Information processing device

Abstract

The challenge is to enable more accurate and easier determination of appropriate participants for upcoming meetings compared to conventional technologies. [Solution] The audio acquisition unit 101 acquires audio data V of meeting M. The meeting determination unit 102 extracts n statements C contained in the audio data V acquired by the audio acquisition unit 101 and identifies the attendee A who made each of the n statements C. The attendee management unit 103 manages the attendee ID, which uniquely identifies the attendee A identified as the speaker by the meeting determination unit 102, by associating it with the meeting theme T and the statement. The candidate proposal unit 106 proposes one or more suitable candidates as attendees for the newly scheduled meeting based on the information managed by the attendee management unit 103 and information that includes at least the theme T of the newly scheduled meeting. This solves the above problem.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus.

Background Art

[0002] Conventionally, there is a technique for making proposals for participants in a newly held meeting based on the content of the remarks of those who attended a previously held meeting and the theme of the newly held meeting (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, including the technology of Patent Document 1, there is a situation where development of a method that can accurately and easily determine appropriate participants in a meeting to be held is desired compared to conventional technologies.

[0005] The present invention has been made in view of such a situation, and an object thereof is to be able to accurately and easily determine appropriate participants in a meeting to be held compared to conventional technologies.

Means for Solving the Problems

[0006] To achieve the above object, an information processing apparatus according to an aspect of the present invention includes: acquisition means for acquiring audio data of a meeting; specification means for specifying each speaker of a speech in a predetermined unit included in the audio data acquired by the acquisition means; management means for associating the theme of the meeting and the speech with the information indicating the speaker specified by the specification means and managing them as first information. A proposal means that proposes one or more suitable candidates as attendees of a meeting based on the first information managed by the management means and the second information which includes at least the theme of a newly to be held meeting, It is equipped with. [Effects of the Invention]

[0007] According to the present invention, it is possible to determine, more accurately and easily than with conventional technology, who is suitable to participate in a meeting. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 2] Figure 1 shows an example of a method used to classify and manage attendees based on the quantity and quality of their contributions during meetings, as outlined in the service overview. [Figure 3] This figure shows a different example from Figure 2 of the method used in the overview of this service shown in Figure 1, which classifies and manages attendees based on the quantity and quality of their contributions in meetings. [Figure 4] This figure shows an example of the configuration of an information processing system to which a server according to one embodiment of the present invention is applied. [Figure 5] Figure 4 is a block diagram showing an example of the server hardware configuration in the information processing system. [Figure 6] This is a functional block diagram showing an example of the functional configuration for executing candidate proposal processing among the functional configurations of the server in Figure 5 that constitute the information processing system in Figure 4. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the drawings.

[0010] First, with reference to Figures 1 to 3, we will describe the outline of the service (hereinafter referred to as "this service") that can be realized by an information processing system (see Figure 4, described later) to which Server 1 according to one embodiment of the present invention is applied.

[0011] Figure 1 shows an overview of the service that can be realized by an information processing system to which a server according to one embodiment of the present invention is applied.

[0012] This service is provided by the service provider (not shown) to the organizer H. Attendees A are those who will attend or have attended meeting M. Organizer H is the person who organizes Meeting M, which attendee A will be attending. Note that attendee A may also be Organizer H. The format of meeting M is not particularly limited, but in this embodiment, both in-person and remote formats are used for meeting M. "In-person format" refers to a meeting where attendees A meet face-to-face in a physical meeting room or similar location. "Remote format" refers to a meeting conducted via the internet. The format of meeting M can be arbitrarily selected by each attendee A according to their own circumstances (e.g., commuting, working from home).

[0013] Figure 1 shows the initial meeting M1 taking place in a conference room of a certain company. Meeting M1 is attended in person by attendees A1 through A3, and remotely by attendees A4 and A5.

[0014] In meeting M1, each of the statements C1 to C5 made by attendees A1 to A5 is acquired by server 1 as audio data V (step S1). Specifically, each of the statements C1 to C3 made by attendees A1 to A3 are input by the voice input function of microphone device 4 placed in the conference room and acquired by server 1 as voice data V. Here, microphone device 4 may be placed as one unit in the conference room as shown in Figure 1, or one unit may be placed in front of each of attendees A1 to A5. In addition, the respective statements C4 and C5 of attendees A4 and A5 are input by the voice input function of an information processing device 2 such as a personal computer operated by attendees A4 and A5 (hereinafter referred to as the "attendee terminal 2") and acquired by the server 1 as respective voice data V4 and V5.

[0015] During or after the progress of the meeting M1, the server 1 extracts n statements C (n is an integer value of 1 or more) respectively included in the acquired voice data V, V4, and V5 and performs analysis (step S2). Here, the analysis method by the server 1 is not particularly limited. For example, the n extracted statements C may be converted into text data and analyzed by text mining. Also, for example, analysis may be performed on the voice data itself without converting the n extracted statements C into text data.

