Web conference assistance system

Abstract

[Problem] To provide a web conference assistance system in which, regarding presentation material or material used in a conference or the like, a computer can automatically propose the material itself and a candidate for each page of the material during a web conference. [Solution] Provided is a web conference assistance system in which a computer performs: a conversation analysis step for analyzing a voice in a conversation in a web conference to obtain analyzed terms that are terms included in the conversation; a related material selection step for selecting, on the basis of the analyzed terms, related material that is related to the conversation or a corresponding page of the material related to the conversation; and a related material display step for displaying an image related to the related material on a conference screen of the web conference.

Description

Web conference support system

[0001] The present invention relates to a web conference support system.

[0002] In recent years, conferences, presentations, and lectures using the web have become increasingly popular. During these web conferences, when a user wants to share a document with all participants, they search for the document and then issue a share command using a share icon. In this way, the document can be shared with other participants during a conference, presentation, or lecture. However, users may find it time-consuming to search for the document, or may not be able to find the desired document.

[0003] Japanese Patent No. 7345909 describes a presentation evaluation system that finds related terms for each page of a presentation document. Finding related terms for each page of a presentation in this way is known.

[0004] Patent No. 7345909

[0005] An object of the present invention is to provide a web conference support system that allows a computer to automatically propose materials and page candidates for presentation materials and materials used in conferences during a web conference.

[0006] The above problem is based on the knowledge that, for example, by using the technology described in the above-mentioned Patent Publication No. 7345909, if related terms for each page of presentation materials are found and stored, a computer can automatically find materials and relevant pages of materials related to conversations during the meeting by analyzing the terms used during the meeting.

[0007] A first invention relates to a web conference support system. In this system, a computer executes a conversation analysis step, a related material selection step, and a related material display step. The computer executes a web conference support method, which includes the conversation analysis step, the related material selection step, and the related material display step. A processor may execute the conversation analysis process, the related material selection process, and the related material display process based on instructions from a program.

[0008] The conversation analysis step is a step in which the computer analyzes the audio of the conversation in the web conference and obtains analyzed terms, which are terms contained in the conversation.The related material selection step is a step in which the computer selects related materials that are related to the conversation or relevant pages of materials related to the conversation based on the analyzed terms.The related material display step is a step in which the computer displays images of related materials on the conference screen of the web conference.This web conference support system allows the computer to automatically analyze the audio during the web conference and automatically suggest candidate materials related to the conference.

[0009] In a preferred example of the web conference support system, a potentially relevant material storage step, in which a computer stores potentially relevant materials, which are materials that may be relevant to the conversation, in a computer memory, is performed before the relevant material selection step. In this example, the relevant material selection step involves the computer selecting relevant materials from the potentially relevant materials based on analysis terms. In this example, materials that may be shared during a web conference can be registered in advance. This narrows down the candidates for shared materials, speeds up computer processing, and prevents irrelevant materials from being suggested.

[0010] A preferred example of the web conference support system further includes a material-related term storage step in which, prior to the related material selection step, the computer associates material-related terms, which are terms related to potentially relevant materials, obtained based on one or more of conversations about potentially relevant materials, presentations using potentially relevant materials, and terms contained in the potentially relevant materials, with the potentially relevant materials and stores them in a storage unit. Then, in the related material selection step, the computer selects relevant materials from the potentially relevant materials based on the analyzed terms and the material-related terms. This system pre-analyzes conversations based on presentation materials, analyzes and stores the terms used in each page and each presentation material, and can use them to select relevant materials. This results in more appropriate candidates for shared materials, speeds up computer processing, and prevents irrelevant materials from being suggested.

[0011] In a preferred example of a web conference support system, the related material display process includes a shared material candidate display process in which a computer displays an image of the related material as a proposed shared material candidate image on the conference screen of the web conference on the terminal of the speaker of the web conference, and a material sharing process in which, when approval input is made from the speaker's terminal, the image of the related material is shared on the conference screen of the web conference on the terminal of the participant of the web conference.

[0012] A second invention relates to a program. The program causes a computer (processor) to execute a web conference support method including a conversation analysis step, a related material selection step, and a related material display step. In other words, the program causes the computer to function as a web conference support system. The program also causes the computer (processor) to execute a conversation analysis process, a related material selection process, and a related material display process. The program may be recorded on a non-transitory information recording medium.

