Server equipment, system, support method, and support program

Abstract

This invention provides a server device, system, support method, and support program to assist the participation of generative AI in communication settings involving multiple users. [Solution] In a system in which multiple server devices and multiple terminals are connected to communicate via a network, the server device 120, which functions as a lecture tuning unit 121, has a prompt generation unit 312 that acquires the input information when information is input by a user in a communication space provided by a communication service and in which multiple users can participate, and instructs a generation AI that operates with an assistant that performs a specified task under a predetermined role to generate response information based on the information, and an input control unit 314 that acquires the response information based on the information generated by the generation AI and controls the input of the response information into the communication space.

Description

Technical Field

[0001] The present disclosure relates to a server device, a system, a support method, and a support program.

Background Art

[0002] In a chat service using generative AI such as ChatGPT, for example, when a user inputs a question, an appropriate answer can be output to the user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] On the other hand, generative AI such as ChatGPT does not have a group discussion function. Therefore, in the case of the above chat service, in a communication place where a plurality of users participate, for example, it is impossible to provide a service in which generative AI participates with a predetermined role and outputs an appropriate answer to a question from a user.

[0005] An object of the present disclosure is to support the participation of generative AI in a communication place where a plurality of users participate.

Means for Solving the Problems

[0006] According to one aspect of the present disclosure, a server device In a communication space provided by a communication service, where multiple users can participate, when information is input by a user, a prompt generation unit instructs a generation AI, which operates using an assistant that acquires the information and performs a specified task under a predetermined role, to generate response information based on the information. The system includes an input control unit that acquires response information based on the information generated by the generation AI and controls the input of the response information into the communication environment. [Effects of the Invention]

[0007] According to this disclosure, it is possible to support the participation of generative AI in communication settings involving multiple users. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example of the system configuration during the preparation phase of the system according to the first embodiment. [Figure 2] This figure shows an example of the hardware configuration of a server device. [Figure 3] This diagram shows an overview of the construction process performed by the lecture tuning unit. [Figure 4] This diagram shows specific examples of assistants. [Figure 5] This figure shows a concrete example of registration data for a lecture. [Figure 6] This diagram shows an overview of the tuning process performed by the lecture tuning unit. [Figure 7A] This is the first diagram showing the flow of the preparation process in the system according to the first embodiment. [Figure 7B] This is a second diagram showing the flow of the preparation process in the system according to the first embodiment. [Figure 7C] This is a third diagram showing the flow of the preparation process in the system according to the first embodiment. [Figure 8]It is a diagram showing an example of the system configuration in the execution phase of the system according to the first embodiment. [Figure 9] It is a diagram showing an overview of the lecture support process by the lecture support department. [Figure 10A] It is the first diagram showing the flow of lecture processing in the system according to the first embodiment. [Figure 10B] It is the second diagram showing the flow of lecture processing in the system according to the first embodiment. [Figure 10C] It is the third diagram showing the flow of lecture processing in the system according to the first embodiment. [Figure 11] It is a diagram showing a specific example of the display screen. [Figure 12] It is a diagram showing an example of the system configuration in the preparation phase of the system according to the second embodiment. [Figure 13] It is a diagram showing an overview of the construction process by the translation tuning department. [Figure 14] It is a diagram showing an overview of the tuning process by the translation tuning department. [Figure 15A] It is the first diagram showing the flow of preparation processing in the system according to the second embodiment. [Figure 15B] It is the second diagram showing the flow of preparation processing in the system according to the second embodiment. [Figure 15C] It is the third diagram showing the flow of preparation processing in the system according to the second embodiment. [Figure 15D] It is the fourth diagram showing the flow of preparation processing in the system according to the second embodiment. [Figure 15E] It is the fifth diagram showing the flow of preparation processing in the system according to the second embodiment. [Figure 16] It is a diagram showing an example of the system configuration in the execution phase of the system according to the second embodiment. [Figure 17] It is a diagram showing an overview of the translation support process by the translation support department. [Figure 18A] It is the first diagram showing the flow of conference processing in the system according to the second embodiment. [Figure 18B] It is a second diagram showing the flow of conference processing in the system according to the second embodiment. [Figure 18C] It is a third diagram showing the flow of conference processing in the system according to the second embodiment. [Figure 19] It is a diagram showing an example of the system configuration in the execution phase of the system according to the third embodiment. [Figure 20] It is a diagram showing a specific example of the output information of the terminal.

Embodiments for Carrying Out the Invention

[0009] Hereinafter, each embodiment will be described with reference to the accompanying drawings. In this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant explanations are omitted.

[0010] [First Embodiment] <System Configuration in Preparation Phase> First, the system configuration of the system according to the first embodiment will be described. The system according to the first embodiment is a system for conducting lectures for students, · In the "Preparation Phase", "Preparation Processes" such as construction of assistants, registration and modification of registration data are performed. · In the "Execution Phase", using the constructed assistants and the registered and modified registration data, "Lecture Processes" are performed in which the generative AI gives lectures to students in the role of a tutor.

[0011] Here, first, the system configuration of the system that performs preparation processes in the preparation phase will be described. FIG. 1 is a diagram showing an example of the system configuration in the preparation phase of the system according to the first embodiment.

[0012] As shown in Figure 1, the system 100 in the preparation phase includes server devices 110, 120, 130, and terminals 140_1 to 140_3. In the system 100 in the preparation phase, server device 120 is connected to server devices 110, 130, and terminals 140_1 to 140_3 via network 160 for communication. Also in the system 100 in the preparation phase, server device 110 is connected to terminals 140_1 to 140_3 via network 160 for communication.

[0013] The server device 110 functions as a communication service provider unit 111 by executing a communication service provision program. The communication service provider unit 111 provides a communication service that allows multiple users to participate and communicate with each other. The communication service provider unit 111 is implemented by software such as Discord.

[0014] The server device 120 functions as a lecture tuning unit 121 by executing a lecture tuning program during the preparation phase. The lecture tuning unit 121 executes the "construction process" and the "tuning process" among the various processes included in the preparation process. In the "construction process", the lecture tuning unit performs the following: • Using the assistant function of the AI ​​chat service provider unit 131 described later, an "assistant" is generated to act as a tutor in a lecture in a specific field for students, in order to operate the AI ​​chat service provider unit 131. • Register the necessary data for providing lectures in specific fields for students in the AI ​​chat service provider unit 131.

[0015] Furthermore, in the "tuning process," the lecture tuning unit 121 is: The AI ​​chat service provider unit 131 is operated in the role of a tutor, and the registration data is corrected through lectures given to simulated students.

[0016] The server device 130 functions as an AI chat service provider unit 131 by executing an AI chat service provider program. The AI ​​chat service provider unit 131 provides a chat service by performing language processing using a large-scale language model. The AI ​​chat service provider unit 131 is implemented by a generative AI such as ChatGPT.

[0017] As described above, the AI ​​chat service provider unit 131 has an assistant function. The assistant function is a function that causes the AI ​​chat service provider unit 131 to operate in order to perform specified tasks under a predetermined role, and the assistant has a defined role, tasks, and methods for performing those tasks.

[0018] For lectures in specific fields aimed at students, the assistant should be: • The role of a tutor, responsible for supervising multiple students, is defined. • Tasks are defined to conduct lectures in a specific field using registered data. • The specific method for performing the task is defined.

[0019] Terminals 140_1 to 140_3 are terminals operated by generators 150_1 to 150_3. During the preparation phase, generators 150_1 to 150_3 input various instructions to the server device 120 via terminals 140_1 to 140_3 to perform tasks such as generating assistants, registering and modifying registration data. Generators 150_1 to 150_3 consist of, for example, instructors (lecturers, associate professors, professors, etc.) who give lectures in a specific field, and assistants (engineers, salespeople) who support the input of various instructions by the instructors.

[0020] <Server hardware configuration> Next, the hardware configurations of server devices 110, 120, and 130 will be described. In this embodiment, server devices 110, 120, and 130 have similar hardware configurations, so here we will describe the hardware configuration of server device 120. Figure 2 is a diagram showing an example of the hardware configuration of a server device. As shown in Figure 2, server device 120 has a processor 201, memory 202, auxiliary storage device 203, interface device 204, communication device 205, and drive device 206. The hardware components of server device 120 are interconnected via a bus 207.

[0021] The processor 201 has various computing devices such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 201 executes various programs (for example, a lecture tuning program) by reading them into memory 202.

[0022] Memory 202 has main memory devices such as ROM (Read Only Memory) and RAM (Random Access Memory). The processor 201 and memory 202 form a so-called computer, and the computer realizes various functions by the processor 201 executing various programs read from memory 202.

[0023] The auxiliary storage device 203 stores various programs and various data used when those programs are executed by the processor 201.

[0024] Interface device 204 is a connection device for connecting an operating device 211 and a display device 212, which are examples of user interface devices. Communication device 205 is a communication device for communicating with server devices 110, 130, terminals 140_1 to 140_3, etc., via network 160.

[0025] The drive device 206 is a device for setting the recording medium 213. The recording medium 213 here includes media that record information optically, electrically, or magnetically, such as CD-ROMs, flexible disks, and magneto-optical disks. The recording medium 213 may also include semiconductor memory such as ROM and flash memory that records information electrically.

[0026] The various programs to be installed on the auxiliary storage device 203 are installed, for example, when the distributed recording medium 213 is set in the drive device 206 and the various programs recorded on the recording medium 213 are read by the drive device 206. Alternatively, the various programs to be installed on the auxiliary storage device 203 may be installed when they are downloaded from the network via the communication device 205.

[0027] <Overview of the construction process by the lecture tuning unit> Next, an overview of the "construction process" performed by the lecture tuning unit 121 of the server device 120 will be described. As mentioned above, in the "construction process", the lecture tuning unit 121 performs the following: • In order to operate the AI ​​chat service provider unit 131 using its assistant function, an "assistant" is generated to act as a tutor in a specific subject lecture for students. • Register the necessary data for providing lectures in specific fields for students in the AI ​​chat service provider unit 131.

[0028] Figure 3 shows an overview of the construction process performed by the lecture tuning unit. As shown in Figure 3, the lecture tuning unit 121 further includes an assistant generation unit 311, a prompt generation unit 312, a prompt correction unit 313, and an input control unit 314. During the construction process, the assistant generation unit 311 and the prompt generation unit 312 are in operation.

[0029] The assistant generation unit 311 generates an assistant using the API 322 provided by the AI ​​chat service provision unit 131. An example in Figure 3 shows that... • Name: Lecture α Tutor, Instructions: Instructions for Lecture α, This shows how the assistants are generated. The assistant generation unit 311 generates, for example, a number of assistants corresponding to the number of lectures.

