Information processing device, data output method, and data output program

Abstract

An information processing device (10) has an acquisition unit (15A) that acquires utterance text corresponding to an utterance voice input in a conference system (3), a model execution unit (15B) that, by inputting the utterance text to a large-scale language model, causes the large-scale language model to output a script in which the structure of a chart corresponding to minutes in the conference system (3) is described in a markup language, and an output unit (15C) that outputs visualization data in which the structure of the chart corresponding to the minutes in the conference system (3) is visualized on the basis of the script output by the large-scale language model.

Description

Information processing device, data output method, and data output program

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

[0002] In remote meetings such as web conferences and video conferences, if you miss a discussion, there is a psychological hurdle of asking questions and it is difficult to look back at the minutes, making it difficult to keep up with the discussion.

[0003] For these reasons, tools that use AI (Artificial Intelligence) to create meeting minutes are becoming more common. Furthermore, because chart-based minutes make it easier to grasp the main points than text-based minutes, visualization technology has been proposed that uses generative AI to generate images from transcribed text.

[0004] Hylable, [Retrieved July 22, 2024], Internet <URL: https: / / www.hylable.com / > AI Minutes Taker, [Retrieved July 22, 2024], Internet <URL: https: / / gijirokukun.com / >

[0005] However, with the above visualization technology, the image output by the generation AI lacks stability, so there is room for improvement in terms of visualizing meeting minutes in real time.

[0006] Therefore, an object of the present invention is to provide an information processing device, a data output method, and a data output program that can realize real-time visualization of meeting minutes.

[0007] In order to solve the above-mentioned problems and achieve the object, the information processing device of the present invention has an acquisition unit that acquires spoken text corresponding to spoken voice input in a conference system, a model execution unit that inputs the spoken text into a large-scale language model and causes the large-scale language model to output a script in which a chart structure corresponding to minutes of a meeting in the conference system is written in a markup language, and an output unit that outputs visualization data in which the chart structure corresponding to minutes of a meeting in the conference system is visualized based on the script output by the large-scale language model.

[0008] According to the present invention, real-time visualization of meeting minutes can be realized.

[0009] FIG. 1 is a block diagram showing an example of the functional configuration of an information processing device. FIG. 2 is a schematic diagram (1) showing an example of an input / output form of an LLM. FIG. 3 is a diagram (1) showing an example of spoken text. FIG. 4 is a diagram (1) showing an example of a system prompt. FIG. 5 is a diagram (1) showing an example of an output of an LLM. FIG. 6 is a schematic diagram (2) showing an example of an input / output form of an LLM. FIG. 7 is a diagram (2) showing an example of spoken text. FIG. 8 is a diagram (2) showing an example of a system prompt. FIG. 9 is a diagram (2) showing an example of an output of an LLM. FIG. 10 is a diagram (1) showing an example of visualized data. FIG. 11 is a diagram (2) showing an example of visualized data. FIG. 12 is a flowchart showing the procedure of data output processing. FIG. 13 is a diagram showing an example of a hardware configuration.

[0010] Hereinafter, an information processing device, a data output method, and a data output program according to the present disclosure will be described with reference to the accompanying drawings. Note that this embodiment merely illustrates one example or one aspect, and the following description does not limit the structure, action, function, properties, characteristics, methods, uses, etc. according to the present disclosure.

[0011] <Overall Configuration> Fig. 1 is a block diagram showing an example of the functional configuration of an information processing device 10. For example, Fig. 1 shows the information processing device 10 that provides a data output function that inputs spoken text corresponding to speech spoken at a meeting into a large-scale language model and outputs visualized data based on a script that describes the structure of a chart corresponding to the minutes of the meeting.

[0012] In one aspect, the information processing device 10 may be realized by a server device. For example, the information processing device 10 can provide the above-described data output function as a cloud service by executing PaaS (Platform as a Service) type middleware or SaaS (Software as a Service) type application.

[0013] 1, the information processing device 10 can be communicatively connected to the conference system 3 via a network NW. For example, the network NW may be realized by any type of communication network, such as the Internet or a local area network (LAN), whether wired or wireless.

[0014] The conference system 3 is a system that provides a communication function for sharing video, audio, materials, etc. between multiple locations. For example, the conference system 3 may be realized by any type of web conference system, such as an on-premise type, a cloud type, or a browser type. However, the conference system 3 is not limited to this, and may also be realized by a video conference system.