[0013] The present invention can provide a web conference support system that allows a computer to automatically suggest the cost itself and candidates for each page during a web conference.

[0014] Fig. 1 is a block diagram illustrating a web conference support system. Fig. 2 is a block diagram illustrating a system for obtaining material-related terms. Fig. 3 is a flowchart illustrating an example of a computer-based web conference support method. Fig. 4 is a conceptual diagram showing an example of a web conference screen.

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the embodiments described below, and includes appropriate modifications of the embodiments below within the scope obvious to those skilled in the art.

[0016] FIG. 1 is a block diagram illustrating an example of a web conference support system. As shown in FIG. 1, this web conference support system 1 includes a voice input unit 3, a conversation analysis unit 5, a related material selection unit 7, and a related material display unit 13. The example of FIG. 1 further includes a material-related term storage unit 9 and a potentially related material storage unit 11. The conversation analysis unit 5, the related material selection unit 7, and the related material display unit 13 perform conversation analysis processing, related material selection processing, and related material display processing, respectively. This system typically includes a server and multiple terminals. In other words, this system may be implemented by one or multiple information processing terminals (computers).

[0017] A computer has an input unit, an output unit, a control unit, a calculation unit, and a memory unit, and each element is connected by a bus or the like to enable the exchange of information. For example, the memory unit may store a control program or various information. When predetermined information is input from the input unit, the control unit reads the control program stored in the memory unit. The control unit then reads the information stored in the memory unit as appropriate and transmits it to the calculation unit. The control unit also transmits the input information as appropriate to the calculation unit. The calculation unit performs calculation processing using the received various information and stores it in the memory unit. The control unit reads the calculation results stored in the memory unit and outputs them from the output unit. In this way, various processes and steps are performed. Each unit or means executes these various processes. A computer may have a processor, and the processor may realize various functions and steps. A computer may be standalone. A computer may have some of its functions distributed between a server and a terminal. In this case, it is preferable that the server and the terminal can exchange information via a network such as the Internet or an intranet. The computer may include a processor and a memory coupled to the processor. The memory may store instructions that, when executed by the processor, cause the computer to perform various processes or function as various elements. The computer may be provided with various training data to construct a learning model and perform various calculations through machine learning. In this case, the computer may perform various analyses using the learning model created through machine learning and deep learning of AI (artificial intelligence).

[0018] Fig. 2 is a block diagram illustrating a system for obtaining document-related terms. In the example of Fig. 2, the system for obtaining document-related terms includes a document-related term storage unit 9, a potentially related document storage unit 11, and a document-related term analysis unit 15. The document-related term storage unit 9, the potentially related document storage unit 11, and the document-related term analysis unit 15 execute a document-related term storage process, a potentially related document storage process, and a document-related term analysis process, respectively.

[0019] Next, an example of a web conference support method will be described. Web conferences include meetings, sales, marketing, explanations, lectures, and speeches conducted via the web. In a web conference, there is a speaker who speaks and participants who listen to the speaker. The speaker's terminal and the participant's terminal are configured to share audio and images via a web server or one of the terminals. The speaker's terminal and the participant's terminal are also simply referred to as the speaker and the participant.

[0020] Potentially relevant material storage process. In this example, a computer performs a process for storing potentially relevant materials. Potentially relevant materials are samples stored in the potentially relevant material storage unit 11 of the computer as materials potentially relevant to the conversation. For example, a web conference organizer (e.g., a speaker, teacher, instructor, or conference proposer) stores presentation materials (e.g., materials in PowerPoint® format) and reference materials (e.g., materials in PDF®, Word®, or Excel® format) on their terminal. In this case, the system may, for example, display an icon for the potentially relevant material on the display unit, and when a material is dragged onto the icon, the material is stored as potentially relevant material. The potentially relevant material may be one page or multiple pages. In this way, potentially relevant materials are stored in the potentially relevant material storage unit 11. In this example, materials that may be shared during a web conference can be registered in advance. This narrows down the number of potential shared materials, speeds up computer processing, and prevents irrelevant materials from being proposed.

