before the interview, by inputting the student's test results, assignment submission status, online learning records, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, information that aggregates the information included in the input information and identifies the weak areas. For example, before the interview, by inputting the student's career aspirations, learning attitude, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, measures for motivation improvement, specific learning plans, etc.

[0051] For example, before the interview, by inputting the resume, claims, papers, articles, SNS posts, etc. of the expert being interviewed into the learned model as input information, the generation control unit 213 may generate, as generation information, an interview brief. For example, before the interview, by inputting information indicating the interview theme into the learned model as input information, the generation control unit 213 may generate, as generation information, questions that match the theme and the needs of the readers.

[0052] For example, before the strategic meeting with the client's management layer, by inputting the latest trends in the industry to which the client company belongs, competitive strategies, consumer behavior data, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, materials for the meeting for formulating strategies. For example, before the strategic meeting with the client's management layer, by inputting the past discussion results, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, information that highlights the strategic issues to be focused on this time (such as entering a new market, reforming the cost structure, etc.).

[0053] For example, before the customer onboarding session, by inputting the usage logs of the SaaS product, etc. into the learned model as input information, the generation control unit 213 may identify the functions that the customer has not yet mastered and the frequently used functions, and generate, as generation information, the key explanation items for the onboarding session. For example, before a customer onboarding session, by inputting information indicating the customer's industry type, role, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, a question guide on which issues should be preferentially asked according to the customer's industry type, role, etc.

[0054] For example, before a quality improvement meeting with a supplier, by inputting the past defective product rate, number of delivery delays, inspection results, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, information calculating the priority of the quality items to be discussed this time. For example, before a quality improvement meeting with a supplier, by inputting improvement requirements for the supplier such as the required quality level, target cost, delivery compliance rate, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, a draft document summarizing the improvement requirements.

[0055] For example, before a client meeting (event planning), by inputting the event scale, theme, budget, venue type, etc. desired by the client from past emails and RFPs into the learned model as input information, the generation control unit 213 may generate, as generation information, information summarizing the input information. For example, before a client meeting (event planning), by inputting successful cases of similar events, cost - effectiveness analysis, etc. into the learned model as input information, the generation control unit 213 may generate, as generation information, a plurality of planning proposals and / or a comparison table for the meeting.

[0056] For example, before a stakeholder meeting, by inputting the latest laws and regulations, statistical data, expert opinions, etc. regarding new policy areas (education policy, environmental regulations, etc.) into the learned model as input information, the generation control unit 213 may generate, as generation information, information summarizing the input information. For example, before a stakeholder meeting, by inputting the past statements, positions, etc. of each stakeholder (such as civic groups, industry groups, NPOs, etc.) into the learned model as input information, the generation control unit 213 may generate information summarizing the input information as generation information.

[0057] For example, before a customer asset management consultation, by inputting the customer's owned assets, investment history, risk tolerance, etc. into the learned model as input information, the generation control unit 213 may generate information indicating an optimal asset allocation plan and candidate products as generation information. For example, before a customer asset management consultation, by inputting the latest economic indicators, stock price trends, bond yields, exchange rate fluctuations, etc. into the learned model as input information, the generation control unit 213 may generate a market report summarizing the input information as generation information.

[0058] For example, before a media briefing meeting, by inputting the past media exposure data related to the company, SNS mentions, news reporting trends, etc. into the learned model as input information, the generation control unit 213 may generate information that summarizes and selects the points to be emphasized in the press briefing as generation information. For example, before a media briefing meeting, by inputting the past media exposure data related to the company, SNS mentions, news reporting trends, etc. into the learned model as input information, the generation control unit 213 may generate information indicating the questions (competitive comparison, USP of new products, concerns) expected from journalists and example answers as generation information.

[0059] For example, before a user community symposium, by inputting the user posting data on online forums, SNS, etc. into the learned model as input information, the generation control unit 213 may extract frequently occurring topics, dissatisfaction points, etc., and generate information specifying the themes to be discussed at the symposium as generation information. For example, before a user community symposium, by inputting the information of the participants as input information into the learned model, the generation control unit 213 may summarize the key points by participant attribute (heavy user, new customer, early adopter, etc.) and generate information indicating a question plan as generation information.

[0060] For example, before customer beta test feedback, by inputting usage logs, feedback forms, etc. as input information into the learned model, the generation control unit 213 may generate information indicating frequently used functions, neglected functions, etc. as generation information. For example, before customer beta test feedback, by inputting a known defect list, improvement request list, etc. as input information into the learned model, the generation control unit 213 may organize the list and generate information scoring which issues are prioritized as generation information.

[0061] For example, before a stakeholder consensus formation meeting, by inputting function requirements etc. submitted from various groups (development team, sales team, customer success team, etc.) as input information into the learned model, the generation control unit 213 may generate a summary considering priority, man-hours, market impact, etc. as generation information. For example, before a stakeholder consensus formation meeting, by inputting the analysis results of the success and / or failure of past releases as input information into the learned model, the generation control unit 213 may generate a draft of function candidates to be introduced in the next version and a phased rollout plan as generation information.

[0062] For example, before a one-on-one interview with team members, by inputting the sales performance, customer response history, CRM, etc. of the members as input information into the learned model, the generation control unit 213 may generate information in which the pipeline status is dashboarded as generation information. For example, before a one-on-one interview with a team member, by inputting the logs, feedback, etc. of past one-on-one interviews as input information into the learned model, the generation control unit 213 may generate, as generation information, topics to be covered in the interview, such as factors for not achieving the target and areas of skill deficiency, which have been extracted.

[0063] For example, before a regular meeting with an important customer, by inputting the utilization status, operation rate, inquiry history, etc. of the products and / or services used by the customer as input information into the learned model, the generation control unit 213 may generate, as generation information, the degree of customer KPI achievement, areas for improvement, etc. For example, before a regular meeting with an important customer, by inputting similar cases in the customer's industry, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, information indicating cases where results have been achieved through measures to promote utilization, introduction of new functions, etc.

[0064] For example, before a regular report meeting with a client, by inputting the contract details with the client as input information into the learned model, the generation control unit 213 may generate, as generation information, information including the progress status, achievement indicators, budget utilization rate, risk items, etc. For example, before a regular report meeting with a client, by inputting past policy proposals, client needs analysis results, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, improvement proposals to be emphasized in this report meeting.

[0065] For example, before a promotion / evaluation interview, by inputting the past performance evaluation sheets, skill test results, 360-degree feedback comments, etc. of the person being evaluated as input information into the learned model, the generation control unit 213 may generate, as generation information, information that has been analyzed or summarized from the input information. For example, before a promotion / evaluation interview, by inputting promotion requirements (job history, achievements, leadership indicators) based on the personnel policy as input information into the learned model, the generation control unit 213 may generate, as generation information, information for evaluating and making a judgment on the input information.

[0066] For example, before the influencer meeting, by inputting influence (number of followers, engagement rate), past collaboration achievements, brand suitability, etc. as input information into the learned model, the generation control unit 213 may generate the scored information as generation information. For example, before the influencer meeting, by inputting influence (number of followers, engagement rate), past collaboration achievements, brand suitability, etc. as input information into the learned model, the generation control unit 213 may generate a brand message, a posting idea tailored to the target customer group, a creative concept, etc. as generation information.

[0067] For example, before the debriefing after the usability test, by inputting the profiles of the test subjects, the devices used, the success rate of the test tasks, the problem log, etc. as input information into the learned model, the generation control unit 213 may generate the information in which the input information is automatically aggregated as generation information. For example, before the debriefing after the usability test, by inputting frequently occurring UI issues, points of insufficient understanding of functions, etc. as input information into the learned model, the generation control unit 213 may generate the information indicating the priority issues to be discussed as generation information.

[0068] For example, before the supplier quality review meeting, by inputting the supplier's delivery compliance rate and defect rate data as input information into the learned model, the generation control unit 213 may generate the information including the areas that need improvement as generation information. For example, before the supplier quality review meeting, by inputting the past improvement requirements, agreement history, etc. as input information into the learned model, the generation control unit 213 may generate the information indicating the points to be pointed out in this review as generation information.

[0069] For example, before the risk assessment meeting, by inputting the corporate risk matrix, compliance reports, internal control assessment results, etc. as input information into the learned model, the generation control unit 213 may generate information indicating the priority order of important risks as the generated information. For example, before the risk assessment meeting, by inputting the measurement data of the effects of past countermeasures as input information into the learned model, the generation control unit 213 may generate information indicating the risk countermeasures that need to be strengthened this time as the generated information.

[0070] For example, before the creative review meeting with the advertising agency, by inputting the measurement data of past campaign effects, brand guidelines, customer survey results, etc. as input information into the learned model, the generation control unit 213 may generate information indicating the direction and essential elements of the advertising expression as the generated information. For example, before the creative review meeting with the advertising agency, by inputting the concept proposal submitted by the agency as input information into the learned model, the generation control unit 213 may generate information listing the parts where the brand tone deviates and areas for improvement as the generated information.

[0071] For example, before the regular meeting with the external legal counsel, by inputting the current contract, various compliance reports, legal amendment information, etc. as input information into the learned model, the generation control unit 213 may generate information indicating the summary of the matters to be consulted with the counsel as the generated information. For example, before the regular meeting with the external legal counsel, by inputting the current contract, various compliance reports, legal amendment information, etc. as input information into the learned model, the generation control unit 213 may generate information in which the points of the risks of unresolved compliance are raised as the generated information.

[0072] For example, before the KPI review meeting, by inputting multiple KPIs such as sales, costs, customer satisfaction, and market indicators as input information into the learned model, the generation control unit 213 may generate information in which the ratio compared to the previous period and / or the target ratio is calculated as the generated information. For example, before the KPI review meeting, by inputting multiple KPIs such as sales, costs, customer satisfaction, and market indicators as input information into the learned model, the generation control unit 213 may generate, as generation information, information including the information on the factors for upward and / or downward fluctuations that should be taken and the information indicating the points for consideration.

[0073] For example, before the regular development team code review meeting, by inputting the repository as input information into the learned model, the generation control unit 213 may generate, as generation information, information summarizing the list of new pull requests, the differences in code changes, test results, etc. For example, before the regular development team code review meeting, by inputting the past bug occurrence patterns as input information into the learned model, the generation control unit 213 may generate, as generation information, information including the code areas to be noted, quality indicators, etc.