[0016] Based on the analysis results of the n statements C, the server 1 estimates the theme T of the meeting M1 (step S3). When the theme T is estimated, the statement C and the theme T are associated and managed. Specifically, for example, when the theme T of "Commercialization of new technology Z" is estimated, the statement C and the theme T of "Commercialization of new technology Z" are associated and managed. Here, the estimation of the theme T by the server 1 can be performed on a per-meeting basis. However, when the meeting M1 is composed of multiple parts (for example, when divided into the first part and the second part, etc.), the estimation of the theme T can also be performed on a per-part basis. Also, the theme T can be estimated on a per-time period basis. For this reason, multiple themes T can also be associated with one statement C. Specifically, for example, the theme T estimated on a per-meeting basis and the theme T estimated on a per-part basis can be associated with the statement C and managed. In this way, in this service, the theme T can be estimated based on the analysis results of the acquired voice data V. However, it is not limited to this, and the organizer H can also manually input the theme T.

[0017] Server 1 identifies the speaker of each of the n statements C contained in the audio data V based on the results of the analysis of the n statements C (step S4). The method used by Server 1 to identify the speaker is not particularly limited; for example, the speaker may be identified based on the recognition result of the audio data V and data indicating the characteristics of attendee A's voice, which has been acquired in advance. Specifically, for example, in the example in Figure 1, each of the statements C1 to C3 of attendees A1 to A3 that may be included in audio data V is identified. In addition, statement C4 of attendee A4 that may be included in audio data V4 is identified, and statement C5 of attendee A5 that may be included in audio data V5 is identified. This allows the individual speaker of each of the n statements C made in meeting M1 to be identified, making it possible to record and manage the content of each statement made by each attendee A in meeting M1. In other words, when server 1 identifies the person who made statement C, it manages the speaker (attendee A), statement C, and theme T in association.

[0018] Server 1 classifies and manages attendees A based on the quantity and quality of their contributions in meeting M (step S5). Specifically, Server 1 classifies attendee A based on the analysis results of statement C, including the amount of speaking time in meeting M1 (e.g., number of statements, number of characters, etc.) and the relationship between the theme T of meeting M1 and the content of statement C. The relationship between the theme T of meeting M1 and the content of statement C can be expressed using an indicator called "theme matching degree" or "theme deviation degree."

[0019] "Theme matching degree" is an index that normalizes the degree of relevance between the content of statement C and the content of theme T. If statement C has a low theme matching degree, it means that statement C is not aligned with the theme T of meeting M1. Conversely, if statement C has a high theme matching degree, it means that statement C is aligned with the theme T of meeting M1.

[0020] Furthermore, "theme deviation" refers to an index that normalizes the degree of relevance between the content of meeting theme T and statement C by comparing them. If the theme deviation of statement C is small, it means that statement C is in line with meeting theme T of meeting M1. Conversely, if the theme deviation of statement C is large, it means that statement C is not in line with meeting theme T of meeting M1. In this service, attendee A is classified and managed based on indicators such as the quantity (e.g., "quantity of speech" mentioned above) and quality (e.g., "degree of theme matching" and "degree of theme deviation" mentioned above) of speeches in meeting M1. Specific examples of the methods for classifying and managing attendee A will be described later with reference to Figures 2 and 3.

[0021] If a subsequent meeting, M2, is scheduled after the conclusion of meeting M, the organizer H sets the requirements for holding meeting M2 (hereinafter referred to as "meeting requirements") (step S6). Meeting requirements are set by the organizer H entering the required information into an information processing device 3 such as a personal computer (hereinafter referred to as "organizer terminal 3"). There are no particular limitations on what can be set as meeting requirements. For example, the date of the next meeting M, the content of the theme T, the number and type of attendees A, etc., can be set as meeting requirements. Specifically, for example, while maintaining theme T, "Regarding the commercialization of new technology Z," as in the previous meeting M1, the meeting requirements could be set so that reliable people make up attendee A in order to facilitate a thorough discussion. Alternatively, the meeting requirements could be set so that people who speak a lot make up the majority of attendee A. Furthermore, to facilitate a free-flowing discussion, the meeting requirement could be deliberately set as "not deciding on theme T."

[0022] Once the organizer H has finished entering the meeting requirements, Server 1 proposes to Organizer H by extracting one or more candidates who are highly suitable for the meeting requirements and presenting them to Organizer H (Step S7). The reason for specifying "one or more" here is that customers may participate in Meeting M. The relevance between the meeting requirements of Meeting M2 and the candidates can be expressed using an indicator called "Requirement Matching Degree." "Requirement Matching Degree" is an indicator that normalizes the degree of relevance between the content of the meeting requirements and the candidate's information. The higher a candidate's "Requirement Matching Degree," the higher their suitability for the meeting requirements, making them more likely to be selected as a candidate to attend Meeting M2. Information indicating one or more selected candidates is not shown graphically, but is displayed, for example, in a ranking format on the organizer's terminal 3.

[0023] Figures 2 and 3 show specific examples of a method for classifying and managing each of the m attendees A (where m is an integer greater than or equal to 1) based on the quantity and quality of their statements C in meeting M. Figure 2(A) shows an example of information stored in a database on server 1 (for example, the attendee DB181 in Figure 6, which will be described later) as a concrete example of a method for classifying and managing each of the m attendees A.

[0024] In other words, the database on server 1 stores information (hereinafter referred to as "attendee information") about each of the m attendees A who were identified as speakers in any of the k meetings M held in the past (where k is an integer of 1 or more). The attendee information includes an ID that uniquely identifies attendee A (hereinafter referred to as "attendee ID"), an ID that uniquely identifies meeting M (hereinafter referred to as "meeting ID"), the content of the theme T, the content of the statement C, the amount of the statement, and the degree of theme matching.