[0021] Document-Related Term Storage Process: For potentially related materials, document-related terms, which are terms related to the potentially related materials, may be analyzed in advance based on one or more of conversations about each potentially related material, multiple presentations about each potentially related material, and terms contained in the potentially related material, and multiple document-related terms for each potentially related material and each page of the potentially related material may be stored. This method of determining document-related terms is publicly known, for example, as described in Japanese Patent No. 7345909. In this case, document-related terms may be stored in association with the identification information of each potentially related material and the identification information of each page of the potentially related material. In this case, the document-related term storage unit 9 may be pre-existing in the terminal. Alternatively, the system may read potentially related materials from the potentially related material storage unit 11 and read information for creating the document-related term storage unit 9 (each related term (document-related term) associated with the identification information of each potentially related material and the identification information of each page of the potentially related material) from the terminal's storage unit, server storage unit, or web storage unit, and create the document-related term storage unit 9. This is an example of a material-related term storage process when the material-related term storage unit 9 is present. For example, if a certain material was shared in a past web conference, such as a past conversation, presentation, explanation, lecture, or speech, a trained model may be constructed using the analysis terms used at that time and the shared material as training data. In this case, when a conversation is analyzed and new analysis terms are input into the trained model, identification information related to the material may be output. In this case, instead of step S103, which will be described later, a process of inputting analysis terms into the trained model may be performed. Furthermore, if a certain material was shared in a past web conference, such as a past conversation, presentation, explanation, lecture, or speech, the analysis terms used at that time may be stored in the material-related term storage unit 9 as material-related terms for the shared material.

[0022] On the other hand, the computer may automatically perform a material-related term analysis process after the potentially related materials are stored in the potentially related material storage unit 11. For example, the material-related term analysis processing unit 15 reads the potentially related materials from the potentially related material storage unit 11. Then, for example, the material-related term analysis processing unit 15 may analyze terms included in the potentially related materials to analyze material-related terms, which are terms related to the potentially related materials or each page of the potentially related materials. The material-related term analysis processing unit 15 may also crawl information on the web related to the potentially related materials to analyze the material-related terms. Furthermore, the material-related term analysis processing unit 15 may perform audio analysis of past conversations or presentations stored in association with the potentially related materials (or use the audio-analyzed past conversations or presentations) to analyze material-related terms related to the potentially related materials or each page of the potentially related materials. When multiple material-related terms are stored in association with the identification information of the potentially related materials, each material-related term may be assigned a coefficient according to its relevance to the potentially related materials. The material-related term analysis processing unit 15 stores material-related terms in association with identification information of possibly related materials or pages of potentially related materials in the material-related term storage unit 9. This makes it possible to find identification information of possibly related materials or pages of potentially related materials associated with the material-related terms in the material-related term storage unit 9, and to read out the relevant possibly related materials or pages of potentially related materials from the relevant material storage unit 11 using the identification information of the relevant materials or pages of the relevant materials.

[0023] For example, the potentially related materials storage unit 11 stores PowerPoint (registered trademark) A, PDF (registered trademark) document B, and PDF (registered trademark) document C as potentially related materials. The material-related term storage unit 9 stores, for example, the identification information of material A: ID Ａ , identification information of each page of material A: ID Ａ．ｐ , identification information of material B and material C: ID Ｂ , ID Ｃ , and associated with each identification information, one or more material-related terms (RW Ａ１ , R.W. Ａ２ , R.W. Ａ３etc.) are stored.

[0024] Conversation Analysis Step (S101) The following steps will be described with reference to Fig. 3. Fig. 3 is a flowchart for explaining an example of a computer-based web conference support method. In Fig. 3, S denotes a step. In the example of Fig. 3, a process for storing potentially relevant materials and a process for storing material-related terms may have been performed in advance.

