Apparatus, methods, and programs for supporting foreign language learning

Abstract

In supporting foreign language learning, this enables learning that is more tailored to specific situations that learners may face than before. [Solution] The device 100 causes the learner terminal 110 to input the learner's attributes and topic data related to the learning topic (S201-S204). The device 100 then causes the learner to further input a scene setting, which is the setting of the scene in which the learner wants to learn (S207-S208). Subsequently, the device 100 requests the generating AI model to generate teaching materials using the topic setting based on the topic data and at least one of the attributes included in the learner data or the learner's role included in the scene setting (S209), and can receive the generated teaching materials (S210). The device 100 then transmits the received teaching materials to the learner terminal 110 (S211).

Description

Technical Field

[0001] The present invention relates to an apparatus, a method, and a program therefor for assisting foreign language learning.

Background Art

[0002] In recent years, with the internationalization of the market, communication between people with different native languages has increased significantly. Especially in business, English functions as a common language. Therefore, the need to learn English is increasing more and more.

[0003] In order to meet such needs, English learning services assuming use in business are provided. For example, learning can be conducted by setting general situations such as meetings, presentations, and negotiations.

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, currently provided English learning services cannot conduct learning along with scenes that learners actually face or are highly likely to face. Therefore, it is not easy to use them in daily work and accumulate experience.

[0005] Similar problems also occur with languages such as Chinese, Spanish, Japanese, and Arabic other than English, and the present invention is applicable to languages other than English.

[0006] Also, for example, even in foreign language classes at school, it is currently not easy for students to conduct learning along with scenes that they actually face or are highly likely to face in the class.

[0007] The present invention has been made in view of the above points, and the problem thereof is to enable learning adapted to more specific scenes than those in the past that learners may face in an apparatus, a method, or a program for assisting foreign language learning.

[0008] In this specification, "scene" refers to the situation in which the learner finds themselves, and more details will be provided later. [Means for solving the problem]

[0009] To achieve this objective, a first aspect of the present invention is a method for supporting foreign language learning, comprising the steps of: receiving a designation of topic data related to a learning topic from a learner terminal used by a learner or an organization terminal used by an organization to which the learner belongs; receiving from the learner terminal at least one of learner data representing the learner's attributes and a scene setting or its identifier, which is a setting of a scene in which the learner wishes to learn; making a request (m) to a generating AI model m to generate teaching materials using the topic setting based on the topic data and the learner's role included in the learner data or the scene setting; and transmitting the generated teaching materials to the learner terminal.

[0010] Furthermore, a second aspect of the present invention is the method of the first aspect, wherein the designation is the designation of the URL of a web page containing the topic data.

[0011] Furthermore, a third aspect of the present invention is the method of the first aspect, wherein the designation is the designation of one or more files.

[0012] Furthermore, a fourth aspect of the present invention is the method of the first aspect, wherein the designation is the learner's input of the topic data.

[0013] Furthermore, a fifth aspect of the present invention is the method of the first aspect, wherein the designation is the designation of the application to be linked.

[0014] Furthermore, a sixth aspect of the present invention is the method of the fifth aspect, wherein the application is an online meeting tool.

[0015] Furthermore, a seventh aspect of the present invention is the method of the sixth aspect, wherein the topic setting includes speaker-specific utterances in a meeting.

[0016] Furthermore, an eighth aspect of the present invention is the method of the first aspect, wherein the designation includes the selection of any of a plurality of topic settings.

[0017] Furthermore, a ninth aspect of the present invention is a method according to any of the first to fifth aspects, wherein the topic is a business in which the learner is engaged or a class conducted in a foreign language that the learner receives.

[0018] Furthermore, a tenth aspect of the present invention is a method according to any of the first to ninth aspects, further comprising the step of making a request (k) to generate the topic setting for a generating AI model k using the topic data.

[0019] Furthermore, an eleventh aspect of the present invention is a method according to any of the first to tenth aspects, wherein the scene setting includes the role of the learner.

[0020] Furthermore, a twelfth aspect of the present invention is the method of the eleventh aspect, wherein the learner's role includes any of the two or three options selected for the content of the topic setting.

[0021] Furthermore, a thirteenth aspect of the present invention is a method according to any of the first to twelfth aspects, wherein the scene is a conversation scene, and the scene setting includes the roles of conversation partners.

