Program and conference analysis device

Abstract

Conducting appropriate analysis across multiple meetings. [Solution] A program for causing a processor to perform an acquisition process that acquires meeting information, which includes information about each of a plurality of meetings held by a predetermined group, and which includes information about the agenda or proceedings of the meeting and information about the date and time the meeting was held; an analysis process that accepts the specification of at least an analysis period and processes information that corresponds to the period, which is extracted from the meeting information based on the date and time information; and an output process that outputs the results of the processing.

Description

Technical Field

[0001] The present invention relates to a program and a conference analysis device.

Background Art

[0002] Conventionally, research and practical application have been advanced regarding assistance such as acquiring the voice and images of a conference and creating minutes of the meeting. In relation to this, an invention of a device for calculating the excitement level of a conference has been disclosed (Patent Document 1). This device extracts feature amounts as a time series along the progress of the conference from conference information, calculates the excitement level as an absolute evaluation value as a time series along the progress of the conference based on the feature amounts, and further corrects the excitement level as an absolute evaluation value to the excitement level as a relative evaluation value.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The invention described in Patent Document 1 calculates the transition of the excitement level within one conference, and does not perform various analyses across a plurality of related conferences. Thus, in the conventional technology, there were cases where suitable analyses could not be performed across a plurality of conferences.

[0005] The present invention has been made in consideration of such circumstances, and one of its objectives is to provide a program and a conference analysis device that can perform suitable analyses across a plurality of conferences.

Means for Solving the Problems

[0006] One aspect of the present invention is a program that causes a processor to perform an acquisition process to acquire meeting information, which includes information about each of a plurality of meetings held by a predetermined group, and which includes information about the agenda or proceedings of the meeting and information about the date and time the meeting was held; an analysis process that accepts the specification of at least an analysis period and processes information that corresponds to the period, which is extracted from the meeting information based on the date and time information; and an output process that outputs the results of the processing.

[0007] Another aspect of the present invention is a meeting analysis device comprising: an acquisition unit that acquires meeting information including information relating to each of a plurality of meetings held by a predetermined group, the meeting information including information relating to the agenda or proceedings of the meeting and information relating to the date and time the meeting was held; an analysis processing unit that accepts the designation of at least an analysis period and processes information that corresponds to the period, extracted from the meeting information based on the date and time information; and an output unit that outputs the results of the processing. [Effects of the Invention]

[0008] According to one aspect of the present invention, suitable analysis can be performed across multiple meetings. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of the configuration and surrounding environment of the conference analysis device 100 according to the present invention. [Figure 2] This figure shows an example of the contents of meeting information 152. [Figure 3] This figure shows an example of the contents of meeting minutes 152A before it was compiled. [Figure 4] This flowchart shows an example of the processing flow performed by the conference analysis device 100. [Figure 5] This figure shows an example of an interface screen IM1 for receiving speaker input. [Figure 6] This figure shows an example of the content of meeting minutes 152B after it has been compiled. [Figure 7]This figure shows the first example of an image generated as a result of the analysis process. [Figure 8] This figure shows a second example of an image generated as a result of the analysis process. [Figure 9] This figure shows a third example of an image generated as a result of the analysis process. [Figure 10] This figure shows a fourth example of an image generated as a result of the analysis process. [Figure 11] This figure shows a fifth example of an image generated as a result of the analysis process. [Modes for carrying out the invention]

[0010] [Summary] The following describes embodiments of the program and meeting analysis device of the present invention with reference to the drawings. The program causes an information processing device having a processor to function as a meeting analysis device. The meeting analysis device analyzes the content of multiple meetings held by a predetermined group, for example, a company. The multiple meetings are basically held at different times, but may include meetings that overlap in time slots. The analysis targets include various things such as speakers, the quality of the meetings, and categorized meeting elements. In the following description, the meeting analysis device will be provided as a cloud service via a network, but the meeting analysis device may also be an on-premise type device installed on the client's equipment.