[0030] In the construction process, the prompt generation unit 312 registers the registration data 323 with the AI ​​chat service provider unit 131 using prompts. The example in Figure 3 shows how registration data for lecture α has been registered as registration data necessary to realize lecture α. The prompt generation unit 312 generates, for example, a number of lecture registration data corresponding to the number of lectures.

[0031] <Specific examples of assistants> Next, we will describe a specific example of an assistant generated by the assistant generation unit 311. Figure 4 shows a specific example of an assistant.

[0032] As shown by the symbol 410 in Figure 4, the assistant's Name is defined as the tutor's name. The example in Figure 4 shows how "Linear Algebra Tutor" is defined as the tutor's name.

[0033] As shown by the symbol 410 in Figure 4, the Assistant's Instructions define the role of the Generative AI. The example in Figure 4 shows how the role of a "linear algebra tutor who teaches linear algebra to freshmen in the Faculty of Science and Engineering" is defined.

[0034] As shown by reference numeral 410 in Figure 4, the Assistant's Instructions define the tasks that the Generating AI should perform. An example of reference numeral 410 is a task, • What kind of registration data will be used for the lecture? How should one approach a student when providing individualized instruction? • What teaching methods will be used for the lectures? • Matters to be covered in the lecture, • Things to do in preparation for the exam, How to give feedback to students, • Display format when displayed in a thread (an example of a communication space provided by the Communication Service Provider Unit 111), This shows how these are defined in multiple sections.

[0035] <Specific examples of registration data for lectures> Next, we will explain a specific example of the registration data for lecture α, which is necessary to realize lecture α, as shown in Figure 3, from the registration data 323 registered in the AI ​​chat service provision unit 131 by the prompt generation unit 312. Figure 5 is a diagram showing a specific example of lecture registration data.

[0036] As shown in the registration data 323 in Figure 5, the registration data for Lecture α includes textbook data, lecture material data, homework assignment data, exam data, etc. When the AI ​​chat service provider 131 operates, the assistant indicated by reference numeral 410 in Figure 4, acting as a Lecture α tutor, performs tasks specified by the Lecture α instructions while referring to the Lecture α registration data. This enables the AI ​​chat service provider 131 to generate appropriate response information for Lecture α.

[0037] <Overview of the tuning process by the lecture tuning department> Next, we will explain the outline of the "tuning process" performed by the lecture tuning unit 121 of the server device 120 during the preparation phase. As described above, the lecture tuning unit 121 operates the AI ​​chat service provider unit 131 in the role of a tutor and performs a "tuning process" to modify the registered data 323 through lectures to simulated students.

[0038] Figure 6 shows an overview of the tuning process performed by the lecture tuning unit. As shown in Figure 6, the lecture tuning unit 121 further includes an assistant generation unit 311, a prompt generation unit 312, a prompt correction unit 313, and an input control unit 314. During the tuning process, the prompt generation unit 312, the prompt correction unit 313, and the input control unit 314 are in operation.

[0039] Furthermore, when the lecture tuning unit 121 starts the tuning process, the communication service provision unit 111 of the server device 110 is assumed to provide the "Lecture α" thread 610. In addition, in the "Lecture α" thread 610, the assistant with the tutor name "Lecture α Tutor" is assumed to be selected. Moreover, in the "Lecture α" thread 610, creators A to C are assumed to be participating as pseudo-students A to pseudo-students C, respectively.

[0040] During the tuning process, the prompt generation unit 312 acquires input information when any simulated student inputs information to the "Lecture α" thread 610. The prompt generation unit 312 sends the acquired input information to the AI ​​chat service provider unit 131 of the server device 130, instructing it to generate response information based on the input information.

[0041] The example in Figure 6 shows how the prompt generation unit 312 acquires simulated questions (1) and (2) entered into the "Lecture α" thread 610 by simulated students A and B. The example in Figure 6 also shows how the acquired simulated questions (1) and (2) are sent to the AI ​​chat service provider unit 131, instructing it to create answers to the simulated questions (1) and (2).

[0042] During the tuning process, when the prompt correction unit 313 receives request information requesting prompt correction in the "Lecture α" thread 610, it sends the request information to the AI ​​chat service provider unit 131 of the server device 130.

[0043] The example in Figure 6 shows how the prompt correction unit 313 receives a request information from simulated student C to correct the prompt in the "Lecture α" thread 610, and sends it to the AI ​​chat service provider unit 131.

[0044] During the tuning process, the input control unit 314 receives a response (an example of response information) sent from the AI ​​chat service provider unit 131 of the server device 130 in response to input information being sent by the prompt generation unit 312. Alternatively, the input control unit 314 receives a response result (an example of response information) sent from the AI ​​chat service provider unit 131 of the server device 130 in response to request information being sent by the prompt correction unit 313.

[0045] The input control unit 314 converts the received response or response result into a display format corresponding to the "Lecture α" thread 610, and inputs the converted response or response result into the "Lecture α" thread 610.

[0046] The example in Figure 6 shows how the input control unit 314 receives the answers (1) and (2) to the simulated questions (1) and (2) entered into the "Lecture α" thread 610 by simulated students A and B, and inputs them into the "Lecture α" thread 610. The example in Figure 6 also shows how the input control unit 314 receives the response result (correction complete) to the request information for prompt correction entered into the "Lecture α" thread 610 by simulated student C, and inputs it into the "Lecture α" thread 610.

[0047] <Preparation Process Flow> Next, we will explain the flow of the "preparation process" in the preparation phase by the entire system 100. Figures 7A to 7C are the first to third diagrams showing the flow of the preparation process by the system according to the first embodiment. In Figures 7A to 7C, the server device 110 is omitted due to space limitations.

[0048] During the preparation process, the input information that generators 150_1 to 150_3 input to the thread via terminals 140_1 to 140_3 will be notified to server device 120 via server device 110. Also, during the preparation process, terminals 140_1 to 140_3 will display the input information that generators 150_1 to 150_3 input to the thread, as well as the response information (answer, response result) sent from server device 120 to server device 110.

[0049] Furthermore, when performing the preparation processes shown in Figures 7A to 7C, it is assumed that the "construction process" by the lecture tuning unit 121 has already been completed. In other words, the preparation processes shown in Figures 7A to 7C will be explained primarily through the "tuning process" by the lecture tuning unit 121.

[0050] As shown in Figure 7A, in step S701, terminal 140_1 issues a Chat command in response to input from generator 150_1 (generator A) and sends it to server device 110. As a result, the communication service provider unit 111 of server device 110 generates a thread (in this case, the "Lecture α" thread 610).

[0051] In step S702, the lecture tuning unit 121 of the server device 120 receives the Chat command sent to the server device 110 by terminal 140_1.

[0052] In step S703, the lecture tuning unit 121 of the server device 120 queries the server device 130 for a tutor name list, which is a list of assistants who can participate as tutors in the generated thread.

[0053] In step S704, the AI ​​chat service provider unit 131 of the server device 130 retrieves the tutor name of the assistant whose role as a tutor is defined among the assistants registered in API 322.

[0054] In step S705, the AI ​​chat service provision unit 131 of the server device 130 transmits the acquired tutor name to the server device 120.

[0055] In step S706, the lecture tuning unit 121 of the server device 120 sends the received list of tutor names to the server device 110 as a tutor name list. As a result, the tutor name list is posted to the thread provided by the communication service provision unit 111 of the server device 110.

[0056] In step S707, terminal 140_1 joins the thread as pseudo-student A, in accordance with instructions from generator 150_1 (generator A). Although not shown in Figure 7A, terminal 140_2 also joins the thread as pseudo-student B, in accordance with instructions from generator 150_2 (generator B). Terminal 140_3 joins the thread as pseudo-student C, in accordance with instructions from generator 150_3 (generator C).

[0057] In step S708, terminal 140_1, in response to instructions from generator 150_1 (generator A), selects one tutor name (in this case, "Lecture α Tutor") from the tutor name list and sends it to server device 110.

[0058] In step S709, the lecture tuning unit 121 of the server device 120 accepts the selection of a tutor name transmitted to the server device 110 by terminal 140_1.

[0059] In step S710, the lecture tuning unit 121 of the server device 120 associates the assistant with the selected tutor name with a thread. Thereafter, the assistant associated with the thread will respond to the input information entered into that thread.

[0060] In step S711, terminal 140_1 creates a pseudo-question (1) in accordance with the instructions of generator 150_1 (generator A).

[0061] In step S712, terminal 140_1 inputs the created pseudo-question (1) into the thread.

[0062] In step S713, the lecture tuning unit 121 of the server device 120 obtains the simulated question (1) entered into the thread from the server device 110.

[0063] In step S714, the lecture tuning unit 121 of the server device 120 identifies the simulated student A who entered the simulated question (1). The lecture tuning unit 121 of the server device 120 also assigns identification information (tag) to the simulated question (1) and sends it to the assistant associated with the thread, thereby instructing the creation of an answer to the simulated question (1).

[0064] In step S715, the AI ​​chat service provider unit 131 of the server device 130 receives a simulated question (1) along with a prompt associated with the assistant.

[0065] In step S716, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for lecture α) specified in the prompt associated with the assistant.

[0066] In step S717, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, creates an answer to the simulated question (1) and issues the answer (1) while referring to the registered data 323.

[0067] In step S718, the lecture tuning unit 121 of the server device 120 receives the answer (1) to the simulated question (1) issued by the AI ​​chat service provision unit 131 of the server device 130.

[0068] In step S721 of Figure 7B, the lecture tuning unit 121 of the server device 120 converts the received response (1) into a display format corresponding to the thread (in this case, the "Lecture α" thread 610).

[0069] In step S722, the lecture tuning unit 121 of the server device 120 inputs the converted answer (1) into a thread provided by the communication service provision unit 111 of the server device 110 (in this case, the "Lecture α" thread 610).

[0070] In step S723_1, terminal 140_1 displays the thread in which answer (1) was entered (in this case, thread 610 of "Lecture α").

[0071] In step S723_2, terminal 140_2 displays the thread in which answer (1) was entered (in this case, thread 610 of "Lecture α").

[0072] In step S723_3, terminal 140_3 displays the thread in which answer (1) was entered (in this case, thread 610 of "Lecture α").

[0073] In step S724, terminal 140_2 creates a pseudo-question (2) in accordance with the instructions of generator 150_2 (generator B).

[0074] In step S725, terminal 140_2 inputs the created pseudo-question (2) into the thread.

[0075] In step S726, the lecture tuning unit 121 of the server device 120 obtains the simulated question (2) entered into the thread from the server device 110.

[0076] In step S727, the lecture tuning unit 121 of the server device 120 identifies the simulated student B who entered the simulated question (2). The lecture tuning unit 121 of the server device 120 also assigns identification information (tag) to the simulated question (2) and sends it to the assistant associated with the thread, thereby instructing the creation of an answer corresponding to the simulated question (2).