[0015] Note that, although an example in which the data output function is provided as a cloud service is given here, the present invention is not limited to this. For example, the data output function may be provided on-premise. Furthermore, the data output function may be packaged as one function of a service or application provided by a service provider that provides services related to remote meetings.

[0016] In addition, the above-described data output function is implemented as a client-server system, but is not limited to this. For example, the above-described data output function may be provided standalone by causing an application running on a user terminal participating in a conference held by the conference system 3 to execute processing corresponding to the above-described data output function on the user terminal.

[0017] <Configuration of Information Processing Device 10> Next, the functional configuration of the information processing device 10 that provides the above-mentioned data output function will be described. Fig. 1 schematically shows blocks related to the data output function of the information processing device 10. As shown in Fig. 1, the information processing device 10 has a communication control unit 11, a storage unit 13, and a control unit 15. Note that Fig. 1 only shows a selection of functional units related to the above-mentioned data output function, and the information processing device 10 may be provided with functional units other than those shown.

[0018] The communication control unit 11 is a functional unit that controls communication with other devices such as the conference system 3. In one embodiment, the communication control unit 11 can be realized by a network interface card such as a LAN card. In one aspect, the communication control unit 11 receives voice data or text data of utterances inputted by voice in a conference held in the conference system 3, or outputs visualized data outputted by the data output function to user terminals participating in the conference.

[0019] The storage unit 13 is a functional unit that stores various types of data. In one embodiment, the storage unit 13 may be realized by internal, external, or auxiliary storage of the information processing device 10. For example, the storage unit 13 stores an utterance text DB 13A and an output log DB 13B. Note that the respective data in the utterance text DB 13A and the output log DB 13B will be described later together with the scenes in which the respective data are referenced or registered.

[0020] The control unit 15 is a functional unit that performs overall control of the information processing device 10. For example, the control unit 15 can be realized by a hardware processor. As shown in FIG. 1 , the control unit 15 has an acquisition unit 15A, a model execution unit 15B, an output unit 15C, and a DB (Database) update unit 15D. Note that the control unit 15 may also be realized by hardwired logic or the like.

[0021] The acquisition unit 15A is a processing unit that acquires speech text corresponding to speech input by the conference system 3. In one aspect, the acquisition unit 15A can acquire speech text in which speech audio data is converted into text data by a transcription function, or so-called transcription function, provided by the conference system 3. When the transcription function is used in this way, it is possible to acquire, along with the speech text, the speaker, a timestamp for the start or end of the speech section, and the like, for each speech section detected from an audio signal input in real time by the conference system 3. Note that while an example using the transcription function has been given here, it goes without saying that a pair of speaker and speech text can be acquired for each speech section by acquiring speech audio data and performing speech section detection, speaker estimation, and dictation.

[0022] The model execution unit 15B is a processing unit that executes input / output control for large-scale language models, so-called LLMs (Large Language Models). In one aspect, the model execution unit 15B inputs the spoken text acquired by the acquisition unit 15A to the LLM, and causes the LLM to output a script in which a structure corresponding to the minutes of the conference system 3 is written in a markup language.

[0023] Here, in order to ensure consistency in the time series of the LLM output, the model execution unit 15B can change the input to the LLM between the spoken text obtained first from a conference held in the conference system 3 and the spoken text obtained second or later.

[0024] 2 is a schematic diagram (1) showing an example of the input / output form of the LLM 5. Fig. 2 shows the input to and output from the LLM 5 when acquiring the initial utterance text. As shown in Fig. 2, the model execution unit 15B inputs to the LLM 5 a user prompt 21 in which the initial utterance text T1 acquired by the acquisition unit 15A is embedded, and a system prompt 41.

[0025] FIG. 3 is a diagram (1) showing an example of utterance text. FIG. 3 shows an example of initial utterance text T1 embedded in the user prompt 21 shown in FIG. 2. As shown in FIG. 3, the initial utterance text T1 is a text version of what Suzuki says at the beginning of the meeting. In this way, the initial utterance text T1, which is associated with the speaker "Suzuki" and the utterance content "Okay, everyone's here. Today's agenda is (omitted) Harada, please tell me if you have learned anything from your internal investigations so far," is embedded in the user prompt 21.