[0074] For example, before the meeting for preparing the earnings presentation for investors, by inputting the actual data for the previous quarter, income statement, balance sheet, cash flow statement, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, information indicating the draft presentation for investors. For example, before the meeting for preparing the earnings presentation for investors, by inputting the actual data for the previous quarter, income statement, balance sheet, cash flow statement, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, the assumed Q&A list for shareholder questions based on past trends.

[0075] For example, before the operation improvement meeting, by inputting the server operation logs, error logs, response time statistics, etc. as input information into the learned model, the generation control unit 213 may generate, as generation information, the overloaded locations, bottlenecks, etc. For example, before the operation improvement meeting, by inputting the past incident report as input information into the learned model, the generation control unit 213 may summarize the room for improvement and generate a list of infrastructure strengthening measures to be considered as generation information.

[0076] For example, before the feedback aggregation meeting, by inputting the NPS questionnaire, customer reviews, SNS mentions, etc. as input information into the learned model, the generation control unit 213 may perform text mining on the customer feedback and generate information visualizing positive or / and negative opinions as generation information. For example, before the feedback aggregation meeting, by inputting the user's dissatisfaction points as input information into the learned model, the generation control unit 213 may generate information that classifies the dissatisfaction points for each specific customer segment (new customers, repeat customers, senior users) as generation information.

[0077] For example, before the new employee training follow-up interview, by inputting the learning test results, online course progress, feedback comments, etc. of new employees as input information into the learned model, the generation control unit 213 may generate information visualizing the proficiency level as generation information. For example, before the new employee training follow-up interview, by inputting the learning test results, online course progress, feedback comments, etc. of new employees as input information into the learned model, the generation control unit 213 may generate information identifying the skill areas to be strengthened (presence skills, product knowledge, understanding of internal processes, etc.) as generation information.

[0078] For example, before the lead nurturing strategy meeting, by inputting the lead score in the CRM, email opening rate, web behavior analysis, etc. as input information into the learned model, the generation control unit 213 may generate information summarizing the customer segments for nurturing and the stage-specific issues as generation information. For example, before a lead nurturing strategy meeting, by inputting past successful campaigns for similar leads and the like as input information into the learned model, the generation control unit 213 may generate information indicating a summarized recommended strategy or the like as generation information.

[0079] For example, before a technology roadmap formulation meeting, by inputting the latest research papers, patent application trends, in-house test results, etc. as input information into the learned model, the generation control unit 213 may generate information indicating candidate areas for technological breakthroughs as generation information. For example, before a technology roadmap formulation meeting, by inputting development resources, budget constraints, etc. as input information into the learned model, the generation control unit 213 may generate information that has drafted a roadmap plan considering development resources, budget constraints, etc. as generation information.

[0080] For example, before an M&A negotiation meeting, by inputting the financial statements of the target company, industry average variations, past M&A success cases, etc. as input information into the learned model, the generation control unit 213 may generate information indicating estimates such as the acquisition price range and synergy effects as generation information. For example, before an M&A negotiation meeting, by inputting information indicating price, non-financial factors, management retention conditions, etc. as input information into the learned model, the generation control unit 213 may generate information that has picked up points predicted to be emphasized by the other party in the negotiation as generation information.

[0081] For example, before a compliance rectification meeting, by inputting an internal audit report, past violation points, compliance score, etc. as input information into the learned model, the generation control unit 213 may generate information listing the areas that need to be rectified as generation information. For example, before a compliance rectification meeting, by inputting past rectification measures as input information into the learned model, the generation control unit 213 may generate information identifying the effects of the past and unresolved risk areas as generation information.

[0082] For example, before the sponsor meeting, by inputting the measurement of the effects by sponsor for past events (brand exposure, number of leads obtained, SNS engagement) as input information into the learned model, the generation control unit 213 may generate information summarizing the input information as generation information. For example, before the sponsor meeting, by inputting the value required by the sponsor (target layer, exposure frequency, in-venue branding location, etc.) as input information into the learned model, the generation control unit 213 may generate information summarizing the input information as generation information.

[0083] For example, before the launch sharing meeting, by inputting customer segment analysis, competitive comparison, message test results, etc. as input information into the learned model, the generation control unit 213 may generate information summarizing the USP (unique value proposition) to be emphasized at the launch as generation information.

[0084] The learned model is an AI (Artificial Intelligence) equipped with a language model such as a Transformer (including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, GPT-3), BERT (Bidirectional Encoder Representations from Transformers), BART (Bidirectional and Auto-regressive Transformer), etc.) or a Recurrent Neural Network (RNN), and includes a generative AI. For example, the generation control unit 213 obtains generation information generated by inputting into the generative AI as the learned model in a manner where the input information is included as part of the prompt. The language model generates text by applying an arbitrary natural language processing (NLP) algorithm.

[0085] The learned model may be able to integratively process multiple types of data multimodally. In this embodiment, the learned model may perform information processing by combining voice and text.

[0086] A language model is an example of a learned model by a machine learning algorithm. Specific algorithms for machine learning include the nearest neighbor method, the naive Bayes method, decision trees, support vector machines, deep learning (deep learning) using neural networks, etc. The learned model can appropriately apply the above algorithms.

[0087] The learned model has, as artificial intelligence, a learned model constructed by a learning method such as supervised learning, unsupervised learning, or semi-supervised learning. In supervised learning, machine learning is performed using teacher data (learning data). The teacher data is composed of a pair of input data for learning and output data (correct data). Also, the language model may be not only a model trained for a specific task but also a general-purpose model that can be generally used for a wide range of tasks.

[0088] The learned model, as artificial intelligence, includes a learning model for general natural language processing such as a large language model (Large Language Models (LLM)) that has learned a vast amount of data. Such a general learning model includes a language model that can handle various tasks without fine-tuning by one-shot learning, few-shot learning, etc. The artificial intelligence used in each functional part of the processor 21 may be a separate learning model or a common general learning model. Natural language processing may include information processing such as morphological analysis, syntactic analysis, semantic analysis, and context analysis.

[0089] The artificial intelligence included in the learned model is capable of performing additional learning. For example, the learned model learns whether the presented generated information has been used by the user. That is, the learned model uses, as teacher data, the generated information presented together with the generated information labeled with the edited generated information as feedback on the generated information created and presented by the artificial intelligence, performs additional learning, and is fine-tuned. As a result, the generated information output from the learning model is optimized and presented to the user.

[0090] FIG. 6 is a diagram showing an example of the minutes screen 5 before a meeting in an embodiment. The minutes screen 5 is a screen for creating minutes before a meeting. The minutes screen 5 includes a meeting information area 50, an information input area 51, a timestamp area 52, a meeting text area 53, a playback area 54, and a generation area 55.

[0091] The meeting information area 50 is an area where an overview of the meeting is displayed. The overview of the meeting includes, for example, information indicating the date and time of the meeting such as "2022 / 4 / 23 10:00-11:00", and information indicating participants, absentees, and the creator of the minutes. The URL of the storage location of the materials used in the meeting may be displayed in the overview of the discussion.

[0092] The information input area 51 is configured to be able to display the results of the input made by the recording user. The input information may be treated as recording information. The recording information may be information consisting of images, sounds, and videos in addition to information consisting of text. The text information in the attached file may be information containing characters, may be itemized information, or may be information consisting only of numbers, symbols, words, etc.

[0093] The timestamp area 52 is configured to be able to display time information regarding the elapsed time of the meeting when information is input to the information input area 51. Since the meeting has not started in the state before the meeting, the time information is not recorded in the timestamp area 52 of FIG. 6.

[0094] The conference text area 53 is configured to display the text generated from the conference audio. In the state before text generation (i.e., the state before or during the conference), no speech recognition information or conference text information is displayed in the conference text area 53.

[0095] The playback area 54 is configured to play the audio during the conference.

[0096] The generation area 55 is an area where generation information is displayed.

[0097] (Activity A2) Subsequently, in response to receiving an instruction to start a conference from the client device 3-1 of the recording user, the processor 21 of the server device 2 starts the processing during the conference. That is, as the processing during the conference, the processor 21 starts counting the elapsed time of the conference and starts recording by the minutes tool. The processor 21 executes the information processing of A3 to A5 in parallel.

[0098] (Activity A3) Subsequently, the information transmission / reception unit 210 receives the audio information acquired by the microphones of the respective users (recording user and participating users) from the client device 3. The acquisition unit 212 acquires the audio information indicating the audio of the conference. Subsequently, the information transmission / reception unit 210 transmits the received audio information from a certain user to the client device 3 operated by the users other than that user. For example, the information transmission / reception unit 210 transmits the received audio information from the client device 3-2 operated by the participating user to the client device 3-1 of the recording user. Subsequently, the processor 31 of the participating user outputs the audio of the audio information from the output unit 34. The recording unit 214 records the audio information received from the client device 3 used by each user in the storage unit 22. Until the conference ends, the processor 21 repeatedly executes the above information processing of Activity A3.

[0099] (Activity A4) In parallel with Activity A3, the processor 31 of the client device 3-1 of the recording user receives input of recording information such as characters into the information input area 71 via an operation on the input unit 35 by the recording user. The information transmission / reception unit 210 receives the recording information from the client device 3-1 used by the recording user. Subsequently, the display control unit 211 appropriately executes processing related to the information input area 61 and the timestamp area 62, which will be described later. Until the meeting ends, the processor 21 repeatedly executes the above information processing of Activity A4. Note that the displayed screen is, for example, the minutes screen 6 to be described later in FIG. 7, but the minutes screen 6 does not necessarily have to be displayed during the meeting. Note that Activity A4 may be omitted.

[0100] Next, while showing FIG. 7, the minutes screen 6 during the meeting will be described. FIG. 7 is a diagram showing an example of the minutes screen 6 during the meeting in the embodiment. The minutes screen 6 is a screen for creating minutes during the meeting. The minutes screen 6 includes a meeting information area 60, an information input area 61, a timestamp area 62, a meeting text area 63, a playback area 64, and a generation area 65. For the meeting information area 60, the information input area 61, the playback area 64, and the generation area 65, refer to the meeting information area 50, the information input area 51, the playback area 54, and the generation area 55.