[0025] Specifically, as shown in Figure 2(A), attendee A1, whose attendee ID is "0001", attended meeting M1 (theme T is "Commercialization of New Technology Z"), whose meeting ID is "101", and made a statement C1 with the content "xxxxxx". Attendee A1 was classified as having a "large" volume of speech and a theme matching degree of "50%". In other words, although the content of statement C1 made by attendee A1 at meeting M1 was large in volume, it was determined that it was not necessarily highly relevant to theme T.

[0026] For example, attendee A2, whose attendee ID is "0002," attended meeting M1 (theme T: "Commercialization of New Technology Z"), whose meeting ID is "101," and made a statement C2 containing the content "xxxxxx." Attendee A2 was classified as having a "small" volume of speech and a theme matching degree of "80%." In other words, the content of statement C2 made by attendee A2 at meeting M1 was small in volume, but was considered appropriate and highly relevant to theme T.

[0027] For example, attendee A3, whose attendee ID is "0003," attended meeting M1 (theme T: "Commercialization of New Technology Z"), whose meeting ID is "101," and made a statement C3 containing the content "xxxxxx." Attendee A3 was classified as having a "medium" volume of contribution and a theme matching degree of "20%." In other words, the content of statement C3 made by attendee A3 at meeting M1 was of average volume, but was deemed to have little relevance to theme T.

[0028] For example, attendee A4, whose attendee ID is "0004," attended meeting M1 (theme T: "Commercialization of New Technology Z") with meeting ID "101" and made a statement C4 containing the content "xxxxxx." Attendee A4 was classified as having a "large" volume of contribution and a theme matching degree of "90%." In other words, the content of statement C4 made by attendee A4 at meeting M1 was considered noteworthy because of its large volume and high relevance to theme T. In this way, all statements C made by attendees A1 through A4 who attended meeting M1 are stored in the database of server 1 and can be classified from two perspectives: the amount of speech and theme matching. This allows organizer H to select more appropriate candidates for attendee A at the upcoming conference M2 by referring to the attendee information stored in the server 1 database.

[0029] This service allows you to classify attendee A using a graph like the one shown in Figure 2(B). Figure 2(B) shows a concrete example of a method for classifying attendees A1 to A4 who attended meeting M1 using graph G, which is represented by a horizontal axis L1 indicating the amount of speaking time and a vertical axis L2 indicating the degree of theme matching. In other words, graph G shown in Figure 2(B) shows an example of classifying four attendees A (attendees A1 to A4) out of m attendees A stored in the database shown in Figure 2(A).

[0030] For example, participant A1 in meeting M1, as shown in Figure 2(A), has a high level of participation and a theme matching score of 50%, so is classified at point P1 on graph G. For example, participant A2 in meeting M1, as shown in Figure 2(A), has a "small" speaking volume and a theme matching degree of "80%", so is classified at point P2 on graph G. For example, participant A3 in meeting M1, as shown in Figure 2(A), has a "medium" level of participation and a theme matching score of "20%", so is classified at point P3 on graph G. For example, participant A4 in meeting M1, as shown in Figure 2(A), has a "high" speaking volume and a theme matching degree of "90%", so is classified at point P4 on graph G.

[0031] Thus, in this service, all of the statements C made by the four attendees A (attendees A1 to A4) who attended meeting M1 are stored in the database on server 1. Furthermore, each of attendees A1 to A4 is classified based on two criteria: the amount of their statements and theme matching. This allows organizer H to quickly grasp the relationship between the amount of speaking time and the degree of theme matching for each participant A1 through A4 by referring to graph G generated by server 1 when selecting candidates for attendee A for the upcoming conference M2. As a result, it becomes possible to quickly make more appropriate selections.

[0032] Figure 3 shows an example of a method for classifying and managing each of the m attendees A, which is an example of information stored in a database on server 1, different from (A) in Figure 2.

[0033] Specifically, the database shown in Figure 3 stores the attendee information for each of the m attendees A who were identified as speakers at any of the k meetings M that were held. The attendee information shown in Figure 3 includes the attendee ID of attendee A, the content of attendee A's experience, the meeting ID of meeting M, the content of the theme T of meeting M, the content of statement C, the amount of speaking time, and the degree of theme matching. In other words, as shown in the example in Figure 3, the experience of attendee A is added to the example in Figure 2 (A), making it possible to select people with a more accurate and realistic approach.

[0034] Specifically, as shown in Figure 3, attendee A1, whose attendee ID is "0001," has experience as a "Z development leader" and attended meeting M1 (theme T is "Commercialization of New Technology Z"), whose meeting ID is "101," and made a statement C1 with the content "xxxxxx." Attendee A1 was classified as having a "large" volume of speech and a theme matching degree of "50%." In other words, although the content of statement C1 made by attendee A1 at meeting M1 was large in volume, it was determined that it was not necessarily highly relevant to theme T.

[0035] For example, attendee A2, whose attendee ID is "0002," is a "sales department manager" and attended meeting M1 (theme T: "Commercialization of new technology Z"), whose meeting ID is "101," and made a statement C2 with the content "xxxxxx." Attendee A2 was classified as having a "small" volume of speech and a theme matching degree of "80%." In other words, the content of statement C made by attendee A2 at meeting M1 was small in volume, but was considered appropriate and highly relevant to theme T.