[0022] Furthermore, a fourteenth aspect of the present invention is a method according to any of the first to twelfth aspects, wherein the scene is a conversation scene, and the scene setting includes a scenario for the conversation.

[0023] Furthermore, a fifteenth aspect of the present invention is a method according to any of the first to twelfth aspects, wherein the scene is a speaking, writing, or reading scene.

[0024] Also, a 16th aspect of the present invention is the method of the 15th aspect, wherein the scene setting includes the purpose of the speaking, writing, or reading.

[0025] Also, a 17th aspect of the present invention is the method of any one of the 1st to 16th aspects, further including a step of making a request (l) to generate the scene setting for the generation AI model l using the topic setting.

[0026] Also, an 18th aspect of the present invention is the method described in any one of the 1st to 16th aspects, wherein the designation of the topic data is received after the reception of the scene setting or its identifier.

[0027] Also, a 19th aspect of the present invention is a program for causing a computer to execute a method for assisting foreign language learning, the method including receiving a designation of topic data related to a topic for learning from a learner terminal used by a learner or an organization terminal used by an organization to which the learner belongs, receiving at least one of learner data representing the attributes of the learner and a scene setting or its identifier which is the setting of the scene in which the learner wishes to learn from the learner terminal, making a request (m) to generate teaching materials for the generation AI model m using the topic setting based on the topic data and the attributes included in the learner data or the role of the learner included in the scene setting, and transmitting the generated teaching materials to the learner terminal.

[0028] Further, a 20th aspect of the present invention is an apparatus for assisting foreign language learning, which receives a specification of topic data related to a topic for learning from a learner terminal used by a learner or an organization terminal used by an organization to which the learner belongs, and receives at least one of learner data representing the attributes of the learner and a scene setting which is a setting of a scene in which the learner wants to learn or an identifier thereof from the learner terminal, makes a request (m) to generate teaching materials for the generation AI model m using a topic setting based on the topic data and the attributes included in the learner data or the role of the learner included in the scene setting, and is configured to transmit the generated teaching materials to the learner terminal.

[0029] Further, a 21st aspect of the present invention is a method for assisting foreign language learning, which includes the steps of receiving a selection of a topic setting that the learner wants to learn from a learner terminal used by the learner or an organization terminal used by the organization to which the learner belongs, receiving at least one of a scene setting which is a setting of a scene in which the learner wants to learn or an identifier thereof from the learner terminal, making a request (m) to generate teaching materials for the generation AI model m using the role of the learner included in the topic setting and the scene setting, and transmitting the generated teaching materials to the learner terminal.

[0030] Further, a 22nd aspect of the present invention is the method of the 21st aspect, wherein the role of the learner is a position with respect to the description of the topic setting.

[0031] Further, a 23rd aspect of the present invention is the method of the 21st aspect, wherein the role of the learner is to ask questions, give explanations or conduct discussions about opinions regarding the description of the topic setting.

[0032] Furthermore, a 24th aspect of the present invention is a program for causing a computer to perform a method for supporting foreign language learning, the method comprising: receiving a selection of a topic setting that the learner wants to learn from a learner terminal used by the learner or an organization terminal used by the organization to which the learner belongs; receiving at least one of a scene setting, which is a setting of a scene in which the learner wants to learn, or an identifier thereof, from the learner terminal; making a request (m) to a generating AI model m to generate teaching materials using the topic setting and the role of the learner included in the scene setting; and transmitting the generated teaching materials to the learner terminal.

[0033] Furthermore, a 25th aspect of the present invention is a device for supporting foreign language learning, which receives a selection of a topic setting that the learner wants to learn from a learner terminal used by the learner or an organization terminal used by the organization to which the learner belongs, receives at least one of a scene setting, which is a setting of a scene in which the learner wants to learn, or its identifier, from the learner terminal, makes a request (m) to a generating AI model m to generate teaching materials using the topic setting and the role of the learner included in the scene setting, and transmits the generated teaching materials to the learner terminal. [Effects of the Invention]

[0034] According to one aspect of the present invention, by making requests to a generating AI model using topic settings based on topic data related to the learning topic, in addition to the learner's attributes or role, it becomes possible to generate practical teaching materials that are suited to specific scenarios that the learner may face. [Brief explanation of the drawing]