[0025] For example, as a web conference progresses, a speaker speaks. Then, a voice input unit 3, such as a microphone on the speaker's device, converts the analog voice into digital information that a computer can process, obtaining the digital conversation. The conversation analysis unit 5 then analyzes the audio of the web conference conversation and obtains analysis terms, which are terms contained in the conversation. The system may store document-related terms and related or similar terms in advance, and determine that an analysis term exists if the analysis term matches these terms. If an analysis term does not exist, the computer may return to the conversation analysis process. If an analysis term exists, the computer may proceed to the next step. The obtained analysis terms may be stored in a storage unit as appropriate. The conversation analysis unit 5 may identify terms characteristic of the conversation in real time and use them as analysis terms. For example, the computer may build a trained model using training data from conferences, lectures, and presentations related to potentially relevant documents, terms contained in the related materials, and past analysis term candidates. The conversation analysis unit 5 may then input the speech analysis results into the trained model to obtain analyzed terms in relation to the real-time conversation.

[0026] Related Material Selection Process: The related material selection unit 7 receives one or more analysis terms. The related material selection unit 7 may perform a process of matching one or more analysis terms with the material-related terms stored in the material-related term storage unit 9 (S103). This process may, for example, be performed repeatedly for all material-related terms to determine whether a material-related term matches one or more analysis terms. If there are no material-related terms that match the analysis term, the process may return to the conversation analysis process. If there is only one material-related term that matches the analysis term, the process may proceed to the next step. The related material selection unit 7 may also perform a process of matching one or more analysis terms or their synonyms with the material-related term by appropriately referencing a thesaurus. For example, if multiple material-related terms are stored in association with the identification information of potentially relevant materials, the identification information of the most relevant potentially relevant material may be determined using the number of matching material-related terms or the sum of the coefficients of the matching material-related terms. In the example of Figure 3, the process of determining whether a material-related term matches one or more analysis terms is performed. However, the method of obtaining identification information of potentially relevant materials based on analysis terms is not limited to this.

[0027] For example, the material-related term storage unit 9 stores the identification information of material B: ID Ｂ In relation to this, the material-related term ID Ｂ１ , ID Ｂ２ , ID Ｂ３ , ID Ｂ４ , are stored as coefficients 10, 5, 3, and 1, respectively. Then, for example, the analysis term at a certain timing is Ｂ１ , and ID Ｂ３ In this case, the related material selection unit 7 determines whether the analysis term is the same as or similar to ID Ｂ１ and ID Ｂ３ The material-related terms in the material-related term storage unit 9 are analyzed using Ｂ１ , and ID Ｂ３ Then, for example, if the threshold for selecting related materials is 8 points, the sum of the coefficients, 13, is compared with the threshold, and the identification information: ID Ｂ Also, ID Ｂ１ , and IDＢ３ If either or both of the above match other material-related terms, the computer may obtain identification information of the potentially related material by calculating and comparing the sum of the coefficients for each piece of identification information. By using either of these processes, the computer can obtain identification information of the potentially related material for the analyzed conversation by reading out identification information of the potentially related material that is stored in association with the material-related term in the material-related term storage unit 9 (S104).

[0028] After obtaining the identification information of the potentially related materials, the related material selection unit 7 reads out, as related materials, the potentially related materials stored in association with the identification information of the potentially related materials from the potentially related material storage unit 11. The related materials are materials related to the conversation or pages of materials related to the conversation. In this way, the related material selection unit 7 can select related materials based on the analysis terms (S105).

[0029] Related Material Display Process: The related material selection unit 7 outputs images of the related materials to the related material display unit 13 (S106). The images of the related materials may be images of the related materials themselves or thumbnails (images that allow a user to grasp the related materials without displaying the entire related materials). Thumbnail processing is well known. Typically, related materials are displayed enlarged on the conference screen when shared. The thumbnail images may be images that are half the scale of the related materials when shared. The related material display unit 13 processes images of the related materials on the conference screen of the web conference. Specifically, the related material selection unit 7 outputs necessary information so that thumbnails of the related materials are displayed as candidate materials to be shared on a portion of the web screen on the speaker's monitor. This web conference support system allows a computer to automatically analyze audio during a web conference and automatically suggest candidate materials relevant to the conference.

[0030] In a preferred example of the web conference support system, the related material display step includes a shared material candidate display step in which a computer displays images of related materials as proposed shared material images on the conference screen of the web conference on the terminal of the speaker of the web conference, and a material sharing step in which, when approval is received from the terminal of the speaker, images of related materials are shared on the conference screen of the terminal of the participant of the web conference. In this example, the computer displays candidate shared materials on the terminal of the speaker, and if the speaker approves, shares the materials with the conference participants. This eliminates the need to search for shared materials and allows the user to check the materials before sharing them. This prevents the wrong materials from being shared.