[0026] FIG. 4 is a diagram (1) showing an example of a system prompt. FIG. 4 also shows an example of the system prompt 41 shown in FIG. 2. As shown in FIG. 4, the system prompt 41 includes an instruction element 41A. This instruction element 41A includes instructions for expressing the structure of a flowchart corresponding to the minutes of a meeting based on inclusion relationships, relevance, and discussion transitions. Furthermore, the instruction element 41A includes instructions for extracting pairs of assertion phrases and the speakers of the assertion phrases for each assertion phrase included in the spoken text. The system prompt 41 also includes an example sentence 41B, embedded as a context element, in Mermaid notation, which has a graph structure with the spoken text as nodes and speakers as edges. Furthermore, the system prompt 41 includes an output format element, as underlined in FIG. 4, that specifies the format of the LLM5 output as Mermaid notation.

[0027] The LLM 5 receives the user prompt 21 and the system prompt 41 and outputs a script SC1 written in Markdown notation, for example, Mermaid notation.

[0028] FIG. 5 is a diagram (1) showing an example of an LLM output. FIG. 5 shows script SC1 output by LLM5 shown in FIG. 2. As shown in FIG. 5, script SC1 describes a flowchart code block that proceeds left or right using Mermaid notation. For example, in the example shown in FIG. 5, five nodes are defined in the flowchart: ID1 "Enhancing security of public web services," ID2 "Instructions from the president," ID3 "High risk," ID4 "Identifying vulnerabilities," and ID5 "Request for report on internal investigation results." Furthermore, four edges are defined in the flowchart: an edge connecting ID1 and ID2, an edge connecting ID1 and ID3, an edge connecting ID1 and ID4, and an edge connecting ID4 and ID5.

[0029] In this way, by implementing prompt engineering that imparts regularity to the chart description in the output of LLM5, it is possible to stabilize the output of LLM5 even when there is variation in the content of the text input to LLM5. Furthermore, by extracting pairs of speech phrases and speakers in units of assertions, even when one speech section contains multiple topics, it is possible to divide it into multiple nodes and describe the flowchart.

[0030] FIG. 6 is a schematic diagram (2) showing an example of the input / output form of the LLM5. FIG. 6 shows the input to and output from the LLM5 when the nth (n is a natural number greater than or equal to 2) utterance text is acquired. As shown in FIG. 6, the model execution unit 15B inputs to the LLM5 a user prompt 22 in which the nth utterance text Tn acquired by the acquisition unit 15A is embedded. Furthermore, the model execution unit 15B inputs to the LLM5 a system prompt 42 in which the first to n-1th utterance texts Tn-1 stored in the utterance text DB 13A and the n-1th script SCn-1 stored in the output log DB 13B, i.e., the output of the LLM5 when the n-1th utterance text is input, are embedded.

[0031] Fig. 7 is a diagram (2) showing an example of utterance text. Fig. 7 shows the second utterance text T2 as an example of the n-th utterance text Tn to be embedded in the user prompt 22 shown in Fig. 6. As shown in Fig. 7, the second utterance text T2 is a text version of what Harada said in response to Suzuki's comment urging him to report on the company's internal investigation. In this way, the second utterance text T2, which is associated with the speaker "Harada" and the utterance "To be honest, I think that the biggest problem is the weak point (omitted)," is embedded in the user prompt 22.

[0032] FIG. 8 is a diagram (2) showing an example of a system prompt. FIG. 8 shows an example of the system prompt 42 shown in FIG. 6. As shown in FIG. 8, the system prompt 42 includes an instruction element 42A. This instruction element 42A includes an instruction to restrict changes to the structure of the flowchart corresponding to the output of the (n-1)th LLM5. Furthermore, the instruction element 42A includes an instruction to add nodes and edges corresponding to the nth spoken text to the description of the graph structure output by the LLM5 when the (n-1)th spoken text is input. The system prompt 42 also includes an output format element, as underlined, that specifies the format of the LLM5 output as Mermaid notation. Furthermore, the system prompt 42 includes the first through (n-1)th spoken texts and the output of the LLM5 when the (n-1)th spoken text is input as a context element 42B.

[0033] The LLM 5 receives the user prompt 22 and the system prompt 42 and outputs a script SCn written in Markdown notation, for example, Mermaid notation.