[0077] In step S728, the AI ​​chat service provider unit 131 of the server device 130 receives a simulated question (2) along with a prompt associated with the assistant.

[0078] In step S729, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for lecture α) specified in the prompt associated with the assistant.

[0079] In step S730, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, creates an answer to the simulated question (2) and issues the answer (2) while referring to the registered data 323.

[0080] In step S731, the lecture tuning unit 121 of the server device 120 receives the answer (2) to the simulated question (2) issued by the AI ​​chat service provision unit 131 of the server device 130.

[0081] In step S732, the lecture tuning unit 121 of the server device 120 converts the received response (2) into a display format corresponding to the thread (in this case, the "Lecture α" thread 610).

[0082] In step S733, the lecture tuning unit 121 of the server device 120 inputs the converted answer (2) into a thread provided by the communication service provision unit 111 of the server device 110 (in this case, the "Lecture α" thread 610).

[0083] In step S734_1, terminal 140_1 displays the thread in which answer (2) was entered (in this case, thread 610 of "Lecture α").

[0084] In step S734_2, terminal 140_2 displays the thread in which answer (2) was entered (in this case, thread 610 of "Lecture α").

[0085] In step S734_3, terminal 140_3 displays the thread in which answer (2) was entered (in this case, thread 610 of "Lecture α").

[0086] In step S741 of Figure 7C, terminal 140_3 generates build and update commands in response to instructions from generator 150_3 (generator C), and inputs request information to the thread requesting correction of the prompt associated with the assistant.

[0087] In step S742, the lecture tuning unit 121 of the server device 120 obtains the request information entered into the thread from the server device 110 and transmits it to the server device 130.

[0088] In step S743, the AI ​​chat service provider unit 131 of the server device 130 modifies the prompt associated with the assistant.

[0089] In step S744, the AI ​​chat service provision unit 131 of the server device 130 sends a message to the server device 120 indicating that the prompt correction has been completed.

[0090] In step S745, the lecture tuning unit 121 of the server device 120 inputs the result of its response to the request information (correction complete) into a thread provided by the communication service provision unit 111 of the server device 110 (in this case, the "Lecture α" thread 610).

[0091] In step S746_1, terminal 140_1 displays the thread in which the corrective action result (correction complete) was entered (in this case, thread 610 of "Lecture α").

[0092] In step S746_2, terminal 140_2 displays the thread in which the corrective action result (correction complete) was entered (in this case, thread 610 of "Lecture α").

[0093] In step S746_3, terminal 140_3 displays the thread in which the corrective action result (correction complete) was entered (in this case, thread 610 of "Lecture α").

[0094] In step S747, terminal 140_3 creates a pseudo-question (2) in accordance with the instructions of generator 150_3 (generator C).

[0095] In step S748, terminal 140_3 inputs the created pseudo-question (2) into the thread.

[0096] In step S749, the lecture tuning unit 121 of the server device 120 obtains the simulated question (2) entered into the thread from the server device 110.

[0097] In step S750, the lecture tuning unit 121 of the server device 120 identifies the simulated student C who entered the simulated question (2). The lecture tuning unit 121 of the server device 120 also assigns identification information (tag) to the simulated question (2) and sends it to the assistant associated with the thread, thereby instructing the creation of an answer to the simulated question (2).

[0098] In step S751, the AI ​​chat service provider unit 131 of the server device 130 receives the simulated question (2) along with the revised prompt associated with the assistant.

[0099] In step S752, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, the registration data for the revised lecture α) specified in the revised prompt that is associated with the assistant.

[0100] In step S753, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, creates an answer to the simulated question (2) while referring to the registered data 323 and issues a revised answer (2)'.

[0101] In step S754, the lecture tuning unit 121 of the server device 120 receives the revised answer (2)' to the simulated question (2) issued by the AI ​​chat service provision unit 131 of the server device 130.

[0102] In step S755, the lecture tuning unit 121 of the server device 120 converts the received corrected answer (2)' into a display format corresponding to the thread (in this case, the "Lecture α" thread 610).

[0103] In step S756, the lecture tuning unit 121 of the server device 120 inputs the converted corrected answer (2)' into the thread provided by the communication service provision unit 111 of the server device 110 (in this case, the "Lecture α" thread 610).

[0104] In step S757_1, terminal 140_1 displays the thread in which the corrected answer (2)' was entered (in this case, thread 610 of "Lecture α").

[0105] In step S757_2, terminal 140_2 displays the thread in which the corrected answer (2)' was entered (in this case, thread 610 of "Lecture α").

[0106] In step S757_3, terminal 140_3 displays the thread in which the corrected answer (2)' was entered (in this case, thread 610 of "Lecture α").

[0107] <System configuration during the execution phase> Next, the system configuration in the execution phase of the system according to the first embodiment will be described. Figure 8 is a diagram showing an example of the system configuration in the execution phase of the system according to the first embodiment.

[0108] As shown in Figure 8, the system 800 in the execution phase includes server devices 110, 120, 130, and terminals 820_1 to 820_n and 820_n+1 to 820_n+m. In the system 800 in the execution phase, server device 120 is connected to server devices 110 and 130 via network 160 so as to be able to communicate. Also in the system 800 in the execution phase, server device 110 is connected to terminals 820_1 to 820_n and 820_n+1 to 820_n+m via network 160 so as to be able to communicate.

[0109] The server device 110 functions as a communication service provider 111. Details of the communication service provider 111 have already been explained using Figure 1, so they will not be explained here.

[0110] The server device 120 functions as a lecture support unit 810 by executing a lecture support program during the execution phase. The lecture support unit 810 is • In response to threads provided by the Communication Service Provider 111, the AI ​​Chat Service Provider 131 participates in the threads in the role of a tutor and assists in providing lectures to students. Perform "Lecture Support Processing".

[0111] The server device 130 functions as the AI ​​chat service provider unit 131. In the execution phase, the AI ​​chat service provider unit 131 is the same unit that was constructed and tuned by the server device 120 during the preparation phase. Details of the AI ​​chat service provider unit 131 have already been explained using Figure 1, so they will not be explained here.

[0112] <Overview of lecture support processing by the Lecture Support Department> Next, we will explain the overview of the "lecture support processing" performed by the lecture support unit 810 of the server device 120 during the execution phase. As mentioned above, the lecture support unit 810 performs the following: • In response to threads provided by the Communication Service Provider 111, the AI ​​Chat Service Provider 131 participates in the threads in the role of a tutor and assists in providing lectures to students. Execute "Lecture Support Processing".

[0113] Figure 9 shows an overview of the lecture support processing performed by the lecture support unit. As shown in Figure 9, the lecture support unit 810 further includes a prompt generation unit 811 and an input control unit 812.

[0114] Furthermore, when the lecture support unit 810 starts lecture support processing, the communication service provision unit 111 of the server device 110 is assumed to provide the "Lecture α" thread 910. Also, the "Lecture α" thread 910 is assumed to have an assistant named "Lecture α Tutor" selected, and for example, three students (student A to student C) are participating in the "Lecture α" thread 910.

[0115] In the lecture support process, the prompt generation unit 811 acquires input information when any student inputs information to the "Lecture α" thread 910. The prompt generation unit 811 sends the acquired input information to the AI ​​chat service provider unit 131 of the server device 130, instructing it to generate response information based on the input information.

[0116] The example in Figure 9 shows how the prompt generation unit 811 retrieves questions (1), (2), and (3) that were entered into the "Lecture α" thread 910 by students A, B, and C. The example in Figure 9 also shows how the retrieved questions (1), (2), and (3) are sent to the AI ​​chat service provider unit 131, instructing it to create answers to questions (1), (2), and (3).

[0117] In the lecture support processing, the input control unit 812 receives a response (an example of response information) sent from the AI ​​chat service provision unit 131 of the server device 130 in response to input information being sent by the prompt generation unit 811.

[0118] The input control unit 812 converts the received response into a display format corresponding to the "Lecture α" thread 910, and inputs the converted response into the "Lecture α" thread 910.

[0119] The example in Figure 9 shows how the input control unit 812 receives the answers (1), (2), and (3) to questions (1), (2), and (3) entered by students A, B, and C into the "Lecture α" thread 910, and inputs them back into the "Lecture α" thread 910.

[0120] <Lecture Processing Flow> Next, we will explain the flow of "lecture processing" in the execution phase by the entire system 800. Figures 10A to 10C are the first to third diagrams showing the flow of lecture processing in the system according to the first embodiment. In Figures 10A to 10C, the server device 110 is omitted due to space limitations.

[0121] During the lecture process, the input information that each student (student A to student C) enters into the thread via terminals 820_1 to 820_3 will be notified to server device 120 via server device 110. In addition, during the lecture process, terminals 820_1 to 820_3 will display the input information that each student (student A to student C) has entered into the thread, as well as the response information (answers) sent from server device 120 to server device 110.

[0122] As shown in Figure 10A, in step S1001, terminal 820_1 issues a Chat command in response to input from student 830_1 (student A) and sends it to server device 110. As a result, the communication service provider unit 111 of server device 110 generates a thread (in this case, the "Lecture α" thread 910).

[0123] In step S1002, the lecture support unit 810 of the server device 120 receives the Chat command sent to the server device 110 by terminal 820_1.

[0124] In step S1003, the lecture support unit 810 of the server device 120 queries the server device 130 for a list of tutor names, which is a list of assistants who can participate as tutors in the generated thread.

[0125] In step S1004, the AI ​​chat service provider unit 131 of the server device 130 retrieves the tutor name of the assistant whose role as a tutor is defined among the assistants registered in API 322.

[0126] In step S1005, the AI ​​chat service provision unit 131 of the server device 130 transmits the acquired tutor name to the server device 120.

[0127] In step S1006, the lecture support unit 810 of the server device 120 sends the received list of tutor names to the server device 110 as a tutor name list. As a result, the tutor name list is posted to the thread provided by the communication service provision unit 111 of the server device 110.

[0128] In step S1007, terminal 820_1 joins the thread as student A, in accordance with instructions from student 830_1 (student A). Although not shown in Figure 10A, terminal 820_2 also joins the thread as student B, in accordance with instructions from student 830_2 (student B). Terminal 820_3 joins the thread as student C, in accordance with instructions from student 830_3 (student C).

[0129] In step S1008, terminal 820_1, in response to instructions from student 830_1 (student A), selects a tutor name (in this case, "Lecture α Tutor") from the tutor name list and sends it to server device 110.

[0130] In step S1009, the lecture support unit 810 of the server device 120 accepts the selection of the tutor name transmitted to the server device 110 by terminal 820_1.