[0011] [composition] Figure 1 shows an example of the configuration and surrounding environment of the conference analysis device 100 according to the present invention. The conference analysis device 100 communicates with a terminal device 10 and an LLM (Large Language Model) server 200, for example, via a network. The terminal device 10 is a terminal device of a user who wishes to use the services provided by the conference analysis device 100, and is a computer device such as a personal computer, tablet terminal, or smartphone. On the terminal device 10, a program such as a browser or a dedicated application program runs to receive the services of the conference analysis device 100. The network includes the internet, cellular network, dedicated line, WAN (Wide Area Network), LAN (Local Area Network), etc.

[0012] The conference analysis device 100 includes, for example, an IF (interface) unit 110, an analysis processing unit 120, and a storage unit 150. The IF unit 110 includes an acquisition unit 112 and an output unit 114. The analysis processing unit 120 includes an analysis processing unit 120 and an other processing unit 124. The components other than the storage unit 150 are realized by a hardware processor such as a CPU (Central Processing Unit) executing a program (software). Some or all of these components are LSIs (Large Scale Integrations), ASICs (Application Specific Integrated Circuits), FPGAs (Field-Programmable Graphite Arrays), etc. The program may be implemented by hardware (including circuitry) such as a Gate Array, GPU (Graphics Processing Unit), or SOC (System On Chip), or by the collaboration of software and hardware. The program may be stored in advance on a storage device such as an HDD (Hard Disk Drive) or flash memory (a storage device equipped with a non-transient storage medium), or it may be stored on a removable storage medium such as a DVD or CD-ROM (a non-transient storage medium) and installed when the storage medium is inserted into a drive device.

[0013] The storage unit 150 is a RAM (Random Access Memory), HDD, flash memory, etc. The storage unit 150 stores meeting information 152, which contains information corresponding to each of multiple meetings. Figure 2 is a diagram showing an example of the contents of the meeting information 152. The meeting information 152 is acquired by the acquisition unit 112. The meeting information 152 associates information such as the meeting ID (which is merely a label and may be omitted if the meeting can be identified by the date and time and agenda) with information such as the date and time of the meeting, agenda, supplementary information on the agenda, number of pages of materials (total and per agenda), attendees, meeting minutes (before organization) 152A, and meeting minutes (after organization) 152B. The supplementary information on the agenda includes information such as the agenda name, category, importance, agenda type (whether it is a new meeting or a continuing meeting that takes over from the previous meeting, and if it is a continuing meeting, the meeting ID of the previous meeting), resolution results, voter turnout, and undecided flag. A category is a classification of the type of proposal, such as label information like "management strategy," "governance," or "financial reporting." The "pending" flag indicates that a resolution was not made or was rejected at this meeting.

[0014] Among the meeting information 152, the holding date and time, agenda items, number of pages of materials (total and for each agenda item), and information on attendees are, for example, information registered by the user before the meeting. The IF unit 110 has, for example, the function of a web server and provides a web page serving as an interface for providing the functions of the meeting analysis device 100 to the terminal device 10. The acquisition unit 112 acquires these pieces of information via this web page. Instead of the web page, a dedicated application program or the like may be used as the interface. The output unit 114 provides the result of the processing performed by the analysis processing unit 120 to the terminal device 10.

[0015] Among the meeting information 152, the supplementary information for the agenda items is, for example, information registered by the user after the meeting or during the meeting. The acquisition unit 112 acquires these pieces of information via the above-mentioned web page.

[0016] FIG. 3 is a diagram showing an example of the content of the unorganized meeting information 152A. The unorganized meeting information 152A includes, for example, a written text (utterance content) of the meeting utterances segmented for each utterance (speech). Each utterance is provided with information such as the speaker, the utterance time (start time and end time of the utterance), and the tone. Since the speaker is, for example, represented together with the job title, the unorganized meeting information 152A may include job information. Further, the unorganized meeting information 152A may include information on the number of utterances (in other words, the utterance frequency) and the total utterance time for each speaker obtained by statistically processing the utterance time. This statistical processing is performed, for example, by the analysis processing unit 120.