[0034] FIG. 9 is a diagram (2) showing an example of the output of the LLM5. FIG. 9 shows the second script SC2 as an example of the script SCn output by the LLM5 shown in FIG. 6. Compared to the script SC1 shown in FIG. 5, the script SC2 shown in FIG. 9 adds four nodes to the flowchart: ID6 "Infinite vulnerability search," ID7 "Complexity and cost of security measures," ID8 "Balance between usability and security," and ID9 "The core of the problem." Furthermore, four edges have been added to the flowchart: an edge connecting ID4 and ID6, an edge connecting ID6 and ID7, an edge connecting ID7 and ID8, and an edge connecting ID8 and ID9. The script SC2 also performs a summary by deleting the ID5 node and the node connected to ID5 from the flowchart.

[0035] In this way, by providing an input log to LLM5 and an output log from LLM5 and implementing prompt engineering to restrict changes to the output log of LLM5, it is possible to ensure consistency in the time series of LLM5 output. Furthermore, by implementing prompt engineering to add nodes and edges corresponding to the latest utterance text to an existing flowchart, it is possible to keep the description of the flowchart up to the latest utterance. Prompt engineering like this also makes fine tuning of LLM5 unnecessary.

[0036] The output unit 15C is a processing unit that executes output control for user terminals and the like included in the conference system 3. In one aspect, the output unit 15C outputs visualization data in which the structure of a chart corresponding to the minutes of the conference in the conference system 3 is visualized based on the script output by the LLM 5. Such visualization data may be displayed on a GUI screen, such as a web conference screen, shared by user terminals participating in a conference held in the conference system 3. For example, the output unit 15C inputs the script output by the LLM 5, for example, a code block in which a flowchart is written in Mermaid notation, into a graph creation tool, thereby causing the graph creation tool to draw graph data.

[0037] Fig. 10 is a diagram (1) showing an example of visualization data. Fig. 10 shows an example in which graph data G1 generated from script SC1 shown in Fig. 5 is displayed in a partial display area of ​​a web conference screen provided by the conference system 3. As shown in Fig. 10, nodes corresponding to the gist of Suzuki's opening remarks are plotted from left to right in the graph data G1. With such graph data G1, even if a user misses Suzuki's opening remarks, the key points of the transition of remarks in the conference can be extracted and visualized, thereby providing support for tracking the transition of remarks in the conference.

[0038] Such graph data can be displayed not only when a request is received from the user terminal, but also in real time as the conference progresses. For example, every time new utterance text is acquired by the acquisition unit 15A, the output unit 15C can update the display data displayed on the web conference screen or the like to the latest graph data generated from the script output by the LLM 5 to which the new utterance text has been input.

[0039] Here, the output unit 15C can apply visualization techniques such as infographics to the graph data generated from the script output by the LLM 5. As just one example, the output unit 15C can highlight, in the graph data, a node, an edge, or a combination thereof that corresponds to the most recent spoken text.

[0040] FIG. 11 is a diagram (2) showing an example of visualization data. FIG. 11 shows an example in which graph data G2 generated from script SC2 shown in FIG. 9 is displayed in a partial display area of ​​a web conference screen provided by the conference system 3. As shown in FIG. 11, nodes corresponding to the gist of Suzuki's remarks at the beginning of the conference and Harada's subsequent remarks are plotted from left to right in graph data G2. In such graph data G2, as shown by hatching in FIG. 11, the node "The Core of the Problem," corresponding to Harada's most recent assertion, is highlighted. This allows the viewer to easily grasp the situation in which the discussion in the conference is focused on Harada's most recent assertion, "The Core of the Problem."

[0041] Although Figures 10 and 11 show flowcharts as examples of graph data, this is merely an example, and other charts such as sequence diagrams, flowcharts, Gantt charts, and class diagrams may also be generated.

[0042] The DB update unit 15D is a processing unit that updates various databases. In one aspect, when new utterance text is acquired by the acquisition unit 15A, the DB update unit 15D additionally registers the new utterance text in the utterance text DB 13A. In another aspect, when an LLM5 output is obtained, the DB update unit 15D additionally registers the LLM5 output in the output log DB 13B. This allows the utterance text and LLM5 output from the start of the conference to the present to be accumulated in chronological order.

[0043] <Processing Flow> Next, a processing flow of the information processing device 10 according to this embodiment will be described. Fig. 12 is a flowchart showing the procedure of data output processing. This processing is merely an example and can be started when a conference is started by the conference system 3.