[0036] For example, attendee A3, whose attendee ID is "0003," is a section chief in the general affairs department. Attended meeting M1 (theme T: "Commercialization of New Technology Z"), with meeting ID "101," and made a statement C3 containing the content "xxxxxx." Attendee A3 was classified as having a "medium" volume of contribution and a theme matching degree of "20%." In other words, the content of statement C3 made by attendee A3 at meeting M1 was considered to have an average volume but little relevance to theme T.

[0037] For example, attendee A4, whose attendee ID is "0004," is the manager of the Intellectual Property Department and attended meeting M1 (theme T: "Commercialization of New Technology Z") with meeting ID "101," where he made a statement C4 containing the content "xxxxxx." Attendee A4 was classified as having a "large" volume of contribution and a theme matching degree of "90%." In other words, the content of statement C4 made by attendee A4 at meeting M1 was noteworthy because of its large volume and high relevance to theme T.

[0038] Thus, in this service, all statements C made by attendees A1 through A4 who attended meeting M1 are stored in the database on server 1. In addition, attendees A are classified based on multiple criteria such as their experience, the amount they speak, and theme matching. This allows organizer H to select candidates for attendee A at the upcoming meeting M2 by referring to the attendee information stored in the server 1 database, enabling more appropriate selection. Specifically, it becomes possible to select attendees based on more detailed criteria, such as, "For the next meeting, we want flexible opinions on new business ventures, so we want to select attendees mainly from outside the development department."

[0039] Next, with reference to Figure 4, we will describe the configuration of an information processing system to which a server 1, according to one embodiment of the information processing device of the present invention, is applied, which enables the provision of the service described above. Figure 4 shows an example of the configuration of an information processing system to which a server according to one embodiment of the present invention is applied.

[0040] The information processing system shown in Figure 4 is configured to include a server 1, an attendee terminal 2, an organizer terminal 3, and a microphone device 4. Server 1, attendee terminal 2, organizer terminal 3, and microphone device 4 are interconnected via a designated network NW such as the Internet or LAN (Local Area Network).

[0041] Server 1 is an information processing device managed by a service provider (not shown). Server 1 performs various processes to realize this service while communicating with attendee terminals 2, organizer terminal 3, and microphone device 4 as needed.

[0042] Attendee terminal 2 is an information processing device operated by attendee A, who is attending meeting M remotely. Attendee terminal 2 can consist of a smartphone, tablet, personal computer, etc. Note that only one attendee terminal 2 is depicted in Figure 4, but this is a simplification to aid in understanding the explanation. In other words, there can actually be as many attendee terminals 2 as there are attendees A who are attending meeting M remotely.

[0043] Organizer terminal 3 is an information processing device operated by the organizer H of conference M. Organizer terminal 3 consists of a smartphone, tablet, personal computer, etc.

[0044] Here, for example, as in the example in Figure 1 above, if there are attendees A of meeting M who attend in person rather than remotely, a microphone device 4 is installed in the meeting room where meeting M is held. The microphone device 4 is not particularly limited as long as it has the function of inputting the voice of attendee A of meeting M to server 1. For this reason, the microphone device 4 may or may not be connected to a network NW. If the microphone device 4 is connected to a network NW, the voice data V is input to server 1 via the network NW. If the microphone device 4 is not connected to a network NW, the voice data V is input to server 1 via a predetermined storage medium. Examples of predetermined storage mediums include voice recorders and smartphones.

[0045] Figure 5 is a block diagram showing an example of the server hardware configuration in the information processing system shown in Figure 4.

[0046] Server 1 comprises a CPU (Central Processing Unit) 11, ROM (Read Only Memory) 12, RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20.

[0047] The CPU 11 executes various processes according to the program recorded in the ROM 12 or the program loaded from the storage unit 18 into the RAM 13. RAM13 also stores data and other information necessary for the CPU11 to perform various processes.

[0048] The CPU 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output interface 15 is also connected to this bus 14. An input / output interface 15 is connected to an input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20.

[0049] The input unit 16 is configured, for example, with a keyboard, and is used to input various types of information. The output unit 17 consists of a display such as an LCD and a speaker, and outputs various information as images and sounds. The memory unit 18 is composed of DRAM (Dynamic Random Access Memory) and stores various types of data. The communication unit 19 communicates with other devices (for example, the attendee terminal 2, the organizer terminal 3, and the microphone device 4 in Figure 2) via a network NW, including the Internet.

[0050] A removable media 40, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately mounted in the drive 20. Programs read from the removable media 40 by the drive 20 are installed in the storage unit 18 as needed. Furthermore, the removable media 40 can store various types of data stored in the storage unit 18, just as the storage unit 18 does.

[0051] Although not shown in the diagram, attendee terminal 2 and organizer terminal 3 in Figure 4 can have essentially the same hardware configuration as shown in Figure 5. Therefore, a description of the hardware configuration of attendee terminal 2 and organizer terminal 3 will be omitted.