[0131] In step S1010, the lecture support unit 810 of the server device 120 associates the assistant with the selected tutor name with a thread. Thereafter, the assistant associated with the thread will respond to the input information entered into that thread.

[0132] In step S1011, terminal 820_1 creates question (1) in accordance with the instructions of student 830_1 (student A).

[0133] In step S1012, terminal 820_1 inputs the created question (1) into the thread.

[0134] In step S1013, the lecture support unit 810 of the server device 120 retrieves the question (1) entered into the thread from the server device 110.

[0135] In step S1014, the lecture support unit 810 of the server device 120 identifies student A who entered question (1). The lecture support unit 810 of the server device 120 also assigns identification information (tag) to question (1) and sends it to the assistant associated with the thread, thereby instructing the creation of an answer to question (1).

[0136] In step S1015, the AI ​​chat service provider unit 131 of the server device 130 receives question (1) along with a prompt associated with the assistant.

[0137] In step S1016, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for lecture α) specified in the prompt associated with the assistant.

[0138] In step S1017, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registration data 323, creates an answer to question (1), and issues answer (1).

[0139] In step S1018, the lecture support unit 810 of server device 120 receives the answer (1) to question (1) issued by the AI ​​chat service provision unit 131 of server device 130.

[0140] In step S1021 of Figure 10B, the lecture support unit 810 of the server device 120 converts the received response (1) into a display format corresponding to the thread (in this case, the "Lecture α" thread 910).

[0141] In step S1022, the lecture support unit 810 of the server device 120 inputs the converted answer (1) into a thread provided by the communication service provision unit 111 of the server device 110 (in this case, the "Lecture α" thread 910).

[0142] In step S1023_1, terminal 820_1 displays the thread in which answer (1) was entered (in this case, thread 910 of "Lecture α").

[0143] In step S1023_2, terminal 820_2 displays the thread in which answer (1) was entered (in this case, thread 910 of "Lecture α").

[0144] In step S1023_3, terminal 820_3 displays the thread in which answer (1) was entered (in this case, thread 910 of "Lecture α").

[0145] In step S1024, terminal 820_2 creates question (2) in accordance with the instructions of student 830_2 (student B).

[0146] In step S1025, terminal 820_2 inputs the created question (2) into the thread.

[0147] In step S1026, the lecture support unit 810 of the server device 120 retrieves the question (2) entered into the thread from the server device 110.

[0148] In step S1027, the lecture support unit 810 of the server device 120 identifies student B who entered question (2). The lecture support unit 810 of the server device 120 also assigns identification information (tag) to question (2) and sends it to the assistant associated with the thread, thereby instructing the creation of an answer to question (2).

[0149] In step S1028, the AI ​​chat service provider unit 131 of the server device 130 receives question (2) along with a prompt associated with the assistant.

[0150] In step S1029, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for lecture α) specified in the prompt associated with the assistant.

[0151] In step S1030, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registration data 323, creates an answer to question (2), and issues answer (2).

[0152] In step S1031, the lecture support unit 810 of server device 120 receives the answer (2) to question (2) issued by the AI ​​chat service provision unit 131 of server device 130.

[0153] In step S1032, the lecture support unit 810 of the server device 120 converts the received response (2) into a display format corresponding to the thread (in this case, the "Lecture α" thread 910).

[0154] In step S1033, the lecture support unit 810 of the server device 120 inputs the converted answer (2) into a thread provided by the communication service provision unit 111 of the server device 110 (in this case, the "Lecture α" thread 910).

[0155] In step S1034_1, terminal 820_1 displays the thread in which answer (2) was entered (in this case, thread 910 of "Lecture α").

[0156] In step S1034_2, terminal 820_2 displays the thread in which answer (2) was entered (in this case, thread 910 of "Lecture α").

[0157] In step S1034_3, terminal 820_3 displays the thread in which answer (2) was entered (in this case, thread 910 of "Lecture α").

[0158] In step S1041 of Figure 10C, terminal 820_2 creates question (3) in accordance with the instructions of student 830_2 (student C).

[0159] In step S1042, terminal 820_3 inputs the created question (3) into the thread.

[0160] In step S1043, the lecture support unit 810 of the server device 120 retrieves the question (3) entered into the thread from the server device 110.

[0161] In step S1044, the lecture support unit 810 of the server device 120 identifies student C who entered question (3). The lecture support unit 810 of the server device 120 also assigns identification information (tag) to question (3) and sends it to the assistant associated with the thread, thereby instructing the creation of an answer to question (3).

[0162] In step S1045, the AI ​​chat service provider unit 131 of the server device 130 receives question (3) along with a prompt associated with the assistant.

[0163] In step S1046, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for lecture α) specified in the prompt associated with the assistant.

[0164] In step S1047, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registration data 323, creates an answer to question (3), and issues answer (3).

[0165] In step S1048, the lecture support unit 810 of server device 120 receives the answer (3) to question (3) issued by the AI ​​chat service provision unit 131 of server device 130.

[0166] In step S1049, the lecture support unit 810 of the server device 120 converts the received response (3) into a display format corresponding to the thread (in this case, the "Lecture α" thread 910).

[0167] In step S1050, the lecture support unit 810 of the server device 120 inputs the converted answer (3) into a thread provided by the communication service provision unit 111 of the server device 110 (in this case, thread 10 of "Lecture α").

[0168] In step S1051_1, terminal 820_1 displays the thread in which answer (3) was entered (in this case, thread 10 of "Lecture α").

[0169] In step S1051_2, terminal 820_2 displays the thread in which answer (3) was entered (in this case, thread 10 of "Lecture α").

[0170] In step S1051_3, terminal 820_3 displays the thread in which answer (3) was entered (in this case, thread 10 of "Lecture α").

[0171] <Specific example of the display screen> Next, a specific example of the display screen of thread 910 when "lecture processing" is performed in system 800 according to the first embodiment will be described. Figure 11 is a diagram showing a specific example of the display screen. Lecture α: Linear Algebra, • Assistant: Assistant to the tutor named "Linear Algebra Tutor" ·Participating students: Student A, B, This shows a specific example of the display screen for thread 910 in this case.

[0172] In Figure 11, symbol 1101 indicates that the assistant of tutor "Linear Algebra Tutor" has given students A and B problems 1 and 2 to check their understanding of linear algebra.

[0173] In Figure 11, reference numeral 1102 indicates that student A submitted their answer to problem 1.

[0174] In Figure 11, symbol 1103 indicates that the assistant of tutor "Linear Algebra Tutor" determined that the answer submitted by student A to problem 1 was incorrect and prompted student A to reconsider.

[0175] In Figure 11, reference numeral 1104 indicates that student B submitted their answer to problem 1.

[0176] In Figure 11, symbol 1105 indicates that the assistant of tutor "Linear Algebra Tutor" determined that the solution submitted by student B to problem 1 was incorrect, and then lectured students A and B on how to solve problem 1.

[0177] <Summary> As is clear from the above description, the server device 120 according to the first embodiment is • When information is entered into a thread provided by the Communication Service Provider Department 111, which allows participation by multiple students, the system retrieves that input information. The AI ​​chat service provider unit 131, which operates using an assistant that conducts designated lectures under the role of a tutor, is instructed to create a response based on the input information. • Retrieve the response created by the AI ​​chat service provider unit 131 and control the system to input the retrieved response into the thread.

[0178] Thus, the server device 120 according to the first embodiment pre-configures an assistant that will deliver the designated lecture. Furthermore, the server device 120 according to the first embodiment has the AI ​​chat service provider 131 participate in the thread provided by the communication service provider 111 in the role of a tutor and assist in delivering lectures to students. As a result, according to the first embodiment, a system for delivering lectures to students can be realized.

[0179] [Second Embodiment] In the first embodiment described above, an assistant for conducting a designated lecture is pre-configured, and the AI ​​chat service provider 131 participates in a thread provided by the communication service provider 111 to assist in conducting lectures for students.

[0180] In contrast, the server device 120 according to the second embodiment has an assistant pre-configured to perform translation into a specified language. Furthermore, in the second embodiment, a case will be described in which the AI ​​chat service provider 131 participates in a thread provided by the communication service provider 111 in the role of a translator, supporting a meeting with participants speaking different languages.

[0181] <System configuration during the preparation phase> First, the system configuration of the system according to the second embodiment will be described. The system according to the second embodiment is a system for participants speaking different languages ​​to conduct a meeting. • In the "preparation phase," preparatory processes such as building the assistant, registering and modifying registration data are performed. In the "execution phase," the generated AI, acting as a translator, uses the built-in assistant and registered and modified registration data to translate the input information entered by participants, thereby facilitating "meeting processing" among participants speaking different languages.

[0182] Here, we will first describe the system configuration that performs the preparation process in the preparation phase. Figure 12 is a diagram showing an example of the system configuration in the preparation phase of the system according to the second embodiment.

[0183] As shown in Figure 12, the system 1200 in the preparation phase includes server device 1210, server device 130, and terminals 1240_1 and 1240_2. In the system 1200 in the preparation phase, server device 1210 is connected to server device 130 and terminals 1240_1 and 1240_2 via network 160 so as to be able to communicate.

[0184] The server device 1210 functions as a communication service provision unit 1211 and a translation tuning unit 1212 by executing a communication service provision program and a translation tuning program during the preparation phase.

[0185] The communication service provider unit 1211 provides a communication service that allows multiple users to participate and communicate with each other. The communication service provider unit 1211 is implemented by software such as Discord.

[0186] The translation tuning unit 1212 executes the "construction process" and the "tuning process" among the various processes included in the preparation process. In the "construction process", the translation tuning unit 1212 performs the following: • In order to operate the AI ​​chat service provider unit 131 using its assistant function, an "assistant" is generated to act as a translator in a meeting with participants speaking different languages. • Register the necessary data for translation in meetings with participants speaking different languages ​​in the AI ​​chat service provider unit 131.

[0187] Furthermore, in the "tuning process," the translation tuning unit 1212 is: The AI ​​chat service provider unit 131 is operated in the role of a translator, and the registered data is modified through a meeting with simulated participants speaking different languages.

[0188] The server device 130 functions as an AI chat service provider unit 131 by executing an AI chat service provider program. The AI ​​chat service provider unit 131 provides a chat service by performing language processing using a large-scale language model. The AI ​​chat service provider unit 131 is implemented by a generative AI such as ChatGPT.

[0189] As described above, the AI ​​chat service provider unit 131 has an assistant function. The assistant function is a function that causes the AI ​​chat service provider unit 131 to operate in order to perform specified tasks under a predetermined role, and the assistant has a defined role, tasks, and methods for performing those tasks.