[0035] [Figure 1] This figure shows a system according to the first embodiment of the present invention. [Figure 2] This figure shows the flow of the method according to the first embodiment of the present invention. [Figure 3]This figure shows an example of topic setting according to the first embodiment of the present invention. [Figure 4] This figure shows another example of topic setting according to the first embodiment of the present invention. [Figure 5] This figure shows an example of a scene setting screen according to the first embodiment of the present invention. [Figure 6] This figure shows an example of a scene setting confirmation screen according to the first embodiment of the present invention. [Figure 7] This figure shows an example of instructions for generating teaching materials according to the first embodiment of the present invention. [Figure 8A] Figure 5 shows an example of teaching materials generated under the topic and scene settings shown. [Figure 8B] This figure shows an example of teaching materials generated when the topic setting is left blank under the scene setting shown in Figure 5. [Modes for carrying out the invention]

[0036] Embodiments of the present invention will be described in detail below with reference to the drawings.

[0037] (First Embodiment) Figure 1 shows a system according to one embodiment of the present invention. The device 100 for supporting foreign language learning communicates with a learner terminal 110 used by the learner and a platform 120 that provides a generative AI model via an IP network such as the Internet. The generative AI model is described exemplarily as being provided by the platform 120 that can communicate with the device 100, but it is also possible to run an application for providing a generative AI model on the device 100 so that the generative AI model is provided by the device 100.

[0038] The device 100 comprises a communication unit 101 such as a communication interface, a processing unit 102 such as a processor or CPU, and a storage unit 103 including a storage device or storage medium such as memory or a hard disk. The device can be configured by executing a program for performing each process or operation in the processing unit 102. The device 100 may include one or more devices, computers, or servers. The program may also include one or more programs and can be recorded on a computer-readable storage medium to form a non-transient program product. The program is stored in a storage device or storage medium such as a database 104 accessible from the storage unit 103 or the device 100 via an IP network, and instructions included in the program can be executed by at least one processor in the processing unit 102. The data described below as being stored in the storage unit 103 may be stored in a storage device or storage medium such as a database 104, and vice versa.

[0039] First, the device 100 transmits input screen display information to the learner terminal 110 to display an input screen for inputting the learner's attributes (S201). On the learner terminal 110, an input screen is displayed using the input screen display information, and the learner using the learner terminal 110 inputs their attributes on the input screen, and these attributes are transmitted from the learner terminal 110 to the device 100 (S202). Attributes can be entered by the learner entering text in the input field or by selecting from a selection menu.

[0040] The input screen display information can be transmitted, for example, as an HTML file, read by the web browser on the learner terminal 110, and displayed on the learner terminal 110's display screen. If a dedicated application is installed on the learner terminal 110, the input screen display information should include the data necessary for that application to display the input screen. Here, "input screen" can take various forms such as a web page, modal window, or popup window when displayed on a web browser, or it can be a screen of a dedicated application when displayed on that application. In any case, any screen that includes an area with input fields for entering necessary information is considered an input screen. Similar techniques can be applied to other screens mentioned in this specification.

[0041] Learner attributes can include occupation, industry, educational background, and academic field. Examples of occupations include sales, marketing, and research and development, while examples of industries include retail, manufacturing, and construction. Occupations can be classified according to occupational classifications such as ISCO, and industries according to industrial classifications such as NAICS.

[0042] Next, the device 100 transmits input screen display information to the learner terminal 110 to display an input screen for inputting topic data related to the topic the learner wishes to learn (S203). On the learner terminal 110, an input screen is displayed using the input screen display information, the learner specifies topic data on the input screen, and this specification is transmitted from the learner terminal 110 to the device 100 (S204). Examples of such topics include the business the learner is engaged in, or classes conducted in a foreign language that the learner is taking. The specification of topic data may also be made from an organizational terminal (not shown) used by the organization to which the learner belongs.

[0043] Topic data can be specified by the learner inputting the topic data, specifying the URL of a web page containing the topic data, specifying one or more files, or specifying an application to link via an API or other method. For example, by linking with an online meeting tool, audio or video data recorded from a meeting can be obtained as topic data. Alternatively, a list of multiple recordings can be obtained, and detailed data can be retrieved for the selected recording. Furthermore, by linking with a calendar tool, schedule data describing the learner's schedule can be obtained as topic data. An example of such a schedule might be "Proposal to ABC Corporation." Alternatively, a list of multiple schedules can be obtained, and detailed data can be retrieved for the selected schedule or for schedules that meet predetermined conditions. An example of a predetermined condition is that the time or period until the start date and time of the schedule is within a predetermined range. Additionally, topic data can be specified by selecting one of a set of topic settings displayed on the learner terminal 110.