[0052] Through the collaboration of various hardware and software components of the information processing system in Figure 4, including Server 1 in Figure 5, it becomes possible to execute various processes, including candidate proposal processing. As a result, the service provider can provide the aforementioned service to the organizer H. "Candidate proposal processing" refers to the process of supporting the selection of candidates to attend the upcoming conference M. The following describes the functional configuration for executing candidate proposal processing, which is performed on Server 1 in Figure 5, which constitutes the information processing system in Figure 4, with reference to Figure 6.

[0053] Figure 6 is a functional block diagram showing an example of the functional configuration for executing candidate proposal processing among the functional configurations of the server in Figure 5 that constitute the information processing system in Figure 4.

[0054] As shown in Figure 6, when Server 1 in Figure 5 performs candidate proposal processing, the following functions operate in CPU 11: voice acquisition unit 101, meeting determination unit 102, information management unit 103, attendee classification unit 104, meeting requirements acquisition unit 105, and candidate proposal unit 106.

[0055] Furthermore, one area of ​​the storage unit 18 of server 1 is provided with an attendee database 181 and a meeting database 182. Attendee DB181 stores attendee information for m attendees A who attended at least one of the k meetings M held. Meeting DB182 stores meeting information for the k meetings M held.

[0056] The audio acquisition unit 101 acquires the audio data V of each of the multiple attendees A who attended meeting M. The audio data V acquired by the audio acquisition unit 101 is stored and managed in the meeting database 182.

[0057] The meeting determination unit 102 determines the content of meeting M. The meeting determination unit 102 also identifies the attendee A who made statement C in meeting M. Specifically, the meeting judgment unit 102 functions as follows: a statement extraction unit 121, a statement analysis unit 122, a theme estimation unit 123, and a speaker identification unit 124.

[0058] The speech extraction unit 121 extracts one or more speeches C contained in the audio data V acquired by the audio acquisition unit 101.

[0059] The speech analysis unit 122 analyzes the n speeches C extracted by the speech extraction unit 121. The method used by the speech analysis unit 122 to analyze speeches C is not particularly limited. As mentioned above, for example, analysis using text mining or analysis targeting the audio data V itself may be performed.

[0060] The theme estimation unit 123 estimates one or more themes T for meeting M based on the results of the analysis by the speech analysis unit 122. The theme estimation by the theme estimation unit 123 can be performed on a meeting M basis, but if meeting M is composed of multiple sections, it can also be performed on a section basis. Furthermore, themes T can also be estimated on a time-slot basis.

[0061] The speaker identification unit 124 identifies the attendee A who made each of the n statements C contained in the audio data V. The method used by the speaker identification unit 124 to identify the attendee A who made statement C is not particularly limited. As described above, the speaker can also be identified based on the recognition result of the audio data V and data indicating the characteristics of the attendee A's voice, which has been acquired in advance.

[0062] The Information Management Department 103 manages the information stored in the Attendees Database 181 and the Meeting Database 182, respectively. Specifically, the Information Management Unit 103 associates the respective attendee IDs of attendees A identified as speakers with the respective themes T of meetings M1 through Mk and the statements C, and manages this as attendee information. As described above, the attendee information is stored and managed in the attendee database 181. Furthermore, the information management unit 103 stores and manages the audio data V acquired by the audio acquisition unit 101 in the conference database 182.

[0063] The attendee classification unit 104 classifies attendee A based on the analysis results of statement C by the statement analysis unit 122. Specifically, the attendee classification unit 104 classifies attendee A based on factors such as the amount of speaking and the degree of theme matching.

[0064] The meeting requirements acquisition unit 105 acquires meeting requirements information. In other words, the meeting requirements acquisition unit 105 acquires the meeting requirements for the scheduled meeting M, which have been entered into the organizer terminal 3, as meeting requirements information. Furthermore, the meeting requirements acquisition unit 105 acquires the meeting requirements for the scheduled meeting M as meeting requirements information based on the results of the analysis performed by the speech analysis unit 122 described above. Specifically, for example, towards the end of a meeting M, the organizer H, who attended meeting M as attendee A, may mention the content of the next meeting M that is scheduled to be held. In such a case, the meeting requirements acquisition unit 105 acquires the meeting requirements for the scheduled meeting M as meeting requirements information, based on the results of the analysis by the statement analysis unit 122 on the statement C made by organizer H.

[0065] The candidate proposal unit 106 proposes one or more suitable candidates as attendees A for the newly scheduled meeting M, based on the attendee information and the meeting requirements information. Specifically, the candidate suggestion unit 106 functions as a candidate extraction unit 161 and a display control unit 162. The candidate selection unit 161 selects one or more suitable candidates to be attendees A of the newly scheduled meeting M, based on the attendee information managed by the information management unit 103 and the meeting requirements information. The display control unit 162 executes control to display one or more candidates selected by the candidate selection unit 161 on the organizer terminal 3.

[0066] Because the information processing device in Figure 5 has the functional configuration described above, the organizer H of meeting M can request the attendance of the most suitable person A for meeting M without having to worry about selecting the attendees A for meeting M. As a result, the time cost of selecting attendees A for meeting M can be reduced, and a more productive meeting can be held with a more accurate selection of participants. Furthermore, it can be used to improve performance evaluations within the organization.

[0067] Although one embodiment of the present invention has been described above, the present invention is not limited to the embodiments described above, and any modifications, improvements, etc. that can achieve the objectives of the present invention are considered to be included in the present invention.