[0017] FIG. 4 is a flowchart showing an example of the processing flow executed by the meeting analysis device 100. First, after the meeting is held (or in real time as the meeting progresses), the acquisition unit 112 acquires the audio data of the meeting (S1). The audio data may be uploaded from the terminal device 10 or uploaded from a recording device to the meeting analysis device 100.

[0018] The order of processing described below is merely an example and can be rearranged as desired. The acquisition unit 112 performs a text conversion process on the audio data to convert it into text data for each speaker (S2). The text conversion process includes, for example, speech recognition processing and morphological analysis processing. The text conversion process may be performed by the conference analysis device 100 or by outsourcing it to an external device.

[0019] Next, the acquisition unit 112 outputs the results of the text conversion process to the terminal device 10 and accepts input of information indicating which attendee made the utterance (i.e., the speaker or utterances) for each utterance (S3). Based on this, the acquisition unit 112 associates the utterances with the speakers and registers the associated information as transcribed text in the meeting minutes information 152A before organization.

[0020] Figure 5 shows an example of an interface screen IM1 for receiving speaker input. On the interface screen IM1, the speaker and utterance content are displayed in association with the utterance time. When icon A1 in the speaker section is operated, a list of attendees obtained from the meeting information 152 is presented as candidate speakers, allowing the user to select a speaker. After this operation and confirmation, the acquisition unit 112 acquires the speaker information. Alternatively, instead of having the user of the terminal device 10 input speaker information, the meeting analysis device 100 may automatically recognize the speaker by analyzing the audio data. In this case, the meeting analysis device 100 may present candidate speakers to the terminal device 10 and obtain user approval. When automatic speaker recognition is performed, data entered several times by the user of the terminal device 10 may be used as correct data.

[0021] Returning to Figure 4, the acquisition unit 112 acquires the tone of speech for each utterance (S4). The acquisition unit 112 inputs the voice data (and text data) into an emotion estimation model that has been trained to estimate emotions based on the content and pitch of the utterance, and acquires the tone of speech. The tone of speech is acquired as one of the predetermined types, such as "positive," "negative," "neutral," or "businesslike." The emotion estimation model may estimate the tone of speech based on both the voice and text data, or it may estimate the tone of speech based only on the text data. In addition to the tone of speech, the acquisition unit 112 may also estimate the speaker's emotions, such as joy, anger, sadness, or happiness. The speaker's emotions are estimated, for example, from a series of text data using a different emotion estimation model for each utterance period. Alternatively, the speaker's emotions may be derived as cumulative information of the tone of speech (for example, "joy" if "positive" is repeated).

[0022] Next, the analysis processing unit 120 aggregates the number of utterances and the duration of utterances (S5). This process creates the pre-organized meeting minutes information 152A. This process may also be performed by the acquisition unit 112.

[0023] Next, the analysis processing unit 120 inputs the unorganized meeting minutes information 152A, a prompt including the instruction to "organize and output the meeting minutes information for each agenda item," and the instruction to "calculate the time spent explaining and discussing the agenda item" to the large-scale language model (hereinafter referred to as LLM) provided by the LLM server 200, thereby obtaining the organized meeting minutes information 152B (S6). This processing may also be performed by the acquisition unit 112. Although there may be embodiments that do not utilize the large-scale language model provided by the LLM server 200 (and obtain the organized meeting minutes information 152B by other methods), the existence of the LLM server 200 is shown here. For example, the meeting analysis device 100 itself may store programs and data for functioning as an LLM in the storage unit 150, and the LLM function may be incorporated as a function of the analysis processing unit 120 or as a function of another functional unit. Alternatively, the meeting analysis device 100 may obtain the organized meeting minutes information 152B by another method instead of LLM.