[0044] As shown in FIG. 12, when the acquisition unit 15A acquires the first utterance text (step S101), the model execution unit 15B generates a user prompt 21 in which the first utterance text is embedded (step S102).

[0045] Then, the model execution unit 15B inputs the user prompt 21 generated in step S102 and the system prompt 41 for the initial utterance to the LLM 5 (step S103).

[0046] Next, the output unit 15C outputs graph data in which the graph structure of the minutes of the conference held in the conference system 3 is visualized based on the script output by the LLM 5 as a result of the input in step S103 (step S104).

[0047] Then, the DB update unit 15D stores the initial utterance text acquired in step S101 and the output of the LLM5 obtained as a result of the input in step S103 in each of the utterance text DB 13A and the output log DB 13B (step S105).

[0048] Thereafter, loop processing 1 is executed, repeating the processing from step S106 to step S112 described below, until a specified end condition is met, such as the end of the conference or receipt of an operation to turn off the data output function.

[0049] That is, the acquisition unit 15A acquires the n-th utterance text (step S106), and the model execution unit 15B then generates the user prompt 22 in which the n-th utterance text is embedded (step S107).

[0050] Furthermore, the model execution unit 15B generates a system prompt 42 in which the first n-1 utterance texts Tn-1 stored in the utterance text DB 13A and the n-1 script SCn-1 stored in the output log DB 13B, i.e., the output of LLM5 when the n-1 utterance text is input, are embedded (step S108).

[0051] Then, the model execution unit 15B inputs the user prompt 22 generated in step S107 and the system prompt 42 for additional comment generated in step S108 to the LLM 5 (step S109).

[0052] Next, the output unit 15C generates graph data in which the graph structure of the minutes of the conference held in the conference system 3 is visualized based on the script output by the LLM 5 as a result of the input in step S109 (step S110).

[0053] Thereafter, the output unit 15C updates the display data displayed on the Web conference screen or the like to the latest graph data generated in step S110 (step S111).

[0054] Then, the DB update unit 15D adds the n-th utterance text acquired in step S106 and the output of LLM5 obtained as a result of the input in step S109 to each of the utterance text DB 13A and the output log DB 13B (step S112).

[0055] By such loop processing 1, graph data can be provided that visualizes the graph structure of the conference minutes in real time, following the progress of the conference held in the conference system 3.

[0056] <Summary> As described above, the information processing device 10 according to this embodiment outputs visualization data based on a script that describes the structure of a chart corresponding to the minutes of a meeting by inputting speech text corresponding to speech speech in a meeting into a large-scale language model. By having the LLM 5 describe the structure of a chart corresponding to the minutes of a meeting in this way, the stability of the image generated from the script can be improved. Therefore, the information processing device 10 according to this embodiment can realize real-time visualization of the minutes of a meeting.

[0057] <Exercising Creative Ability> The matters described in this embodiment, such as specific examples of speech text and prompts, and the types of LLMs, are merely examples and can be changed. Furthermore, the order of processes in the flowchart described in this embodiment can be changed or some processes can be skipped within a consistent range.

[0058] <System> The information including the processing procedures, control procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, any one or more of the functional units of the acquisition unit 15A, model execution unit 15B, output unit 15C, and DB update unit 15D of the information processing device 10 may be configured as separate devices.

[0059] Furthermore, the components of each device shown in the figure are functional concepts and do not necessarily have to be physically configured as shown. In other words, the specific form of distribution and integration of each device is not limited to that shown. In other words, all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Note that each configuration may also be a physical configuration.

[0060] Furthermore, all or any part of the processing performed by the illustrated device may be realized by a program executed by a hardware processor such as an MPU (Micro-Processing Unit) or a CPU (Central Processing Unit), or may be realized as hardware using wired logic.

[0061] <Hardware> Next, an example of the hardware configuration of the information processing device 10 described in this embodiment will be described. For example, the information processing device 10 can be implemented by installing a program that realizes the functions of the information processing device 10 on a computer. For example, the information processing device 10 can be implemented by having the computer execute the program, which is provided as package software or online software. The computer referred to here includes desktop or notebook personal computers, rack-mounted server computers, and the like. Furthermore, the computer also includes smartphones, mobile phones, PHS (Personal Handyphone System) and other mobile communication terminals, as well as PDAs (Personal Digital Assistants). The functions of the information processing device 10 may also be implemented on a cloud server.