[0101] The timestamp area 62 is configured to be able to display time information regarding the elapsed time of the meeting when information is input into the information input area 61. That is, in response to receiving input of information such as characters into the information input area 51, the processor 21 records the time information of the time when the characters are input into the information input area 51 in the timestamp area 52. For example, assume that the processor 21 receives input of characters into the information input area 51 when 2 minutes and 3 seconds have elapsed since the start of the meeting. In this case, the processor 21 records time information such as "02:03" at a position horizontal to the position where the characters are input and within the timestamp area 52. Thereby, the time when the characters are input into the minutes can be recorded.

[0102] The conference text area 63 may be configured to obtain and display the character recognition information in real time from the voice information obtained in activity A3. The conference text area 63 is configured to be able to display one or more pieces of character recognition information that are separated at a predetermined timing and arranged in chronological order, which are the results of the speech during the conference being recognized as characters. This predetermined timing may be any timing as long as the result of character recognition can be grasped as the flow of conversation. For example, it may be the timing when there is a pause for breath, the timing when there is a gap, the timing when a sentence can be determined, such as being delimited by a period, or the timing when the speaker changes. As a result, since the character recognition result is delimited at each predetermined timing, the recording user can easily confirm the result.

[0103] The character recognition information is the information of the characters recognized (the content of the speech) based on the voice information of the conference. Also, for example, the character recognition information may be text in which the voice is digitized by acoustic analysis, phonemes are extracted from the voice by an acoustic model, the phonemes are converted into words, and then output as a sentence. The character recognition information may be associated with speaker information and time information. The speaker information is information that can identify the user who made the speech from which the character recognition is derived. The time information is the information of the time when the voice of the script associated with the script information was uttered during the conference. When the time information is transferred from the conference text area 63 to the information input area 61, it is displayed in the timestamp area 62.

[0104] (Activity A5) In parallel with activities A3 to A4, generated information may be generated during the conference. That is, the generation control unit 213 obtains the generated information generated by inputting the input information into the learned model. More specifically, for example, the generation control unit 213 obtains the generated information generated by inputting the input information into the learned model in a manner that the input information is included as part of the prompt. The display control unit 211 generates screen information that allows the user to visually recognize the generated generated information. Although the generated information is assumed to be displayed in the generation area 65, it is not limited thereto and may be displayed at any location. Note that activity A5 may be omitted.

[0105] By inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and sharing tool information as input information into a learned model, the generation control unit 213 may obtain meeting content information as generated information. As the speech-to-text information, information in which the voice during the meeting is transcribed in real time may be used. The meeting content information here includes at least one of information on the summary of the meeting, information on ToDo items after the meeting, information on decisions made at the meeting, information on key points of the meeting, information on the speeches of the speakers participating in the meeting, information on the names of the related parties of the meeting, information on Q&A sessions of the meeting, information on issues, information on cases under consideration, and information on requests. The meeting content information includes, for example, information summarizing the content of the speech or indicating key points during the ongoing meeting, and may include information indicating key points during the meeting such as "The important issues raised so far are A, B, and C".

[0106] By inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and sharing tool information as input information into a learned model, the generation control unit 213 may obtain reference information as generated information. The reference information is information that supports the conversations of the participants during the meeting, and may be, for example, information related to the content of the discussion, information indicating the summary or key points of past conversations, information that can be an answer to questions raised during the meeting, etc.

[0107] By inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, and speech-to-text information as input information into a learned model, the generation control unit 213 may obtain insight information as generated information.

[0108] By inputting at least one of recorded information, material information, shared tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may acquire discussion information as generated information. The discussion information is information indicating what should be agreed upon or discussed in a meeting. For example, from a statement during a meeting such as "By using Supplier A, the connection with Mr. B can be utilized, so it may be good for both Supplier A and Mr. B", the content of discussing "whether to use Supplier A or not" is generated. The discussion information may be information suggesting or proposing the priority order of a plurality of contents to be agreed upon or discussed. Also, in the example of FIG. 7, in the generation area 65, information indicating what should be discussed in the meeting as discussion information is displayed as an example.

[0109] By inputting at least one of recorded information, material information, shared tool information, past speech information, voice information, and speech-to-text information as input information into a learned model, the generation control unit 213 may acquire analysis information as generated information. The analysis information is information indicating the result of the analysis of each statement. The analysis information is, for example, information on whether statements of the same purport are repeated or / and information obtained by analyzing the amount of statements for each participant.

[0110] By inputting at least one of recorded information, material information, shared tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may acquire result information as generated information. The result information may be information indicating the approval or disapproval of the adoption of a service or product in business negotiations, the decision on promotion or transfer in a personnel interview, the approval or disapproval of passing the selection in an employment interview, etc.

[0111] For example, during a business negotiation, by inputting at least one of the recorded information, document information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may extract key keywords (such as "cost reduction", "short delivery time", "specific function requirements", etc.) from the customer's speech and generate hints as generation information during the conversation with the salesperson. For example, during a business negotiation, by inputting at least one of the recorded information, document information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate answer candidates as generation information from the in-house documents according to the customer's question.

[0112] For example, during an employment interview, by inputting at least one of the recorded information, document information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may textify the interview answers and generate information in which the relevance to the required skills (such as programming experience, management ability, etc.) is immediately evaluated and / or visualized as generation information. For example, during an employment interview, by inputting at least one of the recorded information, document information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate information in which the AI summarizes the key points of the answer as generation information.

[0113] For example, during a meeting, by inputting at least one of the recorded information, document information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate the summary of the discussion points, important decisions, concerns, additional tasks, etc. as generation information. For example, during a meeting, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate related past solutions, person-in-charge information, etc. as generation information when a specific issue resurfaces.

[0114] For example, during a phone call, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may detect a timing when customer satisfaction is likely to decrease through tone analysis of the customer's voice and generate a follow-up comment as generation information. For example, during a phone call, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, if the customer's problem relates to a specific product, the generation control unit 213 may generate the solution steps of past similar cases as generation information.

[0115] For example, during negotiations, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, when the other party requests a modification of the clause text, the generation control unit 213 may generate an alternative text proposal as generation information from similar past contracts and / or in-house standard terms. For example, during negotiations, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate an evaluation as generation information on whether the proposed conditions deviate from in-house guidelines or legal regulations.

[0116] For example, during an interview, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, real-time tagging or / and summarization of positive / negative feedback, improvement requests, etc. from the user's speech. For example, during an interview, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, when the user mentions an interesting point, the generation control unit 213 may generate, as generation information, information for automatically proposing additional questions to dig deeper into it.

[0117] For example, during brainstorming, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may organize in real time the symptoms described by the patient and generate, as generation information, differential diagnosis candidates, supplementary question candidates, etc. For example, during brainstorming, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, appropriate treatment methods and / or recommended items for the next examination.

[0118] For example, during a counseling session, by inputting at least one of recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, when explaining a concept that a student has difficulty with, the generation control unit 213 may generate, as generation information, supplementary examples, visual teaching materials, etc. For example, during an interview, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may analyze the student's reaction and generate another explanation strategy as generation information when the understanding level is low.

[0119] For example, during an interview, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may immediately summarize new information and unique perspectives from the expert's answers and generate text in a state that is easy to be written into an article after the meeting as generation information. For example, during an interview, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate relevant in-depth questions as generation information from the keywords and cases mentioned by the expert.

[0120] For example, during a strategic meeting with the client's management level, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, when the client asks questions about a specific strategic plan, the generation control unit 213 may generate the prepared simulation results and financial models as generation information. For example, during a strategic meeting with the client's management level, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, when the management level indicates concerns, the generation control unit 213 may generate information in which the AI lists similar company cases, alternative measures, etc. as generation information.

[0121] For example, during a customer onboarding session, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, a simple tutorial or text guide for screen sharing immediately for a function that the customer has difficulty understanding. For example, during a customer onboarding session, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, a list of successful cases of other companies, usage methods, etc. for the KPI that the customer aims for.

[0122] For example, during a quality improvement meeting with a supplier, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, information indicating an explanation and compliance confirmation for newly presented quality standards, international standards, etc. For example, during a quality improvement meeting with a supplier, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, performance data of other suppliers so that the negotiation does not get stuck.

[0123] For example, during a client meeting (event planning), by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, related service providers or production plans in response to statements regarding additional elements (booking of specific guests, specific concepts) desired by the client. For example, during a client meeting (event planning), by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may simulate the approximate cost on the spot in response to an additional request and generate an alternative plan as generation information when the budget is exceeded.

[0124] For example, during a stakeholder meeting, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into the learned model, when a stakeholder requests specific data, the generation control unit 213 may generate relevant statistics from a government database or the like as generation information. For example, during a stakeholder meeting, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into the learned model, when a conflict occurs, the generation control unit 213 may generate compromise possible policy options from past similar cases as generation information.

[0125] For example, during a customer asset management consultation, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into the learned model, when a customer asks "What if I add an additional investment of 1 million yen to this fund?", the generation control unit 213 may generate the expected return, risk indicators, etc. as generation information. For example, during a customer asset management consultation, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into the learned model, when a customer is concerned about risks, the generation control unit 213 may generate alternative products, savings investment plans, etc. that are estimated to have low risks as generation information.

[0126] For example, during a media briefing session, by inputting at least one of the recording information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, information (numerical data, exclusive scoops, quotable key phrases, etc.) requested by the reporter side. For example, during a media briefing session, by inputting at least one of the recording information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into a learned model, if a reporter asks an in-depth question, the generation control unit 213 may generate, as generation information, relevant press release excerpts, internal statistics, etc.

[0127] For example, during a user community symposium, by inputting at least one of the recording information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may analyze the statements in real time and generate, as generation information, opinions such as improvement requests, function requirements, dissatisfaction with the UI, and price-value evaluations organized by category. For example, during a user community symposium, by inputting at least one of the recording information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into a learned model, if something noteworthy comes up during the conversation, the generation control unit 213 may generate, as generation information, information that proposes guiding to a simple questionnaire (online form) on the spot.

[0128] For example, during customer beta test feedback, by inputting at least one of the recording information, material information, sharing tool information, past statement information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may immediately summarize the improvement requests and bug reports made during a real-time meeting with beta testers and generate, as generation information, information associated with relevant known bugs and development status. For example, during customer beta testing feedback, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, when a customer indicates dissatisfaction with the UI, the generation control unit 213 may generate a reference example of a successful UI pattern adopted by other products as generation information.