[0190] In the case of a meeting with participants who speak different languages, the assistant will be: • Each participant will be assigned the role of a translator, responsible for translating the input information they submit. The task is defined to translate input information entered in different languages ​​using registered data. • The specific method for performing the task is defined.

[0191] Terminals 1240_1 and 1240_2 are terminals operated by generators 1250_1 and 1250_2. During the preparation phase, generators 1250_1 and 1250_2 input various instructions to the server device 1210 via terminals 1240_1 and 1240_2 to perform tasks such as generating assistants, registering and modifying registration data. Generators 1250_1 and 1250_2 are composed of, for example, translators who perform translations between different languages.

[0192] <Overview of the construction process by the translation tuning unit> Next, an overview of the "construction process" performed by the translation tuning unit 1212 of the server device 1210 will be described. As mentioned above, in the "construction process", the translation tuning unit 1212 performs the following: • In order to operate the AI ​​chat service provider unit 131 using its assistant function, an "assistant" is generated to act as a translator in a meeting with participants speaking different languages. • Register the necessary data for translation in meetings with participants speaking different languages ​​in the AI ​​chat service provider unit 131.

[0193] Figure 13 shows an overview of the construction process by the translation tuning unit. As shown in Figure 13, the translation tuning unit 1212 further includes an assistant generation unit 1321, a prompt generation unit 1322, a prompt correction unit 1323, and an input control unit 1324. During the construction process, the assistant generation unit 1321 and the prompt generation unit 1322 are in operation.

[0194] The assistant generation unit 1321 generates an assistant using the API 322 provided by the AI ​​chat service provision unit 131. An example is shown in Figure 13. • Name: Translator α, Instructions: Instructions for translation α, This shows how the assistants are generated. The assistant generation unit 1321 generates, for example, a number of assistants corresponding to the number of language combinations.

[0195] In the construction process, the prompt generation unit 1322 registers the registration data 323 with the AI ​​chat service provision unit 131 using prompts. The example in Figure 13 shows how the registration data for translation α is registered as the registration data necessary to realize translation α. ​​The registration data for translation α includes dictionary data for performing translation between specific languages ​​(for example, dictionary data for translating from language A to language B, dictionary data for translating from language B to language A), etc. The prompt generation unit 1322 generates, for example, a number of translation registration data corresponding to the number of language combinations.

[0196] <Overview of the tuning process by the translation tuning unit> Next, we will explain the outline of the "tuning process" performed by the translation tuning unit 1212 of the server device 1210 during the preparation phase. As described above, the translation tuning unit 1212 operates the AI ​​chat service provider unit 131 in the role of a translator and performs a "tuning process" to modify the registered data through a meeting of simulated participants speaking different languages.

[0197] Figure 14 shows an overview of the tuning process performed by the translation tuning unit. As shown in Figure 14, the translation tuning unit 1212 further includes an assistant generation unit 1321, a prompt generation unit 1322, a prompt correction unit 1323, and an input control unit 1324. During the tuning process, the prompt generation unit 1322, the prompt correction unit 1323, and the input control unit 1324 are in operation.

[0198] Furthermore, when the translation tuning unit 1212 starts the tuning process, the communication service provision unit 1211 of the server device 1210, • The first communication service provider unit 1311 provides thread 1410 (an example of the first thread) for language A of "Agenda I". • The second communication service provider unit 1312 provides thread 1420 (an example of a second thread) for language B of "Agenda Item I". It shall be assumed that in thread 1410 for language A of "Agenda I", the assistant with translator name "Translator α" is selected. Furthermore, in thread 1420 for language B of "Agenda I", the assistant with translator name "Translator α" is selected.

[0199] Also, thread 1410 for language A of "Agenda I" is: • A producer A, whose native language is Language A (an example of the first language), participates as a pseudo-participant A. • A translator of producer B, whose native language is language B (an example of a second language), is participating under the translator name "Translator α". It shall be considered as such.

[0200] Similarly, thread 1420 for language B of "Agenda I" is: • A producer B whose native language is language B (an example of a second language) participates as a pseudo-participant B. • A translator of a producer A whose native language is Language A (an example of the first language) is participating under the translator name "Translator α". It shall be considered as such.

[0201] During the tuning process, the prompt generation unit 1322 acquires input information when a simulated participant A inputs information in language A into the thread 1410 for language A of "Agenda I". The prompt generation unit 1322 also sends the acquired input information to the AI ​​chat service provider unit 131 of the server device 130, instructing it to generate response information based on the input information.

[0202] The example in Figure 14 shows how the prompt generation unit 1322 acquires a simulated statement (1) entered by simulated participant A into the thread 1410 for language A of "Agenda I," and sends it to the AI ​​chat service provider unit 131 to instruct it to translate the simulated statement (1).

[0203] Furthermore, during the tuning process, the prompt generation unit 1322 acquires the input information when a simulated participant B inputs information in language B into the thread 1420 for language B of "Agenda I". The prompt generation unit 1322 sends the acquired input information to the AI ​​chat service provider unit 131 of the server device 130, thereby instructing it to generate response information based on the input information.

[0204] The example in Figure 14 shows how the prompt generation unit 1322 acquires a simulated statement (2) entered by simulated participant B into the thread 1420 for language B of "Agenda I," and sends it to the AI ​​chat service provider unit 131 to instruct it to translate the simulated statement (2).

[0205] During the tuning process, when the prompt correction unit 1323 receives request information requesting a prompt correction in the thread 1420 for language B of "Agenda I", it sends the request information to the AI ​​chat service provider unit 131 of the server device 130.

[0206] The example in Figure 14 shows how the prompt modification unit 1323 receives request information for prompt modification, which was entered by simulated participant B into thread 1420 for language B of "Agenda I," and sends it to the AI ​​chat service provider unit 131.

[0207] During the tuning process, when request information requesting a prompt correction is input to the thread 1410 for language A of "Agenda I", the prompt correction unit 1323 sends the request information to the AI ​​chat service provider unit 131 of the server device 130.

[0208] The example in Figure 14 shows how the prompt modification unit 1323 receives request information for prompt modification, which was entered by simulated participant A into thread 1410 for language A of "Agenda I," and sends it to the AI ​​chat service provider unit 131.

[0209] During the tuning process, the input control unit 1324 receives a translation result (an example of response information) sent from the AI ​​chat service provider unit 131 of the server device 130 in response to input information being sent by the prompt generation unit 1322. Alternatively, the input control unit 1324 receives a response result (an example of response information) sent from the AI ​​chat service provider unit 131 of the server device 130 in response to request information being sent by the prompt correction unit 1323.

[0210] The input control unit 1324 converts the received translation result or received correspondence result into a display format corresponding to the thread 1410 for language A or the thread 1420 for language B of "Agenda Item I". The input control unit 1324 inputs the converted translation result or converted correspondence result into the thread 1410 for language A or the thread 1420 for language B of "Agenda Item I".

[0211] The example in Figure 14 shows how the input control unit 1324 receives the translation result of a simulated statement (1) entered by simulated participant A into thread 1410 for language A of "Agenda I," and how the translator inputs it into thread 1420 for language B of "Agenda I." The example in Figure 14 also shows how the input control unit 1324 receives the translation result of a simulated statement (2) entered by simulated participant B into thread 1420 for language B of "Agenda I," and how the translator inputs it into thread 1410 for language A of "Agenda I."

[0212] Furthermore, the example in Figure 14 shows how the input control unit 1324 receives the response result (correction complete) to the request information for prompt correction input by pseudo-participant B to thread 1420 for language B of "Agenda Item I". Also, the example in Figure 14 shows how the input control unit 1324 inputs the received response result (correction complete) to thread 1420 for language B of "Agenda Item I".

[0213] Furthermore, the example in Figure 14 shows how the input control unit 1324 receives the response result (correction complete) to the request information for prompt correction that was entered by pseudo-participant A into thread 1410 for language A of "Agenda I". Also, the example in Figure 14 shows how the input control unit 1324 inputs the received response result (correction complete) into thread 1410 for language A of "Agenda I".

[0214] <Preparation Process Flow> Next, we will explain the flow of the "preparation process" in the preparation phase by the entire system 1200. Figures 15A to 15E are the first to fifth diagrams showing the flow of the preparation process in the system according to the second embodiment.

[0215] It should be assumed that the "construction process" by the translation tuning unit 1212 has already been completed before performing the preparation process shown in Figures 15A to 15E. In other words, the preparation process shown in Figures 15A to 15E will be explained primarily in terms of the "tuning process" by the translation tuning unit 1212.

[0216] As shown in Figure 15A, in step S1501, terminal 1240_1 issues a Chat command in response to input from generator 1250_1 (generator A) and sends it to server device 1210.

[0217] In step S1502, the communication service provision unit 1211 of the server device 1210 generates threads (in this case, a thread for language A of "Agenda I" and a thread for language B of "Agenda I").

[0218] In step S1503, the translation tuning unit 1212 of the server device 1210 acquires the Chat command sent by terminal 1240_1.

[0219] In step S1504, the translation tuning unit 1212 of the server device 1210 queries the server device 130 for a list of translator names, which is a list of assistants who can participate as translators in the generated thread.

[0220] In step S1505, the AI ​​chat service provider unit 131 of the server device 130 retrieves the translator name of the assistant whose translator role is defined among the assistants registered in API 322.

[0221] In step S1506, the AI ​​chat service provision unit 131 of the server device 130 transmits the acquired translator name to the server device 1210.

[0222] In step S1507, the translation tuning unit 1212 of the server device 120 notifies the communication service provider unit 1211 of the received list of translator names as a translator name list. As a result, the translator name list is posted to the thread provided by the communication service provider unit 1211 (in this case, thread 1410 for language A of "Agenda I").

[0223] In step S1508, terminal 1240_1 joins the thread (thread 1410 for language A of "Agenda I") as pseudo-participant A, in accordance with instructions from generator 1250_1 (generator A). Although not shown in Figure 15A, terminal 1240_2 joins the thread (thread 1420 for language B of "Agenda I") as pseudo-participant B, in accordance with instructions from generator 1250_2 (generator B).

[0224] In step S1509, terminal 1240_1, in response to instructions from generator 1250_1 (generator A), selects one translator name (in this case, "Translator α") from the list of translator names and sends it to server device 1210.

[0225] In step S1510, the translation tuning unit 1212 of the server device 1210 accepts the selection of the translator name transmitted by terminal 1240_1.

[0226] In step S1511, the translation tuning unit 1212 of the server device 1210 associates the assistant with the selected translator name with a thread (thread 1410 for language A of "Agenda I", thread 1420 for language B of "Agenda I"). Thereafter, the assistant associated with the thread will handle the input information entered into that thread.

[0227] In step S1521 of Figure 15B, terminal 1240_1 creates a pseudo-speech (1) in language A in accordance with the instructions of generator 1250_1 (generator A).