[0062] An example of a computer that executes the above program (data output program) will be described using Fig. 13. As shown in Fig. 13, the computer 1000 includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0063] The memory 1010 includes a read-only memory (ROM) 1011 and a random access memory (RAM) 1012. The ROM 1011 stores a boot program such as a basic input / output system (BIOS). The hard disk drive interface 1030 is connected to a hard disk drive 1090. The disk drive interface 1040 is connected to a disk drive 1100. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to a mouse 1110 and a keyboard 1120, for example. The video adapter 1060 is connected to a display 1130, for example.

[0064] 13, the hard disk drive 1090 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. The storage unit 13 described in the above embodiment is provided in, for example, the hard disk drive 1090 or the memory 1010.

[0065] Then, the CPU 1020 reads out the program module 1093 and the program data 1094 stored in the hard disk drive 1090 into the RAM 1012 as necessary, and executes the above-mentioned procedures.

[0066] The program module 1093 and program data 1094 relating to the data output program are not limited to being stored in the hard disk drive 1090, but may be stored in a removable storage medium, for example, and read by the CPU 1020 via the disk drive 1100. Alternatively, the program module 1093 and program data 1094 relating to the program may be stored in another computer connected via a network such as a LAN or a WAN (Wide Area Network), and read by the CPU 1020 via the network interface 1070.

[0067] 3 Conference system 5 LLM 10 Information processing device 11 Communication control unit 13 Storage unit 13A Speech text DB 13B Output log DB 15 Control unit 15A Acquisition unit 15B Model execution unit 15C Output unit 15D DB update unit

Claims

1. An information processing device comprising: an acquisition unit that acquires spoken text corresponding to spoken voice input in a conference system; a model execution unit that inputs the spoken text into a large-scale language model, and causes the large-scale language model to output a script in which a chart structure corresponding to minutes of a meeting in the conference system is written in a markup language; and an output unit that outputs visualization data in which the chart structure corresponding to minutes of a meeting in the conference system is visualized based on the script output by the large-scale language model.

2. The information processing device described in claim 1, characterized in that the model execution unit inputs into the large-scale language model a prompt that uses the spoken text as input data, has a graph structure in which the spoken text is a node, and the speaker of the spoken text is an edge, and includes example sentences written in the markup language as context elements.

3. The information processing device described in claim 2, characterized in that the model execution unit uses a second spoken text acquired by the acquisition unit subsequent to the first spoken text as the input data and inputs to the large-scale language model a prompt including an instruction element that instructs the addition of nodes and edges corresponding to the second spoken text to the description of the graph structure output by the large-scale language model when the first spoken text is input.

4. The information processing device described in claim 3, characterized in that the model execution unit inputs a prompt to the large-scale language model whose context elements include the first spoken text that was acquired by the acquisition unit prior to the acquisition of the second spoken text and the output of the large-scale language model at the time the first spoken text was input.

5. The information processing device described in claim 2, characterized in that the model execution unit inputs a prompt to the large-scale language model that includes an instruction element to extract pairs of assertion phrases and speakers of the assertion phrases for each assertion phrase contained in the spoken text.

6. The information processing device described in claim 2, characterized in that the output unit highlights a node, edge, or a combination thereof corresponding to the first spoken text in graph data that visualizes a graph structure corresponding to minutes in the conference system.

7. A data output method executed by an information processing device, comprising: an acquisition step of acquiring spoken text corresponding to spoken voice input in a conference system; a model execution step of inputting the spoken text into a large-scale language model, thereby causing the large-scale language model to output a script in which a chart structure corresponding to minutes of a meeting in the conference system is written in a markup language; and an output step of outputting visualization data in which the chart structure corresponding to minutes of a meeting in the conference system is visualized based on the script output by the large-scale language model.

8. A data output program for causing a computer to execute the following steps: an acquisition step of acquiring spoken text corresponding to spoken voice input in a conference system; a model execution step of inputting the spoken text into a large-scale language model, thereby causing the large-scale language model to output a script in which a chart structure corresponding to minutes of a meeting in the conference system is written in a markup language; and an output step of outputting visualization data in which the chart structure corresponding to minutes of a meeting in the conference system is visualized based on the script output by the large-scale language model.

Images