[0068] In the above-described embodiment, the candidate suggestion unit 106 suggests one or more suitable candidates as attendees A for the newly scheduled meeting M based on attendee information and meeting requirements information. However, it is preferable that the attendee information includes the information described below.

[0069] For example, the attendee information may include evaluations of attendee A (e.g., attendee A1) from other attendees A (e.g., attendee A2). Specifically, for example, attendee A of meeting M will have information collected on their evaluation of speakers at each stage of meeting M. In the following explanation, it will be assumed that each attendee A has been given a button (hereinafter referred to as the "Like button") in advance, which, when pressed, indicates that a positive evaluation has been given to the speaker immediately before. In other words, attendee A presses the "Like" button when they understand that a speaker's statement in meeting M is good. The fact that attendee A has pressed the "Like" button is collected by server 1, and server 1 manages information indicating this as part of the attendee information.

[0070] For example, suppose the theme T of the current meeting M is brainstorming on a predetermined topic, and the theme T of the next meeting M will be a deeper exploration of the multiple ideas generated as a result of the brainstorming. Therefore, the next meeting M will be conducted separately for each of the multiple ideas generated as a result of the brainstorming. In such a case, the candidate proposal unit 106 can propose attendee A1, who spoke about an idea, and attendee A2, who clicked the "like" button when A1 spoke, as attendees A for a meeting M with theme T focusing on a particular idea. This is expected to liven up the meeting M with theme T focusing on a particular idea, and to ensure that the meeting M on theme T is conducted more effectively.

[0071] For example, suppose the theme T of meeting M is "to summarize the discussions so far and confirm agreement." In this case, it is possible to propose someone who appropriately clicks the "like" button for other attendees A as attendee A. This prevents someone who respects the opinions of other attendees A or does not agree with those who seem to be a good compromise from being proposed as a candidate for attendee A in meeting M on theme T.

[0072] For example, suppose a newly scheduled meeting M requires careful discussion to ensure no points are overlooked. In such a case, the candidate proposal unit 106 can propose multiple attendees A, who are not necessarily in a relationship to easily "like" each other, as candidates. This is expected to lead to more careful discussion as participants will appropriately and critically evaluate each other's opinions. In this way, the candidate suggestion unit 106 can suggest one or more candidates who are more suitable as attendee A for the newly scheduled meeting M, based on attendee information that includes information indicating that attendee A has pressed the "like" button.

[0073] Furthermore, the display on participant terminal 2 may show that the "Like" button was pressed during meeting M. This allows participant A to share the fact that positive evaluations are being given to each other. In other words, since there is no need to share evaluations by means of voice, for example, meeting M will proceed more smoothly. Moreover, since speakers can see that their comments are being positively evaluated, it is expected that they will be encouraged to speak more, and that meeting M will be a better experience.

[0074] Furthermore, unlike the example above, the fact that the "Like" button was pressed during meeting M may not be shared during the meeting itself. For example, the "Like" button might be pressed based on consideration of the relationship between attendees A (e.g., superior and subordinate). By not sharing the fact that the "Like" button was pressed during meeting M, it is expected that the frequency of such presses will decrease.

[0075] Furthermore, the "Like" button does not have to be used during meeting M, but may be pressed after meeting M has ended. This allows for an evaluation of whether or not attendee A of the current meeting M was appropriate, and the candidate proposal unit 106 can propose one or more candidates who are more suitable as attendee A for the newly scheduled meeting M. Furthermore, the "Like" button may also be able to retrieve information indicating which participant the positive evaluation is for. This would allow a positive evaluation of participant A, who is not speaking at that time, if the "Like" button is pressed during meeting M. Alternatively, if the button is pressed after meeting M has ended, a positive evaluation of any participant A could be given after an overall assessment of meeting M has concluded. In this way, by using the "Like" button, the candidate suggestion unit 106 can use the subjective evaluations of participant A at meeting M to suggest one or more candidates who are more suitable as attendees A at the newly scheduled meeting M. That is, for example,

[0076] Furthermore, the evaluation of a speaker may be based not only on the "like" button, but also on other elements such as the following: In other words, for example, the level of engagement in meeting M can be evaluated by changes in the volume of voices. This makes it possible to evaluate the speakers. Furthermore, for example, fluctuations in voice tone can be used to evaluate whether statement C was perceived positively or negatively in meeting M. For example, in meeting M, attendee A who speaks regularly can receive a good evaluation. That is, attendee A who can speak C regardless of whether it is the first or second half of meeting M, or the circumstances, can receive a good evaluation.

[0077] Furthermore, in the embodiment described above, the relationship between the theme T of meeting M1 and the content of statement C can be expressed using an indicator called "theme matching degree" or "theme deviation degree," but the following indicators can also be adopted. That is, for example, in the embodiment described above, those with a high degree of relationship between theme T and the content of statement C were described as having a high degree of theme matching, and those with a low degree of relationship were described as having a high degree of theme deviation. However, these indicators may be used not only as indicators of high or low relevance, but also as separate indicators as follows.