[0228] In step S1522, terminal 1240_1 inputs the created pseudo-message (1) into the thread (thread 1410 for language A of "topic I").

[0229] In step S1523, the translation tuning unit 1212 of the server device 1210 acquires the pseudo-message (1) entered into the thread (thread 1410 for language A of "topic I").

[0230] In step S1524, the translation tuning unit 1212 of the server device 1210 identifies the pseudo-participant A who input the pseudo-speech (1) and assigns identification information (tag) to the pseudo-speech (1). The translation tuning unit 1212 of the server device 1210 sends the pseudo-speech (1) with the assigned identification information (tag) to the assistant associated with the thread (thread 1410 for language A of "Agenda I"), thereby instructing the translation of the pseudo-speech (1).

[0231] In step S1525, the AI ​​chat service provider unit 131 of the server device 130 receives a simulated statement (1) along with a prompt associated with the assistant.

[0232] In step S1526, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for translation α, such as dictionary data) specified in the prompt associated with the assistant.

[0233] In step S1527, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registered data 323, creates a translation of the simulated statement (1) into language B, and issues the translation result (1).

[0234] In step S1528, the translation tuning unit 1212 of the server device 120 receives the translation result (1) for the simulated statement (1) issued by the AI ​​chat service provision unit 131 of the server device 130.

[0235] In step S1529, the translation tuning unit 1212 of the server device 120 converts the received translation result (1) into a display format corresponding to the thread (in this case, the thread 1420 for language B of "Agenda I").

[0236] In step S1530, the translation tuning unit 1212 of the server device 120 inputs the converted translation result (1) into a thread provided by the communication service provision unit 1211 (in this case, thread 1420 for language B of "Agenda I").

[0237] In step S1531, terminal 1240_2 displays the thread in which the translation result (1) was entered (in this case, thread 1420 for language B of "Agenda I").

[0238] In step S1541 of Figure 15C, terminal 1240_2 creates a pseudo-statement (2) in accordance with the instructions of generator 1250_2 (generator B).

[0239] In step S1542, the created pseudo-speech (2) is entered into the thread (in this case, thread 1420 for language B of "Agenda I").

[0240] In step S1543, the translation tuning unit 1212 of the server device 1210 acquires the pseudo-message (2) entered into the thread (in this case, thread 1420 for language B of "Agenda I").

[0241] In step S1544, the translation tuning unit 1212 of the server device 1210 identifies the simulated participant B who input the simulated statement (2) and assigns identification information (tag) to the simulated statement (2). The translation tuning unit 1212 of the server device 1210 instructs the assistant to translate the simulated statement (2) by sending the simulated statement (2) with the assigned identification information (tag) to the assistant associated with the thread (in this case, the thread 1420 for language B of "Agenda I").

[0242] In step S1545, the AI ​​chat service provider unit 131 of the server device 130 receives a simulated statement (2) along with a prompt associated with the assistant.

[0243] In step S1546, the AI chat service providing unit 131 of the server device 130 refers to the registration data 323 (here, the registration data for translation α such as dictionary data) defined in the prompt associated with the assistant.

[0244] In step S1547, the AI chat service providing unit 131 of the server device 130 refers to the registration data 323 based on the Instruction defined for the assistant, creates a translation of the pseudo utterance (2) into language A, and issues the translation result (2).

[0245] In step S1548, the translation tuning unit 1212 of the server device 120 receives the translation result (2) of the pseudo utterance (2) into language A issued by the AI chat service providing unit 131 of the server device 130.

[0246] In step S1549, the translation tuning unit 1212 of the server device 120 converts the received translation result (2) into a display mode according to the thread (here, the thread 1410 for language A of "topic I").

[0247] In step S1550, the translation tuning unit 1212 of the server device 120 inputs the converted translation result (2) into the thread (here, the thread 1410 for language A of "topic I") provided by the communication service providing unit 1211.

[0248] In step S1551, the terminal 1240_1 displays the thread (here, the thread 1410 for language A of "topic I") into which the translation result (2) is input.

[0249] In step S1561 of FIG. 15D, the terminal 1240_2 generates a build command and an update command according to the instruction of the generator 1250_2 (generator B), and inputs the request information for requesting the modification of the prompt associated with the assistant into the thread.

[0250] In step S1562, the translation tuning unit 1212 of the server device 1210 acquires the request information input to the thread and transmits it to the server device 130.

[0251] In step S1563, the AI chat service providing unit 131 of the server device 130 corrects the prompt associated with the assistant.

[0252] In step S1564, the AI chat service providing unit 131 of the server device 130 transmits to the server device 1210 that the correction of the prompt has been completed.

[0253] In step S1565, the translation tuning unit 1212 of the server device 120 inputs the response result (correction completed) for the request information into the thread provided by the communication service providing unit 1211 (here, the thread for language B of "topic I").

[0254] In step S1566, the terminal 1240_2 displays the thread (here, the thread 1420 for language B of "topic I") in which the correction completion is input.

[0255] In step S1567, the terminal 1240_1 creates a pseudo utterance (1) according to the instruction of the generator 1250_1 (generator A).

[0256] In step S1568, the terminal 1240_1 inputs the created pseudo utterance (1) into the thread (here, the thread 1420 for language A of "topic I").

[0257] In step S1569, the translation tuning unit 1212 of the server device 1210 acquires the pseudo utterance (1) input to the thread.

[0258] In step S1570, the translation tuning unit 1212 of the server device 1210 identifies the pseudo-participant A who input the pseudo-speech (1) and assigns identification information (tag) to the pseudo-speech (1). The translation tuning unit 1212 of the server device 1210 instructs the assistant to translate the pseudo-speech (1) by sending the pseudo-speech (1) with the assigned identification information (tag) to the assistant associated with the thread (in this case, the thread 1410 for language A of "Agenda I").

[0259] In step S1571, the AI ​​chat service provider unit 131 of the server device 130 receives a simulated statement (1) along with a modified prompt associated with the assistant.

[0260] In step S1572, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for translation α, such as dictionary data) specified in the modified prompt that is associated with the assistant.

[0261] In step S1573, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registered data 323, creates a corrected translation of the simulated statement (1) into language B, and issues the translation result (1)'.

[0262] In step S1574, the translation tuning unit 1212 of the server device 1210 receives the translation result (1)' of the simulated statement (1) into language B, which is issued by the AI ​​chat service provision unit 131 of the server device 130.

[0263] In step S1575, the translation tuning unit 1212 of the server device 1210 converts the received translation result (1)' into a display format corresponding to the thread (in this case, the thread 1420 for language B of "Agenda I").

[0264] In step S1576, the translation tuning unit 1212 of the server device 1210 inputs the converted translation result (1)' into a thread provided by the communication service provision unit 1211 (in this case, thread 1420 for language B of "Agenda I").

[0265] In step S1577, terminal 1240_2 displays the thread into which the translation result (1)' was entered (in this case, thread 1420 for language B of "Agenda I").

[0266] In step S1581 of Figure 15E, terminal 1240_1 generates build commands and update commands in response to instructions from generator 1250_1 (generator A), and inputs request information to the thread requesting correction of the prompt associated with the assistant.

[0267] In step S1582, the translation tuning unit 1212 of the server device 1210 acquires the request information entered into the thread and sends it to the server device 130.

[0268] In step S1583, the AI ​​chat service provider unit 131 of the server device 130 modifies the prompt associated with the assistant.

[0269] In step S1584, the AI ​​chat service provision unit 131 of the server device 130 sends a message to the server device 1210 indicating that the prompt correction is complete.

[0270] In step S1585, the translation tuning unit 1212 of the server device 120 inputs the response result (correction complete) to the request information into a thread provided by the communication service provision unit 1211 (in this case, thread 1410 for language A of "Agenda I").

[0271] In step S1586, terminal 1240_1 displays the thread in which correction completion was entered (in this case, thread 1410 for language A of "Agenda I").

[0272] In step S1587, the terminal 1240_2 creates a pseudo utterance (2) according to the instruction of the generator 1250_2 (generator B).

[0273] In step S1588, the terminal 1240_2 inputs the created pseudo utterance (2) into the thread (here, the thread 1420 for language B of "topic I").

[0274] In step S1589, the translation tuning unit 1212 of the server device 1210 acquires the pseudo utterance (2) input into the thread.

[0275] In step S1590, the translation tuning unit 1212 of the server device 1210 identifies the pseudo participant B who input the pseudo utterance (2), and assigns identification information (tag) to the pseudo utterance (2). The translation tuning unit 1212 of the server device 1210 transmits the pseudo utterance (2) with the assigned identification information (tag) to the assistant associated with the thread (here, the thread 1420 for language B of "topic I") to instruct the translation of the pseudo utterance (2).

[0276] In step S1591, the AI chat service providing unit 131 of the server device 130 receives the pseudo utterance (2) together with the corrected prompt associated with the assistant.

[0277] In step S1592, the AI chat service providing unit 131 of the server device 130 refers to the registered data 323 (here, the registered data for translation α such as dictionary data, etc.) defined in the corrected prompt associated with the assistant.

[0278] In step S1593, the AI chat service providing unit 131 of the server device 130 refers to the registered data 323 based on the Instruction defined for the assistant, creates a corrected translation into language A for the pseudo utterance (2), and issues a translation result (2)'.

[0279] In step S1594, the translation tuning unit 1212 of the server device 1210 receives the translation result (2)' of the simulated statement (2) into language A, which is issued by the AI ​​chat service provision unit 131 of the server device 130.

[0280] In step S1595, the translation tuning unit 1212 of the server device 1210 converts the received translation result (2)' into a display format corresponding to the thread (in this case, thread 1410 for language A of "Agenda I").

[0281] In step S1596, the translation tuning unit 1212 of the server device 1210 inputs the converted translation result (2)' into a thread provided by the communication service provision unit 1211 (in this case, thread 1410 for language A of "Agenda I").

[0282] In step S1597, terminal 1240_1 displays the thread into which the translation result (2)' was entered (in this case, thread 1410 for language A of "Agenda I").

[0283] <System configuration during the execution phase> Next, the system configuration in the execution phase of the system according to the second embodiment will be described. Figure 16 is a diagram showing an example of the system configuration in the execution phase of the system according to the second embodiment.

[0284] As shown in Figure 16, the system 1600 in the execution phase includes server device 1210, server device 130, terminals 1620_1 to 1620_n, and terminals 1620_n+1 to 1620_n+m. In the system 1600 in the execution phase, server device 1210 is connected to server device 130 via network 160 for communication. Also in the system 1600 in the execution phase, server device 1210 is connected to terminals 1620_1 to 1620_n and terminals 1620_n+1 to 1620_n+m via network 160 for communication.