[0024] Figure 6 shows an example of the content of the edited meeting minutes 152B. The edited meeting minutes 152B is information arranged chronologically, such as which agenda item was spoken about, the content of the utterance (e.g., a summary of the transcribed text), the speaker, whether it was an explanation or a discussion, the duration of the utterance, and the tone. The tone information in the edited meeting minutes 152B is created by the LLM based on the transcribed text and may differ from the tone information in the unedited meeting minutes 152A.

[0025] The analysis processing unit 120 receives a specification of the period (analysis target period) and analysis content from the terminal device 10 (S7), and processes the information corresponding to the period, which is extracted from the meeting information 152 based on the date and time information (date and time of meeting), based on the analysis content (S8). The period can be specified in units such as days, months, years, quarters, or semi-annually. The analysis content is the perspective from which the user will analyze the acquired meeting information. For example, the analysis content is information that the user wants to visualize chronologically for the specified period, such as the number of agenda items or the number of pages of materials for each agenda item. Note that the analysis content may be set in advance and fixed, or the user may select from several candidates and use them repeatedly. In that case, the processing to receive the specification of the analysis content in S7 is omitted, and only the specification of the period is received. Furthermore, in addition to the period, the analysis processing unit 120 may also receive the specification of specific agenda items and attendees, and process the information corresponding to them based on the analysis content. Then, the output unit 114 outputs the results of the analysis processing to the terminal device 10 (S9).

[0026] [Analysis Processing] The analysis processing unit 120 performs analysis processing as exemplified below. These analysis processes correspond to the "specification of analysis content" received in S7 in the flowchart of Figure 4 above.

[0027] (A) The analysis processing unit 120 may perform a content analysis of the discussions. Specifically, the analysis processing unit 120 may analyze the time allocation for each type of discussion (reporting, decision-making, debate), and the ratio of new agenda items to continuing agenda items. For example, the analysis processing unit 120 may divide the number of new or continuing agenda items entered as supplementary information for each agenda item by the total number of such items. The analysis processing unit 120 may also analyze the resolution results (approval, rejection, abstention) for each agenda item. For example, the analysis processing unit 120 may obtain the approval, rejection, or abstention tags entered as supplementary information for each agenda item, and divide the number of people who voted for each agenda item by the total number of people who should have voted for each agenda item.

[0028] (B) The analysis processing unit 120 may perform an analysis of the depth of the discussion. The depth of the discussion can be expressed, for example, by the complexity of the agenda items (the proportion of agenda items that include multiple themes). For example, the analysis processing unit 120 may calculate an index value indicating the depth of the discussion based on the number of themes covered, the number of pages in the materials, the number of stakeholders, and the scope of impact of the decisions (e.g., company-wide level, departmental level). In this case, the analysis processing unit 120 may assume that the discussion is more complex if there are many keywords from each field appearing in the meeting minutes data, many pages in the materials, many people who spoke during the discussion, and many people and organizations appearing in the meeting minutes data, and then qualitatively evaluate the depth of the discussion and obtain the evaluation result of the depth of the discussion by inputting a prompt to the LLM instructing it to output the evaluation result as a continuous value or in stages.

[0029] (C) The analysis processing unit 120 may also perform an analysis of the efficiency (cost) of the meeting. The efficiency of the meeting can be expressed, for example, as the ratio of explanation time to discussion time. In this case, the analysis processing unit 120 may input instructions to the LLM to calculate the time taken from the start of discussion to questioning, and the time taken from questioning to conclusion, based on the information entered as supplementary information for the agenda, and obtain the results. The analysis processing unit 120 may also input instructions to the LLM to calculate the time taken from the start of questioning to conclusion, based on the time information attached to the meeting minutes data, based on the information entered as supplementary information for the agenda, and obtain the results. The analysis processing unit 120 may also input instructions to the LLM to calculate the time taken from a statement indicating a break to a statement indicating resumption, based on the information entered as supplementary information for the agenda, and obtain the results. The analysis processing unit 120 may also input instructions to the LLM to calculate the time taken from a statement indicating a break to a statement indicating resumption, based on the information entered as supplementary information for the agenda, and obtain the results. Alternatively, the analysis processing unit 120 may input an instruction to the LLM to calculate the time (delay time) from the actual start time to the time when the statement indicating the start of the meeting was made, and obtain the result. Alternatively, the analysis processing unit 120 may refer to the list of attendees, obtain the job title information of each attendee, obtain the salary information for that job title from the salary table based on the job title information, and calculate the meeting cost using the formula: hourly rate × number of meeting hours.