[0078] For example, suppose the theme T of meeting M was "Commercialization of new technology Z." Suppose that at meeting M, participant A made the statement C, "It's not just about commercialization; attention must also be paid to sales channels." If this statement C is considered important, then, although it deviates significantly from the theme, in the context of meeting M, participant A made an important statement C. For example, suppose the theme T of meeting M was "New Technology Z." Suppose that during meeting M, a new idea C was expressed that could lead to the development of new technology Y. Furthermore, suppose various other suggestions C were made during brainstorming. While these suggestions C are highly unrelated to the theme, they can still be considered important. Furthermore, the same applies not only to the degree of theme deviation but also to the degree of theme matching. That is, for example, even if a statement C with a high degree of theme matching is made in the middle of a brainstorming session, the brainstorming session will be interrupted. The degree of theme matching and the degree of theme deviation can be used not only to indicate the degree of relevance, but also as separate indicators to evaluate the situation of meeting M. This allows the candidate proposal unit 106 to propose one or more candidates who are more suitable as attendees A for the newly held meeting M by evaluating attendees A who can make statements C appropriate to the situation of meeting M.

[0079] Furthermore, for example, the evaluation criteria for speakers and attendees A can be varied depending on the stage of meeting M (e.g., the first half, the second half). That is, for example, in the first half of meeting M, those who can speak about the situation or develop the content of the discussions from the previous meeting M can receive a good evaluation. For example, in the latter half of Meeting M, those who can summarize the content of the discussions so far can be given a good evaluation.

[0080] For example, the candidate proposal unit 106 can propose candidates for each stage (e.g., the first half, the second half) of a newly convened meeting M. That is, for example, it can propose someone who can summarize the discussions as a candidate to attend the second half of meeting M.

[0081] For example, if the schedule for a newly scheduled meeting M is set in advance, the candidate proposal unit 106 can propose individuals who are able to participate in that meeting M as candidates. In other words, the candidate suggestion unit 106 may suggest candidates based not only on the theme T of meeting M, but also on information indicating other conditions such as the schedule. Furthermore, the server 1 may output the theme T that is expected to allow meeting M to proceed more appropriately with the candidates suggested by the candidate suggestion unit 106. Specifically, for example, Server 1 can propose many suitable candidates for the aforementioned brainstorming session based on the pre-configured schedule of a new meeting M, and can also propose brainstorming as theme T from among the themes T to be discussed in the future. This will allow for the selection of appropriate themes T within the schedule of several new meetings M to be held in the future, thus enabling more efficient progress for all the meetings M as a whole.

[0082] For example, in the embodiments described above, themes T are estimated on a meeting-by-meeting, department-by-department, and time-slot basis, but the invention is not limited to these. For example, themes T can also be estimated on a statement-by-statement basis. This allows for the estimation of themes T even for unexpected statements, which can be useful in selecting attendees A for future meetings M.

[0083] For example, in the example in Figure 1, each of the statements C1 to C5 from attendees A1 to A5 is shown as a single statement, but this is a simplification to aid in understanding the explanation. In reality, each of the statements C1 to C5 from attendees A1 to A5 could actually be multiple.

[0084] For example, in the example in Figure 2, the amount of participation and the degree of theme matching of participant A are used as indicators for classifying participant A. Similarly, in the example in Figure 3, the experience, amount of participation, and degree of theme matching of participant A are used as indicators for classifying participant A. However, the system is not limited to these; additional indicators may be established, or participant A may be classified using alternative indicators. Furthermore, there is no particular limit to the number of indicators; there may be one, or there may be four or more. For example, participant A's status, influence, gender, age, etc., can be used as indicators. This makes it possible to select participants not only based on the quantity and quality of their past contributions, but also by considering their experience, status, influence, gender, age, etc. As a result, more accurate participant selection becomes possible, leading to more productive meetings.

[0085] Furthermore, the system configuration shown in Figure 4 and the hardware configuration of Server 1 shown in Figure 5 are merely illustrative examples for achieving the objectives of the present invention and are not particularly limited.

[0086] Furthermore, the functional block diagram shown in Figure 6 is merely illustrative and not particularly limiting. In other words, it is sufficient that the information processing system in Figure 4 has the functionality to execute the candidate proposal process described above as a whole, and the functional blocks and databases used to realize this functionality are not particularly limited to the example in Figure 6.

[0087] Furthermore, the location of the functional blocks and database is not limited to Figure 6, but can be any location. In the example shown in Figure 6, the candidate proposal process is controlled by the CPU 11 of Server 1 in Figure 5, which constitutes the information processing system in Figure 4, but the system is not limited to this configuration. For example, at least a portion of the functional blocks and database located on Server 1 may be provided on Attendees Terminal 2, Organizers Terminal 3, Microphone Device 4, or other information processing devices not shown.

[0088] Furthermore, the series of processes described above can be executed by hardware or by software. Furthermore, a single functional block may consist of hardware alone, software alone, or a combination of both.

[0089] When a series of processes are executed by software, the programs that make up that software are installed on a computer or other device from a network or storage medium. The computer may be a computer that is built into dedicated hardware. Furthermore, a computer can be any computer capable of performing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0090] Such recording media containing programs may consist not only of removable media (not shown) distributed separately from the main unit of the device to provide the program to the user, but also of recording media provided to the user in a state where they are pre-installed in the main unit of the device.

[0091] In this specification, the step of describing a program to be recorded on a recording medium includes not only processes that are performed chronologically in that order, but also processes that are not necessarily performed chronologically, but are executed in parallel or individually.