[0285] During the execution phase, the server device 1210 functions as a communication service provision unit 1211 and a translation support unit 1612 by executing a communication service provision program and a translation support program. Details of the communication service provision unit 1211 have already been explained using Figure 12, so their explanation is omitted here.

[0286] Translation support unit 1612, • For threads provided by Communication Service Provider 1211, AI Chat Service Provider 131 participates in the role of a translator and assists in translating participants' statements. Execute "Translation support processing".

[0287] The server device 130 functions as the AI ​​chat service provider unit 131. In the execution phase, the AI ​​chat service provider unit 131 is the same unit that underwent "construction processing" and "tuning processing" by the server device 120 during the preparation phase. Details of the AI ​​chat service provider unit 131 have already been explained using Figure 12, so the explanation will be omitted here.

[0288] <Overview of translation support processing by the Translation Support Department> Next, we will explain the outline of the "translation support processing" performed by the translation support unit 1612 of the server device 120 during the execution phase. As mentioned above, the translation support unit 1612 performs the following: • For threads provided by Communication Service Provider 1211, AI Chat Service Provider 131 participates in the role of a translator and assists in translating participants' statements. Execute "Translation support processing".

[0289] Figure 17 is a diagram illustrating the overview of the translation support process performed by the translation support unit. As shown in Figure 17, the translation support unit 1612 further includes a prompt generation unit 1711 and an input control unit 1712.

[0290] Furthermore, when the translation support unit 1612 starts the translation support process, the communication service provision unit 1211 provides thread 1710 for language A of "Agenda I" and thread 1720 for language B of "Agenda I". In addition, in thread 1710 for language A of "Agenda I" and thread 1720 for language B of "Agenda I", participants are assumed to have selected an assistant with the translator name "Translator α". Moreover, in thread 1710 for language A of "Agenda I" and thread 1720 for language B of "Agenda I", for example, a total of two participants are assumed to be participating as participant A and participant B, respectively.

[0291] In the translation support process, the prompt generation unit 1711 acquires input information when information is entered by either participant into the thread 1710 for language A of "Agenda I" or the thread 1720 for language B of "Agenda I". The prompt generation unit 1711 transmits the acquired input information to the AI ​​chat service provider unit 131 of the server device 130, thereby instructing it to generate response information based on the input information.

[0292] The example in Figure 17 shows how the prompt generation unit 1711 acquires a statement (1) entered by participant A into thread 1710 for language A of "Agenda I," and a statement (2) entered by participant B into thread 1720 for language B of "Agenda I." The example in Figure 17 also shows how the prompt generation unit 1711 sends the acquired statements (1) and (2) to the AI ​​chat service provider unit 131, instructing it to translate statements (1) and (2).

[0293] In the translation support process, the input control unit 1712 receives the translation result (an example of response information) sent from the AI ​​chat service provision unit 131 of the server device 130 in response to the input information being sent by the prompt generation unit 1711.

[0294] The input control unit 1712 inputs the received translation result into either thread 1710 for language A of "Agenda I" or thread 1720 for language B of "Agenda I".

[0295] The example in Figure 17 shows how the input control unit 1712 receives the translation result (1) into language B of a statement (1) entered by participant A into thread 1710 for language A of "Agenda I," and inputs it into thread 1720 for language B of "Agenda I."

[0296] Furthermore, the example in Figure 17 shows how the input control unit 1712 receives the translation result (2) into language A of a statement (2) entered by participant B into thread 1720 for language B of "Agenda I," and inputs it into thread 1710 for language A of "Agenda I."

[0297] <Meeting Procedure Flow> Next, we will explain the flow of "conference processing" in the execution phase of the entire system 1600. Figures 18A to 18C are the first to third diagrams showing the flow of conference processing in the system according to the second embodiment.

[0298] As shown in Figure 18A, in step S1801, terminal 1620_1 issues a Chat command in response to input from participant 1630_1 (participant A) and sends it to server device 1210.

[0299] In step S1802, the communication service provision unit 1211 of the server device 1210 generates threads (in this case, a thread for language A of "Agenda I" and a thread for language B of "Agenda I").

[0300] In step S1803, the translation support unit 1612 of the server device 1210 receives the Chat command sent by terminal 1620_1.

[0301] In step S1804, the translation support unit 1612 of the server device 1210 queries the server device 130 for a list of translator names, which is a list of assistants who can participate as translators in the generated thread.

[0302] In step S1805, the AI ​​chat service provider unit 131 of the server device 130 retrieves the translator names of the assistants registered in API 322 whose translator roles are defined.

[0303] In step S1806, the AI ​​chat service provider unit 131 of the server device 130 transmits the acquired translator name to the server device 1210.

[0304] In step S1807, the translation support unit 1612 of the server device 1210 notifies the communication service provider unit 1211 of the received list of translator names as a translator name list. As a result, the translator name list is posted to the thread provided by the communication service provider unit 1211 (in this case, thread 1410 for language A of "Agenda I").

[0305] In step S1808, terminal 1620_1 joins the thread (thread 1410 for language A of "Agenda I") as participant A, in accordance with the instructions of participant 1630_1 (participant A). Although not shown in Figure 18A, terminal 1620_2 joins the thread (thread 1420 for language B of "Agenda I") as participant B, in accordance with the instructions of participant 1630_2 (participant B).

[0306] In step S1809, terminal 1620_1, in response to instructions from participant 1630_1 (participant A), selects one translator name (in this case, "Translator α") from the list of translator names and sends it to server device 1210.

[0307] In step S1810, the translation support unit 1612 of the server device 1210 receives the translator's name transmitted by terminal 1620_1.

[0308] In step S1811, the translation support unit 1612 of the server device 1210 associates the assistant with the selected translator name with a thread (thread 1410 for language A of "Agenda I", thread 1420 for language B of "Agenda I"). Thereafter, the assistant associated with the thread will handle the information entered into that thread.

[0309] In step S1821 of Figure 18B, terminal 1620_1 creates a statement (1) in language A in response to instructions from participant 1630_1 (participant A).

[0310] In step S1822, terminal 1620_1 inputs the created statement (1) into the thread (thread 1410 for language A of "topic I").

[0311] In step S1823, the translation support unit 1612 of the server device 1210 retrieves the message (1) entered into the thread (thread 1410 for language A of "topic I").

[0312] In step S1824, the translation support unit 1612 of the server device 1210 identifies participant A who entered the statement (1) and assigns identification information (tag) to the statement (1). The translation support unit 1612 of the server device 1210 instructs the assistant to translate the statement (1) by sending the statement (1) with the assigned identification information (tag) to the thread (thread 1410 for language A of "topic I").

[0313] In step S1825, the AI ​​chat service provider unit 131 of the server device 130 receives the message (1) along with a prompt associated with the assistant.

[0314] In step S1826, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for translation α, such as dictionary data) specified in the prompt associated with the assistant.

[0315] In step S1827, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registered data 323, creates a translation of the utterance (1) into language B, and issues the translation result (1).

[0316] In step S1828, the translation support unit 1612 of the server device 120 receives the translation result (1) for the statement (1) issued by the AI ​​chat service provision unit 131 of the server device 130.

[0317] In step S1829, the translation support unit 1612 of the server device 120 converts the received translation result (1) into a display format corresponding to the thread (in this case, the thread 1420 for language B of "Agenda I").

[0318] In step S1830, the translation support unit 1612 of the server device 120 inputs the converted translation result (1) into a thread provided by the communication service provision unit 1211 (in this case, thread 1420 for language B of "Agenda I").

[0319] In step S1831, terminal 1620_2 displays the thread into which the translation result (1) was entered (in this case, thread 1420 for language B of "Agenda I").

[0320] In step S1841 of Figure 18C, terminal 1620_2 creates a statement (2) in accordance with the instructions of participant 1630_2 (participant B).

[0321] In step S1842, the created statement (2) is entered into the thread (in this case, thread 1420 for language B of "Topic I").

[0322] In step S1843, the translation support unit 1612 of the server device 1210 retrieves the utterance (2) entered into the thread (in this case, thread 1420 for language B of "Agenda I").

[0323] In step S1844, the translation support unit 1612 of the server device 1210 identifies participant B who entered the statement (2) and assigns identification information (tag) to the statement (2). The translation support unit 1612 of the server device 1210 instructs the translation of the statement (2) by sending the statement (2) with the assigned identification information (tag) to the assistant associated with the thread (in this case, the thread 1420 for language B of "Agenda I").

[0324] In step S1845, the AI ​​chat service provider unit 131 of the server device 130 receives the message (2) along with a prompt associated with the assistant.

[0325] In step S1846, the AI ​​chat service provider unit 131 of the server device 130 refers to the registration data 323 (in this case, registration data for translation α, such as dictionary data) specified in the prompt associated with the assistant.

[0326] In step S1847, the AI ​​chat service provider unit 131 of the server device 130, under the instructions specified for the assistant, refers to the registered data 323, creates a translation of the utterance (2) into language A, and issues the translation result (2).

[0327] In step S1848, the translation support unit 1612 of the server device 120 receives the translation result (2) of the statement (2) into language A, which is issued by the AI ​​chat service provision unit 131 of the server device 130.

[0328] In step S1849, the translation support unit 1612 of the server device 120 converts the received translation result (2) into a display format corresponding to the thread (in this case, the thread 1410 for language A of "Agenda I").

[0329] In step S1850, the translation support unit 1612 of the server device 120 inputs the converted translation result (2) into a thread provided by the communication service provision unit 1211 (in this case, thread 1410 for language A of "Agenda I").

[0330] In step S1851, terminal 1620_1 displays the thread in which the translation result (2) was entered (in this case, thread 1410 for language A of "Agenda I").

[0331] <Summary> As is clear from the above description, the server device 1210 according to the second embodiment is - When information related to Language A is entered into a Language A thread provided by the Communication Service Provider Unit 1211, which allows participation by multiple participants, that input information is acquired. The AI ​​chat service provider unit 131, which operates using an assistant that translates from language A to language B under the role of a translator, is instructed to translate the input information into language B. The AI ​​chat service provider unit 131 retrieves the translation result of the input information into language B and controls it to input it into the language B thread.

[0332] Thus, the server device 1210 according to the second embodiment pre-configures an assistant that performs the specified translation. Furthermore, the server device 1210 according to the second embodiment has the AI ​​chat service provider 131 participate in the thread provided by the communication service provider 1211 in the role of a translator and assist in translating the participants' statements. As a result, according to the second embodiment, a system can be realized for participants speaking different languages ​​to hold a meeting.

[0333] [Third Embodiment] In the first and second embodiments described above, the communication service provider was described as providing a thread into which text information is entered as a communication space. However, the communication space provided by the communication service provider is not limited to a thread into which text information is entered.