[0030] [Image generation] Examples of the analysis process include the following. For example, the analysis processing unit 120 obtains the number of agenda items and the amount of material for each agenda item for each meeting, and generates an image (an example of information to be displayed on the terminal device 10) that compares and displays them. In the following description, various types of information will be represented as images in the form of graphs, but they may also be represented as images in the form of tables or descriptive text. In any case, the information that is the result of the analysis processing unit 120 will be displayed on the terminal device 10. Figure 7 is a diagram showing a first example of an image generated as a result of the analysis process. As shown in the diagram, the analysis processing unit 120 may, for example, perform monthly aggregation processing regarding the number of agenda items and the number of pages of materials, and generate an image that represents them as monthly data in a graph. This allows users to check whether the number of agenda items and the amount of material are excessive, and guides them to allocate time to the topics of interest.

[0031] Furthermore, the analysis processing unit 120 may derive information on discussion time and explanation time in the meeting and generate an image showing their time-series changes. Discussion time is the time occupied by utterances described as "discussion" in the compiled meeting minutes information 152B, and explanation time is the time occupied by utterances described as "explanation" in the compiled meeting minutes information 152B. Figure 8 shows a second example of an image generated as a result of the analysis processing. As shown in the figure, the analysis processing unit 120 may perform monthly aggregation processing on the total and percentage of discussion time and explanation time, and generate an image that represents them as monthly data in a graph. This allows viewers to confirm whether or not the explanation time on the day has been compressed by providing thorough pre-meeting explanations, and whether or not the necessary discussion time has been secured.

[0032] Furthermore, the analysis processing unit 120 may generate an image showing the degree to which each agenda item was discussed in the meeting in chronological order. The agenda item's theme is, for example, information that further details the "category" mentioned above. If the category is "management strategy," then the agenda item's theme might be something like "Mid-term management strategy for XX year," with additional limiting information such as specifying a time period. Figure 9 shows a third example of an image generated as a result of the analysis processing. As shown in the figure, the analysis processing unit 120 may perform monthly aggregation processing on the total time and percentage of time that each agenda item's theme was discussed, and generate an image that represents these as monthly data in a graph. This helps to move the meeting away from being merely a formal confirmation body and guide it towards becoming a forum for discussing themes that improve corporate value.

[0033] Furthermore, the analysis processing unit 120 may generate an image summarizing the speaking time for each agenda item for each executive attending the meeting. Figure 10 shows a fourth example of an image generated as a result of the analysis processing. As shown in the figure, the analysis processing unit 120 may, for example, aggregate the speaking volume (speaking time, or the amount of information resulting from the transcription of the speech) for each agenda item for each executive over a quarter and generate an image representing this as a graph.

[0034] Furthermore, the analysis processing unit 120 may generate an image that shows the volume of materials for each agenda item in comparison with time information. Time information refers to, for example, the discussion time and explanation time in a meeting. Figure 11 shows a fifth example of an image generated as a result of the analysis processing. As shown in the figure, the analysis processing unit 120 may generate a graph image for each agenda item that allows for comparison of the number of pages of materials, discussion time, and explanation time. This allows viewers to recognize whether excessive time is being spent on explanations, whether there is insufficient time for discussion on certain agenda items, or whether the total amount of materials is reasonable.

[0035] [Meeting Quality] The analysis processing unit 120 may not only generate images that have undergone the statistical processing described above, but may also derive index values ​​indicating the quality of the meeting using the methods described below. The elements (values) that form the basis of the index values ​​may be calculated by the analysis processing unit 120 itself using a rule-based method, or they may be obtained by requesting calculation from an LLM. The analysis processing unit 120 may also generate images that perform time-series comparisons of the index values, comparisons for each agenda item, comparisons for each attendee, etc.