[0092] In summary, the information processing device to which the present invention applies only needs to have the following configuration, and can take various forms. In other words, the information processing apparatus to which the present invention is applied is: An acquisition means (for example, the audio acquisition unit 101 in Figure 6) for acquiring audio data (for example, the audio data V in Figure 1) of a meeting (for example, the 1st to the kth meeting M in Figure 1), A means for identifying the speaker (for example, attendee A in Figure 1) of each predetermined unit of speech (for example, the n speeches C mentioned above) contained in the audio data acquired by the acquisition means (for example, the meeting determination unit 102 in Figure 6), A management means (for example, the information management unit 103 in Figure 6) manages the information indicating the speaker identified by the identification means (for example, the attendee ID in Figure 2), the theme of the meeting (for example, theme T in Figure 1), and the statement as first information (for example, attendee information in Figure 1), Based on the first information managed by the management means and the second information (for example, the meeting requirements information in Figure 1) which includes at least the theme of the newly scheduled meeting (for example, the kth meeting M in Figure 1), a proposal means (for example, the candidate proposal unit 106 in Figure 6) proposes one or more suitable candidates to attend the meeting. It is equipped with.

[0093] In other words, when audio data from a meeting is acquired, predetermined units of speech contained in that audio data are extracted, and the speaker of each predetermined unit of speech is identified. Then, the information indicating the identified speaker is associated with the meeting's theme and the content of the speech, and managed accordingly. When a new meeting is scheduled, one or more suitable candidates are proposed as attendees of the new meeting based on the managed information and information that includes at least the theme of the new meeting. This allows conference organizers to invite the most suitable individuals to attend the conference without having to worry about selecting attendees. As a result, the time cost of selecting meeting attendees can be reduced, and the risk of selection errors can be minimized, leading to more productive meetings.

[0094] Furthermore, the system includes an estimation means (for example, the theme estimation unit 123 in Figure 6) that estimates one or more themes of the meeting based on the results of analyzing the audio data acquired by the acquisition means. The aforementioned management means is The information indicating the speaker can be managed as the first information by associating it with the theme estimated by the estimation means and the statement itself.

[0095] In other words, the theme of the meeting is estimated based on the audio data of the meeting. Then, the estimated meeting theme and the content of the statements are associated with information indicating the identified speaker and managed accordingly. This allows for the management of discussions by associating them with the topic (which may differ from the original topic) even if the discussion temporarily deviates from the original topic during the meeting. As a result, it can handle real-world meetings that don't always proceed according to the original theme. Specifically, for example, if the original theme of a meeting is "invention discovery," the conversation might temporarily shift to the theme of "fundraising." Even in such cases, comments about "fundraising" will not be associated with the meeting's theme of "invention discovery."

[0096] Furthermore, the means of control are, The information indicating the speaker is further associated with information regarding the speaker's experience (for example, the experience of the attendee included in the attendee information in Figure 3), and managed as the first information. The suggestion means can suggest one or more suitable candidates to attend the newly scheduled meeting, based on the first information managed by the management means and the second information.

[0097] In other words, information identifying a speaker is associated with and managed information about that speaker's experience. Then, when a new meeting is scheduled, one or more suitable candidates are proposed as attendees of the new meeting based on the managed information and information that includes at least the theme of the new meeting. This makes it possible to select speakers for meetings based not only on the quantity and quality of their contributions, but also on their experience. As a result, it becomes possible to achieve even more accurate personnel selection. [Explanation of Symbols]

[0098] 1...Server, 2...Attendee terminals, 3...Organizer terminal, 4...Microphone device, 11...CPU, 12...ROM, 13...RAM, 14...Bus, 15...Input / Output interface, 16...Input unit, 17...Output unit, 18...Storage unit, 19...Communication unit, 20...Drive, 40...Removable media, 101...Audio acquisition unit, 102...Conference judgment unit, 103...Information management unit, 104... • Attendee Classification Unit, 105...Meeting Requirements Acquisition Unit, 106...Candidate Proposal Unit, 121...Statement Extraction Unit, 122...Statement Analysis Unit, 123...Theme Estimation Unit, 124...Speaker Identification Unit, 161...Candidate Extraction Unit, 162...Display Control Unit, 181...Attendee DB, 182...Meeting DB, A...Attendees, H...Organizers, M...Meeting, V...Audio Data, C...Statements, T...Themes, NW...Network

Claims

1. A means of acquiring audio data from a meeting, A means for identifying the speaker of each utterance in a predetermined unit contained in the audio data acquired by the acquisition means, A management means manages the information indicating the speaker identified by the aforementioned identification means, associating it with the theme of the meeting and the aforementioned statement, as first information. A proposal means that proposes one or more suitable candidates as attendees of a meeting based on the first information managed by the management means and the second information which includes at least the theme of a newly to be held meeting, An information processing device equipped with the following features.

2. The system further comprises estimation means for estimating one or more themes of the meeting based on the results of analyzing the audio data acquired by the acquisition means, The aforementioned management means is The information indicating the speaker is managed as the first information by associating the theme estimated by the estimation means with the statement. The information processing apparatus according to claim 1.

3. The aforementioned management means is The information identifying the speaker is further associated with information regarding the speaker's experience and managed as the first information. The proposal means proposes one or more suitable candidates to attend the newly scheduled meeting, based on the first information managed by the management means and the second information. The information processing apparatus according to claim 1 or 2.

Images