[0129] For example, during a stakeholder consensus formation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, information automatically sorted by consensus stage, such as "functions essential for this release" and "functions to be considered in the future", from the content of the speech. For example, during a stakeholder consensus formation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may immediately estimate the development period and additional costs when prioritizing a specific function, and generate options as generation information during the conversation.

[0130] For example, during a one-on-one interview with a team member, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate training materials and e-learning content as generation information according to the sales skill areas cited by the member as issues. For example, during a one-on-one interview with a team member, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate realistic KPI targets as generation information.

[0131] For example, during regular meetings with important customers, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may perform text analysis on the customer's speech and generate, as generation information, a new service guide, upsell, etc. with potential issues and desired areas of the customer tagged. For example, during regular meetings with important customers, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, a solution flow, FAQ, etc. generated from in-house knowledge for the points where the customer is troubled.

[0132] For example, during regular client reporting meetings, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, when the client doubts the effect of a specific measure, the generation control unit 213 may generate, as generation information, related datasets, statistical analysis results, graphs, etc. For example, during regular client reporting meetings, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, when unexpected questions, new issues, etc. arise, the generation control unit 213 may generate, as generation information, alternative approach plans, industry benchmarks, etc.

[0133] For example, during promotion / evaluation interviews, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, cases of employees within the company who have similar skill sets and have been promoted, and their subsequent career paths, etc. For example, during a promotion / evaluation interview, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate a training program, a mentor system, etc. tailored to the weaknesses of the person being evaluated as generation information.

[0134] For example, during an influencer meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may analyze the conditions (such as compensation, number of posts, posting period, etc.) presented by the influencer, immediately compare them with internal standards, and generate information indicating room for negotiation as generation information. For example, during an influencer meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate win-win content as generation information from successful cases favorable to the brand side and analysis of the influencer's past posts.

[0135] For example, during a debriefing after a usability test, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may, during the discussion of the test results, automatically generate information listing improvement measures and retest ideas from the members' statements as generation information. For example, during a debriefing after a usability test, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, points where the current UX has improved and / or deteriorated compared to other products or past versions.

[0136] For example, during the supplier quality review meeting, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, transcription information, and insight information as input information into the learned model, the generation control unit 213 may generate past similar cases or other supplier cases as generation information in response to the supplier's reply. For example, during the supplier quality review meeting, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, transcription information, and insight information as input information into the learned model, the generation control unit 213 may generate delivery date improvement measures, quality assurance contract conditions, etc. as generation information.

[0137] For example, during the risk assessment meeting, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, transcription information, and insight information as input information into the learned model, the generation control unit 213 may generate relevant internal regulations and past incident response records as generation information for the newly mentioned risk factors. For example, during the risk assessment meeting, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, transcription information, and insight information as input information into the learned model, the generation control unit 213 may compare risk mitigation measures in terms of cost, effectiveness, etc. and generate the best plan as generation information.

[0138] For example, during the creative review meeting with the advertising agency, by inputting at least one of the recorded information, material information, sharing tool information, past statement information, voice information, transcription information, and insight information as input information into the learned model, the generation control unit 213 may generate the consistency check result with the brand requirements for the new visual proposal presented by the agency as generation information. For example, during a creative review meeting with an advertising agency, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, information indicating a proposal for an optimal color palette or copy based on past creative A / B test results.

[0139] For example, during a regular meeting with an external legal advisor, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, countermeasures against a legal amendment proposal presented by a lawyer, past case information, etc. For example, during a regular meeting with an external legal advisor, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may automatically generate an amendment to the contract terms and generate, as generation information, information indicating verification of the approval or disapproval thereof.

[0140] For example, during a KPI review meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may analyze in real time the reasons for KPI fluctuations and generate, as generation information, correlations with data from other departments and outlier patterns. For example, during a KPI review meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may respond immediately to a data confirmation request from the attendees and generate, as generation information, detailed charts, tables, etc.

[0141] For example, during the regular code review meeting of the development team, by inputting at least one of the recorded information, material information, sharing tool information, past release information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, similar code implementation examples, improvement plans, and existing framework usage examples for the pointed-out locations. For example, during the regular code review meeting of the development team, by inputting at least one of the recorded information, material information, sharing tool information, past release information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, conflict resolution, optimization points, etc.

[0142] For example, during the meeting for preparing the financial results presentation for investors, by inputting at least one of the recorded information, material information, sharing tool information, past release information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, various financial indicators, graphs, etc., in response to additional comments from the management team and requests for competitive comparison. For example, during the meeting for preparing the financial results presentation for investors, by inputting at least one of the recorded information, material information, sharing tool information, past release information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, proposed solutions for common concerns for investors (such as a decrease in profit margin and an increase in inventory).

[0143] For example, during the operation improvement meeting, by inputting at least one of the recorded information, material information, sharing tool information, past release information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, information in which costs, effects, introduction performance data, etc. are compared for a new tool introduction plan. For example, during the operation improvement meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information into the learned model as input information, the generation control unit 213 may generate a load test scenario as generation information to facilitate the evaluation of performance improvement measures.

[0144] For example, during the feedback aggregation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information into the learned model as input information, the generation control unit 213 may generate, each time an improvement request is made, past improvement success cases and related customer data as generation information. For example, during the feedback aggregation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information into the learned model as input information, the generation control unit 213 may generate, when a product improvement measure or a UX revision plan is presented, information on the estimated execution cost and expected effect as generation information.

[0145] For example, during the follow-up interview for new employee training, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information into the learned model as input information, the generation control unit 213 may generate, in response to questions from the trainees, links to relevant in-house manuals, e-learning materials, etc. as generation information. For example, during the follow-up interview for new employee training, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information into the learned model as input information, the generation control unit 213 may generate a revised learning plan as generation information for areas with insufficient progress.

[0146] For example, during a lead nurturing strategy meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, a predicted conversion rate, an expected CTR improvement, etc. for a new campaign measure being considered by team members. For example, during a lead nurturing strategy meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, a content proposal tailored to the lead's interest topics (specific product fields, price ranges).

[0147] For example, during a technology roadmap formulation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, a reference framework, tools, etc. for solving the technical issues presented during the discussion. For example, during a technology roadmap formulation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may generate, as generation information, information obtained by simulating the development period, cost increase or decrease, etc. when switching to another technology option.

[0148] For example, during an M&A negotiation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into a learned model, the generation control unit 213 may immediately execute a simulation for the conditions presented by the other party and generate, as generation information, the post-acquisition cash flow, ROI, etc. For example, during an M&A negotiation meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may refer to the conditions of past similar transactions and industry standard terms, and generate information indicating the evaluation of validity as generation information.

[0149] For example, during a compliance rectification meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, related legal explanations, internal regulations, rectification cases, etc. for the pointed-out matters. For example, during a compliance rectification meeting, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, information that can compare the reduction effects such as man-hours, costs, and risks for each countermeasure plan.

[0150] For example, during a sponsor consultation, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, a booth layout plan and a privilege menu in accordance with the sponsor's requests. For example, during a sponsor consultation, by inputting at least one of the recorded information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, information that can compare the sponsor package price and comparison with other sponsors (such as exposure level) in order to smoothly advance the negotiation.

[0151] For example, during the launch sharing meeting, by inputting at least one of the recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, the degree of KPI achievement, successful cases of appeal points, etc. during the launch of past similar products, for the proposed launch campaign. For example, during the launch sharing meeting, by inputting at least one of the recording information, material information, sharing tool information, past speech information, voice information, speech-to-text information, and insight information as input information into the learned model, the generation control unit 213 may generate, as generation information, a real-time trial calculation of the price and / or promotion strategy and the possibility of obtaining the target market share.

[0152] (Activity A6) Subsequently, during the execution of A3 to A5, the information transmission / reception unit 210 of the server device 2 receives an instruction to end the meeting, such as the pressing of a meeting end button from the client device 3 of the recording user. The recording unit 214 ends the counting of the elapsed time of the meeting and also ends the recording.

[0153] (Activity A7) Subsequently, as shown in FIG. 8, information processing related to speech recognition is executed. FIG. 8 is a diagram for explaining the information processing related to speech recognition. As shown in FIG. 8, the generation control unit 213 acquires conference text information indicating text generated based on the voice information and the term information. More specifically, the acquisition unit 212 receives term information in addition to an instruction to generate conference text information via an operation by the user on the input unit 35. First, the generation control unit 213 acquires speech recognition information from the voice information. The speech recognition information is information indicating text obtained by performing speech recognition on the voice of the voice information. That is, the speech recognition information may be interpreted as information indicating text in a state before the correction process based on terms acquired from the term information is performed on the text obtained by performing speech recognition on the voice of the voice information. Further, the speech recognition information may be interpreted as information indicating text in a state before fluctuations in notation or fillers are eliminated, which is the text obtained by performing speech recognition on the voice of the voice information. Instead of the voice information, video information may be acquired and the voice included in the video information may be used to perform a similar process.

[0154] Also, it is not necessary to perform real-time speech recognition by acquiring voice information during a meeting, and a recorded file may be uploaded. That is, the acquisition unit 212 may acquire voice information by receiving an upload of voice information including the voice of the meeting from the user.

[0155] In addition, the acquisition unit 212 acquires term information from which terms related to the meeting can be acquired. For example, the acquisition unit 212 may acquire term information by acquiring a list including terms uploaded by the user or input by the user via the input unit 35. For example, the acquisition unit 212 extracts terms by inputting the term information into a learned model. The term information is information including text information, and may include, for example, any noun (e.g., personal name, place name, product name, in-company term, technical term, industry term, etc.), verb or adjective (e.g., run, search, beautiful, dangerous, etc.). Further, the term information may include, as text, terms not listed in any dictionary (such as the goo Japanese dictionary, Weblio dictionary, Digital Daijisen, etc.). Terms not listed in any dictionary are assumed to be coined words, in-company terms, industry terms, personal names, etc. The term information may include data in the state of an image from which text information can be acquired by OCR (Optical Character Recognition / Reader). The term information may be text directly input by the user into the minutes creation tool. The term information may be a document indicating the minutes of a meeting different from the current meeting (minutes of past meetings), and may also be acquired from past minutes information. The term information as the minutes includes at least one of recording information, transcription information, and meeting text information. The term information may be material information. The term information may be in any data format, and may be in formats such as txt, pdf, docs, rtf, odt, jpg, png, gif, tif, bmp, svg, xlsx, csv, ods, pptx, odp, etc. Further, the file may be a compressed file including files in the above-mentioned data formats, and in that case, the data format may be zip, rar, tar, gz, bz2, 7z, lzma, xz, etc. A plurality of term information files may be specified. The memory control unit 216 may store the analyzed terms in the memory unit 22 or the memory unit 32. The display control unit 211 may display a screen on which the stored terms are visible in response to an instruction via an operation on the input unit 35 by the user.