[0036] (1) For example, the analysis processing unit 120 may derive an index value that indicates a favorable value (e.g., a high value) if the degree of involvement of outside directors in the meeting is high (specifically, the amount of speaking by outside directors or the amount of questions asked of outside directors is high), indicating that opinions have been collected more broadly. To "derive an index value that indicates a favorable value" means, for example, assuming that the index value is obtained by a weighted sum of multiple elements, that the partial index value for the element in question is set to a favorable value. The index value is not limited to a continuous value, but may also be expressed in stages such as high, medium, and low. The same applies hereafter.

[0037] (2) The analysis processing unit 120 may also derive an index value that indicates a more active discussion if the degree of diversity among speakers is high (specifically, if the range of positions among speakers is wide, or if there are many speakers). The analysis processing unit 120 may also derive an index value that indicates a more active discussion if there are many negative utterances.

[0038] (3) The analysis processing unit 120 may also derive an index value that is more favorable the higher the continuity of discussions between meetings (i.e., the higher the level of completion of discussions that were not concluded in the previous meeting), or conversely, the higher the level of completion of the meetings.

[0039] (4) The analysis processing unit 120 may also derive an indicator value that indicates a favorable value, such as the amount of overlap in agenda items in meetings between different levels, like the executive board and the board of directors, which means that unnecessary time has been reduced.

[0040] (5) The analysis processing unit 120 may also derive an indicator value that indicates a favorable outcome if the agenda items are consistent across multiple meetings, but the content of the explanations and questions and answers differ, indicating that thorough discussions have taken place in the later meetings. In this case, if the order of the meetings is correct (for example, in the order of lower to higher levels within the company hierarchy), the unit may derive an indicator value that indicates an even more favorable outcome.

[0041] (6) The analysis processing unit 120 may also derive an index value that indicates a favorable value when a specific sequence of processes (for example, explanation → opinion / discussion → question and answer) occurs in the meeting.

[0042] (7) The analysis processing unit 120 may also derive an undesirable index value if there are many detailed questions and answers in the meeting, indicating that the materials were insufficient.

[0043] (8) Alternatively, the analysis processing unit 120 may obtain the results of a questionnaire given to the meeting attendees and input a prompt to the LLM instructing it to divide the attendees' questionnaire results information and the number of people who gave a positive response by the total number of questionnaires. The analysis processing unit 120 may also perform a similar process within its own algorithm.

[0044] (9) The analysis processing unit 120 may also count the number of agenda items that had negative comments or comments prompting reconsideration from the compiled meeting minutes information 152B, and input a prompt to the LLM to output the ratio to the total number of agenda items, thereby obtaining the ratio as the quality of the meeting.

[0045] (10) Alternatively, the analysis processing unit 120 may input a prompt to the LLM instructing it to count and output the number of organizations appearing in each agenda item from the organized meeting minutes information 152B, and obtain the result. The analysis processing unit 120 may also perform a similar process within its own algorithm.

[0046] In addition to the above functions, the meeting analysis device 100 may also have a function to automatically create meeting minutes from transcribed text or the like.

[0047] According to the embodiments described above, meeting information is obtained that includes information about each of multiple meetings held by a predetermined group, and that includes information about the agenda or proceedings of the meeting and information about the date and time the meeting was held. The system accepts at least a specified period for analysis and processes the information that corresponds to the analysis period, which is extracted from the meeting information based on the date and time information, thereby enabling suitable analysis across multiple meetings.