[0334] For example, in the first embodiment, the communication space provided by the communication service provider 111 may be a thread into which voice is input.

[0335] Alternatively, the server device 1210 according to the second embodiment may include, for example, a virtual space providing unit that provides a virtual space in which participants participate as avatars, instead of the communication service providing unit 1211. In the third embodiment, the differences from the second embodiment will be explained below, focusing on the case in which a voice conference is processed by participants speaking different languages ​​in a virtual space in which participants participate as avatars.

[0336] <System configuration during the execution phase> First, we will describe the system configuration in the execution phase of the system according to the third embodiment. Figure 19 is a diagram showing an example of the system configuration in the execution phase of the system according to the third embodiment.

[0337] As shown in Figure 19, the system 1900 in the execution phase comprises server device 1210, server device 130, and terminals 1920_1 to 1920_n. In the system 1900 in the execution phase, server device 1210 is connected to server device 130 via network 160 for communication. Also in the system 1900 in the execution phase, server device 1210 is connected to terminals 1920_1 to 1920_n via network 160 for communication.

[0338] During the execution phase, the server device 1210 functions as a virtual space provision unit 1911 and a translation support unit 1612 by executing a virtual space provision program and a translation support program.

[0339] The virtual space provider unit 1911 provides a virtual space (an example of a communication space) for multiple participants to participate as avatars and communicate with each other.

[0340] Details of the translation support unit 1612 have already been explained using Figure 16, so the explanation will be omitted here. However, the translation support unit 1612 according to the third embodiment converts the translation result (an example of response information) into audio data and inputs it into the virtual space.

[0341] <Specific examples of terminal output information> Next, we will describe specific examples of output information output at terminals 1920_1 and 1920_2. Figure 20 shows a specific example of terminal output information. The example in Figure 20 shows a case where virtual space A, which is a virtual space for language A, and virtual space B, which is a virtual space for language B, are provided by the virtual space provider unit 1911. The example in Figure 20 also shows a situation where five participants have joined the virtual space provided by the virtual space provider unit 1911.

[0342] Of these, code 2010 indicates virtual space A as viewed by participant 1630_1 (participant A), whose native language is language A, on terminal 1920_1. As shown in code 2010, in virtual space A in which participant 1630_1 participates, four other participants participate as avatars. Also, as shown in code 2010, participant 1630_1 speaks in language A, and the avatars of the other four participants also speak in language A in virtual space A. However, the speech of the avatars of the other four participants includes speech that was originally in language B and translated into language A.

[0343] On the other hand, code 2020 represents virtual space B as viewed by participant 1630_2 (participant B), whose native language is language B, on terminal 1920_2. As shown in code 2020, in virtual space B in which participant 1630_2 participates, four other participants participate as avatars (similar to virtual space A). Also, as shown in code 2020, participant 1630_2 speaks in language B, and the avatars of the four other participants also speak in language B in virtual space B. However, the speech of the avatars of the four other participants includes speech that is translated from speech in language A into language B.

[0344] <Summary> As is clear from the above description, the server device 1210 according to the third embodiment is - When audio in language A is input into a virtual space provided by the virtual space provider unit 1911, in which multiple participants can participate as avatars, the input audio is acquired. The AI ​​chat service provider unit 131, which operates using an assistant that translates from language A to language B under the role of a translator, is instructed to translate the input speech into language B. The AI ​​chat service provider unit 131 retrieves the translation result of the input voice into language B and controls it to input it into the virtual space of language B.

[0345] Thus, the server device 1210 according to the third embodiment pre-configures an assistant that performs the specified translation. Furthermore, the server device 1210 according to the third embodiment has the AI ​​chat service provider 131 participate in the virtual space provided by the virtual space provider 1911 in the role of a translator and assist in translating the participants' statements. As a result, according to the third embodiment, a system can be realized for participants speaking different languages ​​to conduct a conference by voice.

[0346] [Other embodiments] In the first embodiment described above, the system 800 is • Server device 110 that functions as a communication service provision unit 111, • Server device 120 that functions as lecture support unit 810, • Server device 130 that functions as an AI chat service provider 131, The system was described as having the above features. However, the system configuration of system 800 is not limited thereto, and some or all of the functions of the communication service provision unit 111 may be implemented in the server device 120. Alternatively, some or all of the functions of the AI ​​chat service provision unit 131 may be implemented in the server device 120. Alternatively, some of the functions of the lecture support unit 810 may be implemented in the server device 110 or the server device 130.

[0347] Furthermore, in the second embodiment described above, the system 1600 is • Server device 1210 that functions as a communication service provision unit 1211 and a translation support unit 1612, • Server device 130 that functions as an AI chat service provider 131, The system was described as having the above features. However, the system configuration of system 1600 is not limited thereto, and some or all of the functions of the communication service provision unit 1211 and the translation support unit 1612 may be implemented in a server device other than server device 120. Alternatively, some or all of the functions of the AI ​​chat service provision unit 131 may be implemented in server device 1210.

[0348] Furthermore, in the third embodiment described above, the system 1900 is • Server device 1210 that functions as a virtual space provision unit 1911 and a translation support unit 1612, • Server device 130 that functions as an AI chat service provider 131, The system was described as having the above features. However, the system configuration of system 1900 is not limited thereto, and some or all of the functions of the virtual space provisioning unit 1911 and the translation support unit 1612 may be implemented in a server device other than the server device 1210. Alternatively, some or all of the functions of the AI ​​chat service provisioning unit 131 may be implemented in the server device 1210.

[0349] Furthermore, in the above embodiments, each functional unit (for example, the function of the lecture tuning unit 121) was described as being implemented by a single server device, but each functional unit may be implemented by multiple server devices.

[0350] It should be noted that the present invention is not limited to the configurations shown in the above embodiments, including combinations with other elements. These aspects can be modified without departing from the spirit of the present invention and can be appropriately determined according to their application. [Explanation of symbols]

[0351] 100: System 110: Server device 111: Communication Services Department 120: Server device 121: Tuning section for lectures 130: Server device 131: AI Chat Service Provider 311: Assistant Generation Unit 312: Prompt generation unit 313: Prompt Correction Section 314: Input Control Unit 321: Large-scale language models 322 :API 810: Lecture Support Department 811: Prompt generation unit 812: Input Control Unit 1200: System 1210: Server device 1211: Communication Services Department 1212: Translation Tuning Section 1311: 1st Communication Service Provider Department 1312: Second Communication Service Department 1321: Assistant Generation Unit 1322: Prompt generation unit 1323: Prompt Correction Section 1324: Input Control Unit 1600: System 1612: Translation Support Department 1711: Prompt generation unit 1712: Input Control Unit 1900: System 1911: Virtual Space Provision Department

Claims

1. In a communication space provided by a communication service, where multiple users can participate, when information is input by a user, a prompt generation unit instructs a generation AI, which operates using an assistant that acquires the information and performs a specified task under a predetermined role, to generate response information based on the information. An input control unit that acquires response information based on the information generated by the generation AI and controls the input of the response information into the communication field. A server device having the following features.

2. The input control unit controls the input of the converted response information, which is obtained by converting the response information based on the information into a manner appropriate to the communication setting. The server device according to claim 1.

3. The prompt generation unit assigns identification information to the information to identify each of the multiple users. The server device according to claim 2.

4. The aforementioned assistant has a defined method for performing tasks in which, under the role of a tutor who guides multiple students participating in a thread, he or she provides instruction on a specific subject using pre-registered data. The prompt generation unit instructs the generation AI, which operates using the assistant designated by the student among the multiple assistants, to generate response information based on the information. The server device according to claim 3.

5. The aforementioned registration data includes: Data registered during the preparation phase, or In the preparation phase, when the converted response information is input to the thread by the input control unit, and when a request information is input to the thread by a simulated student, the registered data is modified based on the request information to obtain the modified data. The server device according to claim 4, comprising at least one of the following.

6. The aforementioned registration data includes: This includes textbook data, lecture materials data, assignment data, or exam data corresponding to the aforementioned specific field. The server device according to claim 4.

7. The aforementioned assistant has a defined method for performing the task of translating information entered by multiple participants in a thread, using pre-registered data between specific languages, under the role of a translator. The prompt generation unit instructs the generation AI, which operates using a specified assistant among the multiple assistants, the translator designated by the participant, to generate response information based on the information. The server device according to claim 3.

8. The aforementioned communication service provides a first thread into which information is input in a first language, and a second thread into which information is input in a second language. When information is input to the first thread by the participant in the first language, the input control unit controls the input to the second thread based on the translated response information obtained by translating the information into the second language. When information is input to the second thread by the participant in the second language, the input control unit controls the input of the converted response information, which is the information translated into the first language, to the first thread. The server device according to claim 7.

9. The aforementioned registration data includes: Data registered during the preparation phase, or In the preparation phase, when the converted response information is input to the first thread or the second thread by the input control unit, and when a pseudo-participant inputs request information to the first thread or the second thread, the registered data is modified based on the request information to obtain the modified data. The server device according to claim 8, comprising at least one of the following.

10. The aforementioned registration data includes: The system includes either dictionary data for translating from the first language to the second language, or dictionary data for translating from the second language to the first language. The server device according to claim 8.

11. The communication space provided by the aforementioned communication service is a virtual space in which the multiple users can participate as avatars, and voice is input into the virtual space via the avatars. The server device according to claim 1.

12. A server device that provides a communication service in which multiple users participate in a communication space, A server device according to any one of claims 1 to 11, A system equipped with these features.

13. A server device according to any one of claims 1 to 11, A generative AI, which operates using an assistant that performs a designated task under a predetermined role based on information entered into a communication space provided by a communication service and in which multiple users can participate, generates response information using a server device and A system equipped with these features.

14. A server device that provides a communication service in which multiple users participate in a communication space, A server device according to any one of claims 1 to 11, A generative AI, operating with an assistant that performs a specified task under a predetermined role, generates response information based on information input into the communication environment, and a server device. A system equipped with these features.

15. Computers In a communication space provided by a communication service that allows participation by multiple users, when information is input by a user, the process of instructing a generative AI, which operates using an assistant that acquires the information and performs a specified task under a predetermined role, to generate response information based on the information, The process involves acquiring response information based on the information generated by the aforementioned generation AI, and controlling the input of the response information into the communication space. A method to support the execution of this task.

16. On the computer, In a communication space provided by a communication service that allows participation by multiple users, when information is input by a user, the process of instructing a generative AI, which operates using an assistant that acquires the information and performs a specified task under a predetermined role, to generate response information based on the information, The process involves acquiring response information based on the information generated by the aforementioned generation AI, and controlling the input of the response information into the communication space. A support program to enable execution.

Images