[0156] The generation control unit 213 obtains conference text information by inputting voice information, the text included in the transcription information, and the terms extracted from the term information into a learned model. For example, the generation control unit 213 obtains conference text information based on the terms extracted from the term information and the voice indicated by the voice information for the text included in the transcription information. More specifically, for example, when the text "Tanaka Koki" is included in the text included in the transcription information, and the voice information and the term information including the term "Tanaka Mitsuki" are input into the learned model, the part of "Tanaka Koki" is converted to "Tanaka Mitsuki", and the conference text information (refer to the conference text areas 63 and 73 in FIGS. 7 and 9) is output. Also, the learned model may be configured to be able to extract terms with reference to the number of appearances of each term from the text included in the term information. The learned model may be configured to newly extract terms without using the results of past term extractions. Further, the generation control unit 213 obtains conference text information as text in a state where the orthographic variations between words included in the text included in the transcription information are resolved by inputting the transcription information into the learned model. Orthographic variations are presumed to have the same meaning but occur due to some characters of the terms being different. For example, "minutes creation tool" and "minutes system" (refer to the conference text areas 63 and 73 in FIGS. 7 and 9), "patent application" and "patent filing", etc. may be applicable. Also, the generation control unit 213 obtains conference text information as text in a state where the text corresponding to the fillers included in the text included in the transcription information is removed by inputting the transcription information into the learned model. Fillers indicate words that have no meaning such as "um" (refer to the conference text areas 63 and 73 in FIGS. 7 and 9), "uh", "ah", "well", "um", "well", "um", etc. From another perspective, fillers may include interjections. Fillers may include the concept of keba. Further, the generation control unit 213 may output conference text information in a state where the spoken language included in the text of the transcription information is converted into written language by inputting the transcription information into the learned model.The process of converting spoken language into written language includes processes such as converting "nanode / node / dakara / desukara" into "sonotame / tame", converting "yappari" into "yabari", etc. When the conference text information is acquired, the display control unit 211 generates screen information that can display the conference text information shown in FIG. 9 in the conference text area 73. As a result, since the text of the conference in a mode replaced with appropriate terms can be acquired, it is possible to support the creation of a more accurate minutes from the viewpoint of converting the conference audio into text.

[0157] Next, the minutes screen 7 after the conference will be described with reference to FIG. 9. FIG. 9 is a diagram showing an example of the minutes screen 7 after the conference in the embodiment. The minutes screen 7 is a screen for creating the minutes after the conference. The minutes screen 7 includes a conference information area 70, an information input area 71, a timestamp area 72, a conference text area 73, a playback area 74, and a generation area 75. For the conference information area 70, the information input area 71, the timestamp area 72, the conference text area 73, the playback area 74, and the generation area 75, refer to the conference information area 60, the information input area 61, the timestamp area 62, the conference text area 63, the playback area 64, and the generation area 65.

[0158] The time stamp area 72 after the conference displays time information recorded during the conference. Also, the time stamp area 72 after the conference displays time associated with time information of the conference text information transcribed from the conference text area 73. In addition to the above-mentioned functions, the time stamp area 72 is configured to be able to display time information related to the elapsed time of the conference from the start of the speech of the related conference text information when the conference content information is output to the information input area 71. Also, the time stamp area 72 is configured to be able to display time information related to the elapsed time of the conference input to the related information input area 71 when the conference content information is output to the information input area 71. When the information is input to the information input area 71 in bullet points, the processor 21 gives a time stamp (such as "02:14" and "04:48" in the example of FIG. 9) to each item in the bullet points. In response to the selection of the time stamp, the display control unit 211 moves the position of the seek bar to the elapsed time of the conference associated with the time stamp. Also, the time may be displayed on the seek bar in the playback area 74. For example, if 30 seconds have elapsed since the start of a meeting and the background to the inquiry is something like, "I heard about it from an acquaintance and it caught my attention so I searched for it," processor 21 may link and display the words "Developing new customers" at the position corresponding to 30 seconds on the seek bar of playback area 74.

[0159] The conference text area 73 displays conference text information or transcription information. When both conference text information and transcription information are acquired, the display of either information may be switchable. The conference text area 73 in FIG. 9 displays conference text information. The conference text information is information on the text of the conference generated based on the transcription information and the audio information. The conference text information may be text output by inputting the transcription information to the generation AI. Comparing the conference text information and the transcription information, the conference text information more accurately reflects the contents of the conference. The conference text information may also be linked to speaker information and time information. The time information is displayed in the time stamp area 72 when the text is transcribed from the conference text area 73 to the information input area 71.

[0160] When the user clicks and selects the time information in the time stamp area 72, the playback unit 215 plays back the audio of the meeting from the time associated with the time information. The playback of the audio of the meeting may end when all the playback of the audio of the part associated with the meeting text information in the meeting text area 73 has been completed, or all the audio after the time associated with the time information may be continuously played back. Thereby, the user can easily play back the audio of the time related to the meeting text information.

[0161] (Activity A8) Subsequently, the processor 31 of the client device 3 receives a predetermined operation via the operation of the input unit 35 by the user. When the processor 31 receives a predetermined operation, it transmits information specifying the predetermined operation to the server device 2 via the communication unit 33 and the network NW. The processor 21 of the server device 2 receives information specifying the predetermined operation from the client device 3 via the network NW and the communication unit 23.

[0162] (Activity A9) For example, when the predetermined operation received in Activity A8 is an operation including an instruction to acquire generation information, the processor 21 of the server device 2 proceeds with information processing in Activity A10. Also, for example, when the predetermined operation received in Activity A8 is an operation including an instruction to complete the minutes of the meeting, the processor 21 proceeds with information processing in Activity A11.

[0163] (Activity A10) When the predetermined operation received in Activity A8 includes an instruction to acquire generated information, the generation control unit 213 acquires the generated information generated by inputting the input information into the learned model in response to determining that the meeting has ended. More specifically, for example, the generation control unit 213 acquires the generated information generated by inputting the input information into the learned model in a manner that the input information is included as part of the prompt. The display control unit 211 generates screen information that enables the user to visually recognize the generated generated information. The generated information may be meeting content information. Note that the displayed screen is assumed to be, for example, the minutes screen 7 described above in FIG. 9, the mail screen 8 described below in FIG. 10, a screen on a sharing tool (not shown), etc., but is not limited to these screens.

[0164] By inputting at least one of past speech information, speech-to-text information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model as input information, the generation control unit 213 may acquire meeting content information as the generated information. The meeting content information here includes at least one of information on the summary of the meeting, information on ToDo items after the meeting, information on the decisions made at the meeting, information on the key points of the meeting, information on the speeches of the speakers participating in the meeting, information on the names of the persons related to the meeting, information on the Q&A of the meeting, information on issues, information on the cases under consideration, and information on requests in the ongoing meeting.

[0165] By inputting at least one of past speech information, speech-to-text information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model as input information, the generation control unit 213 may acquire a message used on an arbitrary communication tool as the generated information. As the message information, for example, a mail such as the mail screen 8 shown in FIG. 10 is generated. FIG. 10 is an example of the mail screen 8 generated by the learned model. The mail screen 8 includes a destination area 80, a subject area 81, an attachment area 82, and a body area 83.

[0166] For example, the generation control unit 213 may generate a message on any tool. The tool may be, for example, a communication tool, a schedule adjustment tool, etc. The tool may be arbitrarily selected by the user for each meeting, or the tool may be used when there is a reference to the tool to be used during the meeting, or a pre-defined tool may be used. In the example of FIG. 10, a mailer is used as the tool.

[0167] For example, the generation control unit 213 may specify the destination to which the message is to be sent. For example, when the message is a notice of the next meeting, the generation control unit 213 identifies, from the content of the meeting, the participants of the previous meeting, the participants of the current meeting, the persons who appeared in the speech of the meeting, etc., and from these contents, identifies the participants of the next meeting. For example, the generation control unit 213 acquires information such as the email addresses, phone numbers, names, etc. of the identified participants, which is necessary for issuing the meeting notice to the participants of the next meeting. In the example of FIG. 10, the email address identified by the generation control unit 213 is input as the destination to which the message is to be sent in the destination area 80.

[0168] For example, when the generation control unit 213 uses a tool that allows input of a subject such as a mailer, it may generate a subject. The subject may be arbitrarily composed of, for example, a case number (which may include the concept of an arrangement number), a summary of the meeting, a request, a next action, etc. For example, the generation control unit 213 may obtain "P001" as the case number and "Notice of the next meeting" as the request, and generate a subject of "Notice of the next meeting / P001". In the example of FIG. 10, the subject generated by the generation control unit 213 is input in the subject area 81.

[0169] For example, the generation control unit 213 identifies a file related to a meeting based on input information including at least one of the speech recognition information, meeting text information, audio information, video information, recording information, material information, and sharing tool information. Further, the generation control unit 213 may identify a file used for screen sharing or the like in the meeting. The generation control unit 213 may attach the identified file to the message, or may embed a link that can access the identified file in the message. Also, for example, the generation control unit 213 identifies the topic, purpose, etc. of the next meeting from the input information including at least one of the speech recognition information, meeting text information, audio information, video information, recording information, material information, and sharing tool information. Subsequently, the generation control unit 213 may obtain information required by the meeting body from the input information including at least one of the speech recognition information, meeting text information, audio information, video information, recording information, material information, and sharing tool information, and generate materials that are assumed to be used in the next meeting. For example, the generation control unit 213 may obtain materials or text from a sharing tool (for example, a file storage tool, a communication tool, etc.), or generate next meeting material information planned to be used in the next meeting on the Internet based on the information included in the materials or the text. The materials to be output may be files with extensions such as documents (doc, docx, odt, pdf, rtf, txt), presentations (ppt, pptx, odp, pps, ppsx, key), spreadsheets (xls, xlsx, csv, ods), images (jpg, jpeg, png, gif, bmp, tiff, svg), videos (mp4, avi, mov, wmv, flv), audio (mp3, wav, aac, flac), compressed files (zip, rar, 7z, tar, gz), and others (exe, dll, iso, html, htm, json, xml). The generation control unit 213 may determine the design of the materials according to a template set by the user or the default template. Also, the generation control unit 213 generates materials including statistical data, figures, graphs, etc. as required. The next meeting material information is an example of the generated information. In the example of FIG. 10, the generated document is attached to the attached document area 82 by the generation control unit 213.