[0031] FIG. 4 is a conceptual diagram showing an example of a web conference screen. FIG. 4(a) shows an example of a web screen before sharing is approved. FIG. 4(b) shows an example of a web screen after sharing is approved. In the example shown in FIG. 4(a), a camera on the speaker's device captures the speaker. The captured image of the speaker is then displayed on the speaker's monitor 21 as well as on the participant's monitor 23. As the speaker continues the conversation (explanation, lecture), a sharing candidate proposed image 25 is displayed. This sharing candidate proposed image 25 is an image related to a related document. The speaker can visually recognize the sharing candidate proposed image 25 while the conversation is ongoing and understand which related document (potentially relevant document) it refers to. An approval icon 27 associated with the sharing candidate proposed image 25 is displayed on the monitor 21. While the speaker is continuing the conversation, the speaker clicks the approval icon 27 using a mouse or the like. This input regarding approval is then made on the speaker's device. The image related to the related document is then shared in the web conference. As shown in FIG. 4(b), the related document is displayed on the monitors of not only the speaker but also the participants. Meanwhile, as the speaker continues the conversation, a new sharing candidate suggestion image 29 is displayed on the speaker's monitor 21. In this way, this system automatically analyzes the conversation and can suggest materials that may be shared in real time, eliminating the need for the speaker to go to the trouble of searching for materials to share.

[0032] A second invention relates to a program. The program causes a computer (processor) to execute a web conference support method including a conversation analysis step, a related material selection step, and a related material display step. In other words, the program causes the computer to function as a web conference support system. The program also causes the computer (processor) to execute a conversation analysis process, a related material selection process, and a related material display process. The program may be recorded on a non-transitory information recording medium. Examples of information recording media include CDs, CD-ROMs, DVDs, USB memory, hard disks, and disks on a server.

[0033] The present invention relates to a web conference support system and can be used in the information-related industry.

[0034] 1 Web conference support system 1 3 Voice input unit 5 Conversation analysis unit 7 Related material selection unit 9 Material-related term storage unit 11 Related material storage unit 13 Related material display unit

Claims

1. A web conference support system in which a computer performs the following steps: a conversation analysis step in which the computer analyzes the audio of a conversation in a web conference and obtains analysis terms that are terms included in the conversation; a related material selection step in which the computer selects, based on the analysis terms, related materials that are materials related to the conversation or pages of materials related to the conversation; and a related material display step in which the computer displays images related to the related materials on the conference screen of the web conference.

2. A web conference support system according to claim 1, wherein the computer performs a potentially relevant material storage step, which is a step of storing potentially relevant materials, which are materials that may be related to the conversation, in a memory unit of the computer, before the relevant material selection step, and the relevant material selection step is a step of selecting the relevant materials from the potentially relevant materials based on the analysis terms.

3. A web conference support system according to claim 2, wherein the computer further includes a material-related term storage step, which is a step of storing, before the related material selection step, material-related terms, which are terms related to the potentially related material obtained based on one or more of a conversation about the potentially related material, a presentation using the potentially related material, and terms included in the potentially related material, in a storage unit in association with the potentially related material; and the related material selection step is a step of selecting the relevant material from the potentially related materials based on the analysis terms and the material-related terms.

4. The web conference support system of claim 1, wherein the related material display step includes: a shared material candidate display step, which is a step of displaying an image of the related material as a proposed image to be shared on the conference screen of the web conference on the terminal of the speaker of the web conference; and a material sharing step, which is a step of sharing the image of the related material on the conference screen of the web conference on the terminal of a participant of the web conference when approval is input from the terminal of the speaker.

5. A program that causes a computer to execute a web conference support method, including: a conversation analysis step of analyzing audio of a conversation in a web conference and obtaining analyzed terms that are terms included in the conversation; a related material selection step of selecting related materials that are materials related to the conversation or pages of materials related to the conversation based on the analyzed terms; and a related material display step of displaying images related to the related materials on the conference screen of the web conference.

6. A non-transitory information recording medium storing the program according to claim 5.

Images