[0048] Although embodiments for carrying out the present invention have been described above using examples, the present invention is not limited in any way to these embodiments, and various modifications and substitutions can be made without departing from the spirit of the present invention. [Explanation of Symbols]

[0049] 10 Terminal devices 100 Conference Analysis Device 110 IF section 112 Acquisition Department Output section of 114 120 Analysis Processing Unit 150 Storage section 152 Meeting Information 152A Meeting minutes (before editing) 152B Agenda information (after sorting)

Claims

1. In the processor, A process for acquiring meeting information that includes information about each of multiple meetings held by a specified group, and which includes information about the agenda or proceedings of the meeting and information about the date and time the meeting was held. An analysis process that accepts at least the designation of an analysis period and processes the information corresponding to the analysis period, which is extracted from the meeting information based on the date and time information, Output processing to output the results of the above processing, A program to execute.

2. The aforementioned processor performs the analysis process as follows: Further requests for specifications of the analysis content are accepted, and information corresponding to the analysis period is processed based on the analysis content. The program according to claim 1.

3. The aforementioned processor performs the analysis process as follows: The information from multiple meetings that fall within the aforementioned analysis period is processed together. The program according to claim 1.

4. The information relating to the aforementioned proposal includes some or all of the following: proposal name, category, importance, proposal type, resolution result, and voter turnout. The program according to claim 1.

5. The information relating to the aforementioned proposals includes information on the volume of materials for each of the aforementioned proposals, The information regarding the volume of materials for each agenda item is information registered by users as meeting information before the meeting was held. The program according to claim 1.

6. The aforementioned processor performs the analysis process as follows: Accepting the designation of a specific proposal, From the aforementioned meeting information, the information corresponding to the specific agenda item and the aforementioned analysis period will be processed. The program according to claim 1.

7. The information relating to the proceedings includes some or all of the following information: the speaker, the content of each speaker's utterance, the duration of each speaker's utterance, the number or frequency of each speaker's utterances, and the tone of speech. The program according to claim 1.

8. The information regarding the aforementioned proceedings includes information about the speaker's position, The program according to claim 1.

9. The aforementioned processor performs the analysis process and the output process as follows: The terminal device displays information comparing the number of agenda items and the amount of material for each agenda item for each meeting. The program according to claim 1.

10. The aforementioned processor performs the analysis process and the output process as follows: The terminal device is configured to derive information on the discussion time and explanation time at the aforementioned meeting, and to display information showing the time-series changes of the discussion time and explanation time. The program according to claim 9.

11. The aforementioned processor performs the analysis process and the output process as follows: The terminal device displays information showing the degree to which each agenda item was addressed in the aforementioned meeting, in chronological order. The program according to claim 1.

12. The aforementioned processor performs the analysis process and the output process as follows: The terminal device displays aggregated information on the amount of speech each speaker and each agenda item spoken. The program according to claim 1.

13. The aforementioned processor performs the analysis process and the output process as follows: The terminal device displays information comparing the volume of materials for each of the aforementioned agenda items with time information. The program according to claim 1.

14. The aforementioned processor, In the aforementioned analysis process, information on the discussion time and explanation time at the meeting is derived, The aforementioned time information is information on the discussion time and explanation time at the meeting. The program according to claim 13.

15. The aforementioned processor, In the aforementioned analysis process, an indicator value showing the quality of the meeting is derived. The program according to claim 1.

16. The aforementioned processor, In the aforementioned analysis process, an index value indicating the quality of the meeting is derived based on the degree of involvement of the outside directors. The program according to claim 15.

17. The aforementioned processor, In the aforementioned analysis process, an index value indicating the quality of the meeting is derived based on the degree of diversity among the speakers. The program according to claim 15.

18. The aforementioned processor, To derive an index value indicating the quality of the meeting based on the continuity of discussions between meetings, The program according to claim 15.

19. An acquisition unit that acquires meeting information, which includes information about each of multiple meetings held by a specified group, and which includes information about the agenda or proceedings of the meeting and information about the date and time the meeting was held. An analysis processing unit that accepts at least a specified period for analysis and processes information corresponding to the period, extracted from the meeting information based on the date and time information, An output unit that outputs the result of the above processing, A conference analysis device equipped with the following features.

Images