[0170] For example, the generation control unit 213 generates a message based on input information including at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and sharing tool information. The message may include, for example, meeting content information. The generation control unit 213 outputs the message in a draft state to the message input field on the tool. In the example of FIG. 10, the message generated by the generation control unit 213 is output in a draft state in the text area 83. The draft is an example of the generated information. In the example of FIG. 10, the message is about the notice of the next meeting, but it may also be about the summary or gratitude of the meeting. The message may be configured such that the portions estimated by the learned model where the customer has shown particular interest are emphasized.

[0171] As described above, the message information is assumed to be used in any communication, and in addition to the example of the mailer shown in FIG. 10, information that can be used in LINE, Chatwork, Slack, WeChat, etc. may be generated.

[0172] By inputting at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and sharing tool information as input information into the learned model, the generation control unit 213 may obtain information indicating the notice of the meeting as the generated information. For example, the generation control unit 213 specifies the date and time of the next meeting from the input information. For example, when the generation control unit 213 obtains the content to the effect that "at the meeting, the next meeting will be held from 15:30 on July 21, 2025 (Monday)", it generates a notice of the meeting starting at 15:30 on July 21, 2025 (Monday). Various methods of meeting notice will be described below. More specifically, for example, the generation control unit 213 may output a message in a draft state to the message input field on the tool, as shown in FIG. 10, to notify of a meeting starting at 15:30 on July 21, 2025 (Monday). As another example, more specifically, for example, the generation control unit 213 may set the meeting time based on the content and participants of the next meeting. First, for example, the generation control unit 213 identifies, as the participants of the next meeting, the people who appear in the input information, the people assigned to the relevant projects in the sharing tool, and the like. The generation control unit 213 identifies the schedules of the identified participants of the next meeting from the schedule adjustment tool. The generation control unit 213 identifies a plurality of candidates from the available dates and times among the schedules of the participants of the next meeting. Here, the generation control unit 213 may identify the candidates excluding Saturdays, Sundays, and holidays. The generation control unit 213 may identify the candidates excluding the time before the start of work and after the end of work. As another example, more specifically, for example, when the generation control unit 213 cannot refer to the schedules of the identified participants of the next meeting, it selects several dates and times through the schedule adjustment tool among the sharing tools. Then, the generation control unit 213 may give an instruction to the schedule adjustment tool to present the identified date and time candidates to the participants of the next meeting.

[0173] The generation control unit 213 identifies, as input information, at least one of the character recognition information, meeting text information, voice information, video information, recording information, material information, and sharing tool information, and identifies the tasks to be worked on after the meeting and the responsible persons for the tasks. For example, the generation control unit 213 may perform a process of setting and saving the tasks to be worked on and the responsible persons for the tasks in a sharing tool (such as a task management tool).

[0174] The generation control unit 213 acquires information indicating improvement measures generated based on at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and shared tool information. For example, the generation control unit 213 may generate information such as "Since the price negotiation took a long time in this conversation, it would be good to prepare a price comparison sheet by the next time" as an improvement measure. The improvement measures may be generated based on various perspectives (satisfaction level, agreement level, remaining issue level, etc.) of the conversation content.

[0175] For example, after a business negotiation, by inputting input information including at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and shared tool information into a learned model, the generation control unit 213 may generate, as generation information, a meeting minutes that organizes the points of agreement, remaining issues, next actions, etc. based on the voice recording of the business negotiation. For example, after a business negotiation, by inputting input information including at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and shared tool information into a learned model, the generation control unit 213 may generate, as generation information, a customized thank-you email text corresponding to the business negotiation content and attach related white papers and case studies.

[0176] For example, after an employment interview, by inputting input information including at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and shared tool information into a learned model, the generation control unit 213 may generate, as generation information, an evaluation report summarizing the strengths, concerns, cultural fit, etc. of the candidate. For example, after an employment interview, by inputting input information including at least one of the speech-to-text information, meeting text information, voice information, video information, recording information, document information, and shared tool information into a learned model, the generation control unit 213 may generate, as generation information, an email text that describes gratitude, next steps, etc. based on an appropriate template.

[0177] For example, after a meeting, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may organize who did what by when from the minutes of the meeting and register it in a sharing tool such as a project management tool. For example, after a meeting, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, an email to all relevant parties including a summary of the meeting content, a task list, etc.

[0178] For example, after a phone call, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may automatically document the procedures and solutions carried out during the call and generate, as generation information, an email to the customer. For example, after a phone call, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, a draft of a guidance email regarding product updates, additional support, etc. that the customer is likely to need in the near future.

[0179] For example, after a negotiation, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may organize the negotiation content, reflect the parts that need to be revised in the contract draft again, and then generate a new version as generation information. For example, after negotiation, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, information that can adjust the next review schedule and / or a draft report email to legal, management, etc. In the following, although the adjustment of the schedule and the registration in the calendar are described as being performed via the schedule adjustment tool, it is not limited thereto, and it may be performed via other sharing tools.

[0180] For example, after an interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, a report that organizes user needs, improvement points, proposed ideas, etc. based on the interview results. For example, after an interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, information that automatically tasks the collected feedback to the product development team and prompts its reflection in the improvement sprint.

[0181] For example, after a brainstorming session, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, a concise medical record and patient explanation materials (such as life improvement advice, medication instructions, etc.). For example, after a brainstorming session, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, a reminder email for the patient by automatically scheduling the next examination schedule and rehabilitation reservation.

[0182] For example, after an interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, based on the interview result, information obtained by automatically optimizing (such as setting additional question sets, online video links, etc.) the curriculum for students as generated information. For example, after an interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as needed, a contact email that briefly summarizes the learning situation and support proposals for the guardians as generated information.

[0183] For example, after an interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, based on the interview record, a draft article text that summarizes the key points as generated information. For example, after an interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generated information, reliable sources for the data, statistical values, etc. mentioned by the experts.

[0184] For example, after a strategic meeting with the client's management level, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may organize the decisions in chronological order and generate, as generated information, the assignment of responsible persons, the execution schedule, etc. For example, after a strategic meeting with the client management layer, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may set a regular checkpoint in the execution phase and generate a schedule adjustment email as generation information.

[0185] For example, after a customer onboarding session, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate a summary that combines the functions to be acquired, setting tasks, training video links, etc. as generation information. For example, after a customer onboarding session, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, information on the adjustability of the next follow-up schedule and / or information obtained by scoring the customer's proficiency and satisfaction for reference in the upsell strategy.

[0186] For example, after a quality improvement meeting with a supplier, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, information documenting the agreed improvement measures, schedule, quality inspection protocol, etc. For example, after a quality improvement meeting with a supplier, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may propose the schedule for the next quality check meeting and generate generation information as information for automatically sending calendar invitations to the supplier and the in-house responsible persons.

[0187] For example, after a client meeting (event planning), by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate the latest event plan document as generation information based on the agreed matters. Further, the generation control unit 213 may automatically send the generation information to the client. For example, after a client meeting (event planning), by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may list the necessary arrangements (such as catering, acoustics, venue decoration, etc.) and generate information that can be registered in a sharing tool such as a project management tool as generation information.

[0188] For example, after a stakeholder meeting, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate an official meeting record that organizes the opinions, agreed matters, and matters to be considered at the next meeting as generation information. For example, after a stakeholder meeting, by inputting input information including at least one of the transcription information, meeting text information, audio information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may, based on the meeting results, update the policy proposal document and report, and generate a draft for submission to a superior organization as generation information.

[0189] For example, after a customer asset management consultation, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a report summarizing the key points of the selected investment plan as generation information. Further, the generation control unit 213 may automatically send the generation information to the client. For example, after a customer asset management consultation, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may schedule the regular check timing and generate a follow-up email notice to the client as generation information. Further, the generation control unit 213 may automatically send the generation information to the client.

[0190] For example, after a media briefing meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a press release or a draft of a follow-up email based on the meeting content as generation information. For example, after a media briefing meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a keyword list for tracking future reports as generation information.

[0191] For example, after a user community symposium, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may conduct a cross-sectional analysis of multiple symposiums and generate a summary report summarizing customer insights from quantitative and / or qualitative perspectives as generation information. For example, after a user community symposium, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may automatically reflect the extracted needs in the product development roadmap and generate information for cooperation with the sharing tool as the generated information.

[0192] For example, after customer beta test feedback, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate information for registering new improvements and bug fixes in the Issue tracking tool as the generated information in order to facilitate reflection in the development sprint. For example, after customer beta test feedback, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate an update note for the next release that organizes the items to be improved as the generated information.

[0193] For example, after a stakeholder consensus formation meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate a roadmap reflected in a sharing tool such as a project management tool as the generated information based on the consensus content. For example, after a stakeholder consensus formation meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate emails, Slack notifications, etc. that summarize the consensus points and task assignments to the relevant parties as the generated information.

[0194] For example, after a one-on-one interview with a team member, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a one-on-one meeting record summarizing the goals, improvement measures, and next checkpoint as generation information. The generation control unit 213 may notify the person himself / herself, the supervisor, etc. via an arbitrary sharing tool. For example, after a one-on-one interview with a team member, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate information for registering the set target deadline and follow-up interview date in the calendar as generation information.

[0195] For example, after a regular meeting with an important customer, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may form an action plan including the agreed-upon improvement measures and generate a customer report as generation information. For example, after a regular meeting with an important customer, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may re-evaluate the user score (NPS) and utilization rate and generate the next improvement policy as generation information.

[0196] For example, after a regular client report meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, a summary of the countermeasures agreed upon in the report meeting and follow-up survey items in chronological order. The generation control unit 213 sends the generated information to the client, the in-house team, etc. For example, after a regular client meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may reflect improvement measures to be tried by the next time in a sharing tool such as a project management tool and generate information for automatically assigning responsibilities as generated information.