[0044] Furthermore, the device 100 can obtain the web browser browsing history from the learner terminal 110 and use at least one of the number of views, viewing frequency, and viewing time calculated based on the browsing history to determine the URL of a web page containing topic data suitable for learning. Alternatively, a browser extension may be installed on the learner terminal 110 to analyze the browsing history and send at least one of the number of views, viewing frequency, and viewing time for each web page from the learner terminal 110 to the device 100, or send the URL of a learning-suitable page determined using the analysis results from the learner terminal 110 to the device 100. This allows the learner to obtain the topic settings described later without having to specify topic data, and promotes learning that is of high interest to the learner.

[0045] The device 100 can, if necessary, use the topic data to request the generating AI model (hereinafter also referred to as the "first generating AI model") to generate a topic setting (hereinafter also referred to as the "first request") (S205), and receive the generated topic setting (S206). More specifically, the device 100 can use the topic data to create an instruction to generate a topic setting (hereinafter also referred to as the "first instruction"), and make a first request including the first instruction to the generating AI model. The first instruction may include at least one of the number of characters and the format. The format may be one or more items included in the generated topic setting.

[0046] For example, if the topic data is specified as a URL or a file, the first instruction may be an instruction to summarize the content of the web page displayed at that URL or the content of that file. Alternatively, if the specified topic data is text entered by the learner or schedule data obtained from a calendar tool, the first instruction may be an instruction to provide an explanation, investigation, analysis, or other description of that topic data.

[0047] Figure 3 shows an example of topic setting according to the first embodiment of the present invention. In this example, when the learner inputs "OpenAI and SoftBank Group collaborate" as topic data, the first generating AI model is made to investigate the input "Collaboration between OpenAI® and SoftBank® Group," and the results are used as topic setting. On the topic setting viewing screen 300 viewed by the learner, only a part of the generated topic setting may be displayed, or a summary of the topic setting may be displayed. Figure 3 is also an example of a selected topic setting from a set of topic settings displayed on the learner terminal 110.

[0048] Figure 4 shows another example of topic setting according to the first embodiment of the present invention. This is an example in which, when audio data obtained by API integration with an online meeting tool is used as topic data, the utterances of each speaker, or their summaries, generated by distinguishing the speakers, are used as topic settings. In this example, when the transcription tab is selected, the utterances of each speaker or their summaries are displayed, and in the summary tab, the meeting summary, participants, and headings (chapters) are displayed, and all or part of the content displayed in the summary tab may be included in the topic settings.

[0049] In this specification, "AI model" refers to a machine learning model that has been trained to predict an output for a given input, and "generative AI model" refers to a large-scale language model (LLM) that has been trained to generate an output not included in the given input. As a generative AI model, an LLM applying a transformer architecture is particularly preferred, but it is expected that the name of the architecture may change as technology advances. Therefore, in this specification, "transformer architecture" includes architectures that use one or more features of a transformer architecture or improvements thereof.

[0050] The topic setting may be the topic data itself if the topic data is text entered by the learner or schedule data obtained from a calendar tool, or it may be a summary of the text or schedule data generated by a generative AI model. Furthermore, if the topic data is specified by a web page URL, the topic setting may be the text itself written on that web page, a summary of all or part of that text generated by the first generative AI model, or text obtained by searching a pre-created embedded database using that text. Also, if the topic data is one or more uploaded files, the topic setting may be the text itself written on all or part of one or more of those files, a summary of all or part of that text generated by the first generative AI model, or text obtained by searching a pre-created embedded database using all or part of that text. The topic setting may also consist of the text itself written on one of the files, and summaries of all or part of the text written on all or part of the other files. The required embedded database can be created as a database on device 100 or a database accessible from device 100 by having the learner or the organization to which the learner belongs upload one or more files and index those files. Thus, topic setting can be determined based on topic data and does not necessarily have to involve requirements for the first generative AI model.