[0197] For example, after a promotion / evaluation interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate information for inputting the interview results and decision items (such as promotion approval, salary revision, future goals, etc.) into the evaluation management system as generated information. For example, after a promotion / evaluation interview, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a summary email of feedback, improvement suggestions, etc. to the evaluated person as generated information.

[0198] For example, after an influencer meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a draft of the contract as generated information based on the agreed content. The generation control unit 213 may send it to the legal department via an arbitrary sharing tool. For example, after an influencer meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may automatically reflect the agreed posting plan in the calendar of the SNS posting tool and generate information for setting reminders as generated information.

[0199] For example, after a debriefing following a usability test, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, a design requirements document including agreed UI modifications and user feedback countermeasures. For example, after a debriefing following a usability test, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, the next sprint with improvement tasks added to the triage list of designers and engineers.

[0200] For example, after a supplier quality review meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, an improvement plan document that organizes the agreed matters. Further, the generation control unit 213 may generate, as generation information, information that can be linked to an ERP tool for assignee assignment and schedule. For example, after a supplier quality review meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information that can notify the candidate schedule for the next review to the supplier, the in-house responsible person, etc.

[0201] For example, after a risk assessment meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information that can document the defined action items and register them in the risk management system. For example, after a risk assessment meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information capable of notifying the schedule of the next regular review and the preparation of necessary materials.

[0202] For example, after a creative review meeting with an advertising agency, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may summarize the finally agreed-upon creative requirements and generate, as generation information, the next revision request email, shared documents, etc. For example, after a creative review meeting with an advertising agency, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information capable of registering the schedule of the next progress confirmation meeting in the calendars of relevant parties.

[0203] For example, after a regular meeting with an external legal advisor, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, an implementation plan for the agreed-upon countermeasures. The generation control unit 213 may also generate, as generation information, information indicating the task assignment to relevant departments. For example, after a regular meeting with an external legal advisor, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, the schedule of the next legal audit and a draft follow-up letter.

[0204] For example, after a KPI review meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a report such as an agreed improvement measure and strategic direction as generation information. The generation control unit 213 transmits the generation information to management, relevant teams, etc. via a sharing tool. For example, after a KPI review meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may task the improvement measures and generate, as generation information, information that can be registered in the schedule for progress confirmation at the next meeting.

[0205] For example, after a regular development team code review meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information for automatically allocating the agreed correction instructions to tasks for each developer. For example, after a regular development team code review meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, a document summarizing the revised guidelines and updated coding convention points.

[0206] For example, after a meeting for preparing an earnings presentation for investors, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, the final version of the script and slides. The generation control unit 213 sends the generation information to the IR team, management, etc. via a sharing tool. For example, after an earnings presentation preparation meeting for investors, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may determine the Q&A response scenario for the day of the presentation and generate information that can register reminders as generated information.

[0207] For example, after an operation improvement meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate information that can automatically reflect the agreed improvement measures, setting changes, and hardware enhancement plans to the configuration management tool as generated information. For example, after an operation improvement meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate information that can automatically notify relevant parties of the schedule for the next test implementation date and evaluation meeting as generated information.

[0208] For example, after a feedback aggregation meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate a CX improvement plan document summarizing the agreed matters (UI improvement, support enhancement, price strategy review) as generated information. For example, after a feedback aggregation meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into a learned model, the generation control unit 213 may generate information that can link the task assignment and deadline setting by team and notify the schedule for the next follow-up meeting as generated information.

[0209] For example, after the follow-up interview for new employee training, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate a training plan document including the agreed improvement measures as generation information. The generation control unit 213 may transmit the generation information to the person himself / herself, the supervisor, etc. via the sharing tool. For example, after the follow-up interview for new employee training, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may register additional tasks and supplementary teaching materials, and may generate, as generation information, information that can automatically register the next interview schedule in the calendar.

[0210] For example, after the leadership strategy meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may automatically set the agreed campaign content in the marketing tool, and may generate, as generation information, information that can register the email sending schedule. For example, after the leadership strategy meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate, as generation information, information that can share the candidate schedule for the next KPI evaluation meeting.

[0211] For example, after the technology roadmap formulation meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may refine the agreed technology roadmap and generate, as generation information, information that can be reflected in a sharing tool such as a project management tool. For example, after a technology roadmap planning meeting, by inputting input information including at least one of transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information capable of automatically scheduling a review meeting when the next milestone is achieved.

[0212] For example, after an M&A negotiation meeting, by inputting input information including at least one of transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may summarize the agreed-upon content and generate, as generation information, an LOI (Letter of Intent) or a Term Sheet draft. For example, after an M&A negotiation meeting, by inputting input information including at least one of transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information capable of automatically registering the next due diligence schedule and legal checkpoints in the schedule.

[0213] For example, after a compliance rectification meeting, by inputting input information including at least one of transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may document the agreed-upon rectification plan and generate, as generation information, information capable of automatically executing task assignments and deadline notifications to each department. For example, after a compliance rectification meeting, by inputting input information including at least one of transcription information, meeting text information, voice information, video information, recording information, document information, and sharing tool information into a learned model, the generation control unit 213 may generate, as generation information, information capable of automatically registering the next follow-up audit schedule.

[0214] For example, after a sponsor meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate a contract draft including the agreed conditions as generation information. The generation control unit 213 may send the generation information to the sponsor or the like via the sharing tool. For example, after a sponsor meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate information that can be registered in the calendar for the next progress confirmation meeting or the material submission deadline as generation information.

[0215] For example, after a launch sharing meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate the agreed marketing plan and creative brief as generation information. The generation control unit 213 sends the generation information to the advertising agency, sales team, etc. via the sharing tool. For example, after a launch sharing meeting, by inputting input information including at least one of the transcription information, meeting text information, voice information, video information, recording information, material information, and sharing tool information into the learned model, the generation control unit 213 may generate information that can be registered in the task management tool for the kick-off schedule and review cycle as generation information.

[0216] (Activity A11) When the predetermined operation received at A8 includes an operation for completing the minutes, the storage control unit 216 stores the created minutes in the storage unit 22.

[0217] After that, the information processing system 1 terminates the information processing.

[0218] According to the present disclosure, in various tools, information can be output more appropriately or convenience can be improved more effectively.

[0219] [Others] The program is a program that causes one or more computers to execute each of the following steps. The information processing system 1 includes one or more computers that execute the program. Regarding the information processing system 1 according to the above-described embodiment, it may be a program that causes a computer to function as the processor 21 of the information processing system 1. It may also be an information processing method executed by the information processing system 1.

[0220] The server device 2 may be in an on-premises form or a cloud form. As the cloud-form server device 2, for example, in the form of SaaS (Software as a Service), cloud computing, etc., the above functions and processes may be provided.

[0221] In the above embodiment, the server device 2 performs various storage and controls. However, instead of the server device 2, a plurality of external devices may be used. That is, various information and programs may be distributed and stored in a plurality of external devices using blockchain technology or the like.

[0222] At least one of the devices included in the information processing system 1 may be installed outside Japan. For example, the server device 2 may be installed outside Japan, and each terminal device may be installed within Japan. Similarly, a user may access the server device 2 installed within Japan using his or her terminal device from outside Japan. According to such an aspect, a more convenient experience can be provided to the user in various management forms.

[0223] Note that the position where the generated information is output is not limited to the generation areas 55, 65, 75, and may be any position such as the information input areas 51, 61, 71, etc., or may be displayed on a page different from the minutes screens 5, 6, 7.

[0224] In a modified example, the minutes screen 5 before the meeting and the minutes screen 6 during the meeting may not be displayed, and only the minutes screen 7 after the meeting may be displayed. Also, for example, the minutes screen 9 shown in FIG. 11 may be displayed after the meeting. FIG. 11 is a diagram showing an example of the minutes screen 9 in the modified example. The minutes screen 9 is a screen for creating the minutes before the meeting. The minutes screen 9 includes a meeting text area 90, a generation area 91, and an information input area 92. Meeting text information or speech-to-text information may be displayed in the meeting text area 90. At least one of the meeting text information, the speech-to-text information, and the generation information may be displayed in the generation area 91. The information displayed in the meeting text area 90 or the generation area 91 may be transcribed in the information input area 92, or recording information may be input by the user.

[0225] Also, the learned model of the embodiment may generate any of the following generation information in response to the input of any input information (which may include at least one of speech-to-text information, meeting text information, voice information, video information, recording information, material information, and sharing tool information). The learned model may generate the generation information at at least one timing among before the meeting, during the meeting, and after the meeting. · Information for customer data analysis, analyzing customer attribute data, website visit history, email opening rate, and past purchase history, and scoring potential prospects · Information for a personalized approach, capable of automatically selecting and sending content (white papers, blog articles, product comparison tables) that matches the interests and concerns of prospects · Information for predicting customer behavior, predicting the possibility that a potential customer will take some action (inquiry, purchase), and sending targeted advertisements or reminder emails just before that · Information for handling routine inquiries, operating a chatbot that automatically answers frequently asked questions such as basic product information, price, inventory status, and delivery date · Information for proposing next improvement measures, when a salesperson proposes to a customer, the AI shows the optimal discount rate, optional products, and campaigns from past successful patterns · Information indicating effective ad copy, images, and video concepts for each target for ad creative generation · Information that can predict assumed CVR, CTR, etc. before a campaign starts and enable optimal budget allocation and channel selection for campaign performance prediction · Information indicating attractive recruitment terms by referring to similar job types and market data for automatic job description generation · Information indicating keyword candidates that are likely to be prominently displayed on search engines and job sites for keyword optimization · Information that is scored by analyzing applicants' resumes and work histories to extract important skills and / or experiences for resume screening · Information indicating an in-depth question list according to the position and candidate's history for interview question proposal · Information indicating employees with a high risk of leaving the company from employee satisfaction questionnaires and performance data for turnover tendency analysis · Information indicating a skill improvement plan and promotion model tailored to each employee for career path proposal · Information that can automatically respond to common questions and support processes with a chatbot for FAQ bots · Information that can automatically identify difficult inquiries and allocate them to appropriate personnel for escalation determination · Information indicating points where customers feel dissatisfied from past inquiry data for inquiry log analysis · Information that determines sentiment from customer email and chat logs for sentiment analysis and shows preventive measures for customers with signs of accumulating dissatisfaction · Automatically summarize and classify support documents for automatic document organization and present them in a form that is easy for agents to refer to. · Information indicating learning content that summarizes frequently asked questions and best practices for new operators for training content creation · Information that can extract amount, payment due date, and supplier information from PDF and scan data and automatically input it into the ERP system for invoice analysis · Information indicating the detection results of unusually high expense settlements or unnatural spending patterns for fraud detection · Information indicating the cash flow forecast for several months ahead, taking into account sales forecasts, spending patterns, exchange rate fluctuations, etc. for cash flow prediction · Information indicating cost reduction potential, based on business processes and supplier contract information, for cost reduction proposals · Information where the risks and / or returns of investment projects are simulated based on market data and industry trends for scenario analysis · Information indicating the optimal asset allocation from the assets held, target return, and risk tolerance for portfolio optimization · Information indicating payment terms, contract periods, cancellation terms, etc. from non-disclosure agreements, sales contracts, service contracts, etc. for important clause extraction · Information indicating suspicious clauses or omissions when compared with relevant laws and internal regulations for compliance checks · Information indicating a summary report that quickly searches past cases and relevant regulations for legal database searches · Information indicating areas with violation risks when launching new businesses or product launches for risk assessment · Information that enables automatic notification when the contract renewal time or payment deadline approaches for automatic reminders · Information indicating negotiation room in terms of price and conditions from past similar contract data for negotiation support · Information indicating potential bugs and security holes as a code analysis tool for bug detection · Information indicating code formatting based on coding rules for style guide compliance checks · Information indicating the optimal architecture and design patterns according to requirements for design pattern proposals · Information indicating the assumed load and required server resources based on system requirements for load prediction · Information automatically showing API specifications from code comments and annotations for automatic API document generation · Information that can summarize long technical materials into a short and easy-to-understand form for the summary of the developer guide, reducing the learning cost · Information showing a Gantt chart based on task estimation, resource allocation, and dependency analysis for automatic schedule optimization · Information showing the risks of delivery delays and cost overruns from past project data for risk prediction · Information aggregating customer hearings, support tickets, and voices on SNS and indicating priorities for aggregating customer feedback · Information extracted from a large number of documents and meeting memos, organizing core requirements as a backlog for extracting functional requirements · Information summarizing the project progress and showing reports for the team and upper management for automatic generation of status reports · Information organizing prediction scenarios, merits, and / or demerits for multiple policy options for decision-making analysis · Information showing reference visual concepts (colors, fonts, layout examples) from text requirements for image generation · Information identifying areas where users are likely to leave and showing improvement plans from past UI test data for user behavior analysis · Information showing a dynamic UI layout according to region, language, and access device for UI optimization according to user profiles · Information extracting insights from AB tests and heatmap analysis and showing points for next design improvement for summarizing user test results · Information automatically classifying and tagging design materials for tagging images and icons · Information showing design components along new brand guidelines for supporting the update of the design system · Information showing future demand using past sales data, seasonal factors, and trend information for a demand prediction model · Information showing the optimal replenishment time and order quantity calculated based on inventory levels and demand prediction for automatic ordering · Information indicating the shortest and lowest-cost delivery route determined based on consideration of transportation costs and / or time between multiple locations for optimizing the delivery route · Information indicating a prior warning of delivery delay risk using weather, traffic conditions, and supplier information for predicting delay risk · Information indicating the optimal inventory placement based on the movement of goods and similar order trends for improving picking efficiency · Information indicating the automatic evaluation of supplier delivery quality and delivery compliance rate and showing the priority for supplier evaluation · Information indicating lesion locations from X-rays, MRIs, CTs, etc. for medical image analysis · Information indicating a predicted diagnosis from past similar cases and guidelines based on patient symptom data for case matching · Information indicating the optimal treatment plan (drug selection, surgical method, rehabilitation program) based on the individual data of a patient for protocol proposal · Information indicating the side effect risk when multiple drugs are used together for side effect prediction · Information indicating pamphlets and explanatory texts that simplify and summarize difficult medical information for patients for creating medical explanatory materials · Information that can automatically remind of regular health check-up and re-examination times based on the medical history and examination results of a patient for follow-up proposal · Information indicating teaching materials for reinforcement learning by identifying areas of weakness from the proficiency data of learners for learning data analysis · Information indicating problem sets and tasks tailored to the progress of each student for individual optimal learning · Information indicating scoring by natural language analysis of descriptive answers and according to certain scoring criteria for test grading · Information indicating areas for improvement and supplementary explanations based on the answering tendencies of students for feedback generation · Information indicating a summary for slides by extracting key points from long teaching materials for content summarization · Information indicating practice problems with gradually increasing difficulty according to a specified topic for example creation · Information indicating the extraction and / or summarization of only relevant information from a vast number of news sources for news aggregation · Information indicating trend keywords and highly topical themes for keyword analysis, and suggesting article themes · Information indicating initial article text based on given topics and requirements for draft generation · Information indicating source verification of proper nouns and data in an article for fact-checking assistance, showing credibility · Information indicating content format and / or distribution timing tailored to the reader group by analyzing past article viewing trends for reader analysis · Information indicating titles, headings, meta tags, internal link strategies, etc. for SEO optimization

[0226] Furthermore, it may be provided in each of the aspects described below.

[0227] (1) An information processing system, comprising one or more processors capable of performing the following steps: In a first acquisition step, audio information indicating the audio of a meeting is acquired. In a second acquisition step, term information from which terms related to the meeting can be acquired is acquired. In a generation control step, meeting text information in which at least a part of the audio of the meeting is texturized based on the audio information and the term information is acquired. An information processing system.

[0228] (2) In the information processing system according to (1) above, In the generation control step, the meeting text information is acquired by inputting the audio information and the speech-to-text information including the text obtained by speech-to-text conversion of the audio of the audio information into a learned model. An information processing system.

[0229] (3) In the information processing system according to (2) above, In the generation control step, the meeting text information is acquired by inputting the terms obtained from the term information into the learned model. An information processing system.

[0230] (4) In the information processing system according to any one of (1) to (3) above, In the generation control step, by inputting the term information into a learned model, the term is extracted from the term information. Information processing system.

[0231] (5) In the information processing system according to any one of (1) to (4) above, The term information is information indicating the minutes of a meeting different from the said meeting. Information processing system.

[0232] (6) In the information processing system according to any one of (1) to (5) above, The term information includes terms not listed in a predetermined dictionary. Information processing system.

[0233] (7) In the information processing system according to any one of (1) to (6) above, In the generation control step, by inputting the speech-to-text information indicating the text obtained by speech recognition of the speech information into a learned model, the meeting text information is obtained as text in a state where the orthographic variations between words included in the text are eliminated. Information processing system.

[0234] (8) In the information processing system according to any one of (1) to (7) above, In the generation control step, by inputting the speech-to-text information indicating the text obtained by speech recognition of the speech information into a learned model, the meeting text information is obtained as text in a state where the text corresponding to fillers included in the text is removed. Information processing system.

[0235] (9) An information processing method executed by an information processing system, An information processing method comprising each process executed by the processor of the information processing system according to any one of (1) to (8). Information processing method.

[0236] (10) A program, for causing the processor of the information processing system according to any one of (1) to (8) to function. Program.

[0237] Of course, this is not the limit.

[0238] Finally, although various embodiments of the present invention have been described, these are presented as examples and are not intended to limit the scope of the invention. The novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the gist of the invention. The embodiments and their modifications are included in the scope and gist of the invention and are also included in the invention described in the claims and its equivalent scope.

Explanation of Signs

[0239] 1: Information processing system 2: Server device 21: Processor 210: Information transmission / reception unit 211: Display control unit 212: Acquisition unit 213: Generation control unit 214: Recording unit 215: Reproduction unit 216: Memory control unit 22: Memory unit 23: Communication unit 3: Client device 31: Processor 32: Memory unit 33: Communication unit 34: Output unit 35: Input unit 5: Minutes screen 50: Meeting information area 51: Information Input Area 52: Timestamp Area 53: Information Input Area 54: Playback Area 55: Generation Area 6: Minutes Screen 60: Meeting Information Area 61: Information Input Area 62: Timestamp Area 63: Meeting Text Area 64: Playback Area 65: Generation Area 7: Minutes Screen 70: Meeting Information Area 71: Information Input Area 72: Timestamp Area 73: Meeting Text Area 74: Playback Area 75: Generation Area 8: Mail Screen 80: Destination Area 81: Subject Area 82: Attachment Area 83: Body Area 9: Minutes Screen 90: Meeting Text Area 91: Generation Area 92: Information Input Area NW: Network

Claims

1. An information processing system, comprising one or more processors capable of executing the following steps: In a first acquisition step, voice information indicating the voice of a meeting is acquired; In a second acquisition step, term information from which terms related to the meeting can be acquired is acquired; In a generation control step, meeting text information in which at least a part of the voice of the meeting is texturized is acquired by inputting the voice information, the term information, and text conversion information including text obtained by converting the voice of the voice information into text into a learned model; An information processing system.

2. In the information processing system according to Claim 1, in the second acquisition step, the term information is acquired from material information from which terms related to the meeting can be acquired, and the material information is information acquired from materials planned to be used in the meeting. An information processing system.

3. In the information processing system according to Claim 1, in the generation control step, the terms are extracted from the term information by inputting the term information into a learned model. An information processing system.

4. In the information processing system according to Claim 1, the term information is information indicating the minutes of a meeting different from the meeting. An information processing system.

5. In the information processing system according to Claim 1, the term information includes terms not listed in a predetermined dictionary. An information processing system.

6. In the information processing system according to Claim 1, in the generation control step, the meeting text information is acquired as text in a state where the variation in the notation between words included in the text is eliminated by inputting text conversion information indicating text obtained by converting the voice of the voice information into text into a learned model. An information processing system.

7. In the information processing system according to Claim 1, in the generation control step, the meeting text information is acquired as text in a state where text corresponding to fillers included in the text is removed by inputting text conversion information indicating text obtained by converting the voice of the voice information into text into a learned model. An information processing system.

8. An information processing method executed by an information processing system, comprising each process executed by the processor of the information processing system according to any one of Claims 1 to 7. An information processing method.

9. A program, A program for causing the processor of the information processing system according to any one of claims 1 to 7 to function as a program.

Images