[0051] If the topic data consists of one or more uploaded files, the topic setting may be a summary generated by the first generative AI model for all or part of those files, associated with the label corresponding to each file. For example, on the input screen, a learner can input the label "Our Product" and specify a file containing information about their product, and input the label "Competitor Product" and specify a file containing information about the competitor product, thereby inputting topic data. The first generative AI model can then summarize the text in each file to generate a topic setting in the following format. Some templates for topic settings The following is information about {label_1}. {contents_1} The following is information about {label_2}. {contents_2}

[0052] For example, you could set the variable {label_1} to the value "Our Company's Products" and the variable {contents_2} to the value of a summary of a file containing information about your company's products. Here, we have explained an example where the learner inputs the labels themselves, but you could also have the first generative AI model generate labels from the subject, summary, features, etc., of the text contained in the specified file. Furthermore, you can assign labels to any multiple topic data, not just file uploads.

[0053] When a learner uploads one or more files, they may include files containing one or more sets of source and translation terms related to a learning topic. Such files may be designated as topic data by individual learners, or they may be uploaded by the organization to which the learner belongs and designated as common to that organization in the services provided by device 100.

[0054] Then, the device 100 transmits input screen display information to display an input screen for the learner to input scene settings, which are the settings for the scene in which the learner wants to learn (S207). On the learner terminal 110, an input screen is displayed using the input screen display information, the learner inputs scene settings on the input screen, and the input is transmitted from the learner terminal 110 to the device 100 (S208). As shown in Figures 3 and 4, a screen that displays topic settings and also shows buttons such as "Ask a question," "Explain," and "Discuss" corresponds to the "input screen" described here, because when the learner selects one of these buttons, the corresponding scene is set.

[0055] Figure 5 shows an example of a scene setting screen according to the first embodiment of the present invention. The scene setting screen 500 may include a pull-down menu 510 for scene settings. This represents an overview of the scene. The scene setting screen 500 may include an input field 511 for the learner's role, and if the scene is a conversation scene, it may further include an input field 512 for the role of the conversation partner. In the case of a conversation scene, the scene setting screen 500 may further include an input field 513 for details of the scene, such as the scenario of the conversation. Scenes that can be set include speaking scenes as shown in Figure 5, as well as writing and reading scenes. Speaking scenes include not only conversation scenes with a conversation partner but also scenes without a conversation partner. In the case of a conversation scene, the scenario of the conversation could be included in the scene setting, but in other speaking, writing, and reading scenes, their purposes, etc., may be included in the scene setting as details of the scene.

[0056] In a writing scenario, an example would be creating a proposal document for a client. In this case, the learning method could involve generating example sentences in the learner's native language as learning material, outputting them to the learner's device, and then grading and correcting the translated text entered by the learner in the language they wish to learn. In a reading scenario, an example would be understanding a client's requirements. In this case, the learning method could involve generating the client's requirements as text or audio in the foreign language the learner wishes to learn, outputting them to the learner's device, and then grading and correcting the content that the learner understands and writes in the foreign language or their native language.

[0057] In the example in Figure 5, when a learner selects one of several options from the pull-down menu 510, which contains an overview of the scene setting, the learner's role and other details included in the selected scene setting are automatically filled in the respective input fields, and the learner can modify them as needed. In addition to creating one or more scene settings in advance in this way, it is also possible to generate scene settings using a generative AI model, which will be described later. Furthermore, learners may be able to select from two or three positions on the content of the topic setting from a pull-down menu or other options. For example, if the topic setting is a description of a theme such as politics or economics, learners may be able to select from two options, "agree" and "disagree," or three options, "agree," "neutral," and "disagree," and the selected position may be made at least part of the learner's or conversation partner's role. Specifically, examples include including positions and actions such as "asking questions from a disagreeing position" or "discussing from a supporting position" as roles for the learner or conversation partner. "Agree" and "disagree" are just examples; they may also be liberal and conservative, and they do not have to be completely exclusive. Another possible scenario involves allowing learners to select one or more opinions, such as supporting, neutral, or opposing, regarding the topic description, and then having the learner's role be to ask questions, provide explanations, or engage in discussions about the selected opinion.

[0058] In Figure 5, the topic setting is described in the section labeled "Selected Content," and using Zoom® as an example, the following text is provided as a description of the company's product: "Zoom is a platform that provides a variety of communication tools, including video conferencing, webinars, cloud phone calls, and team collaboration. High-quality audio and video conferencing facilitates effective discussions with screen sharing and breakout room features. The webinar function supports up to 1,000 participants and enables two-way communication through Q&A and polling features. It also offers a robust cloud phone service, chat functions for team collaboration, and file sharing. Furthermore, AI-powered Zoom Docs streamlines document creation and editing, enabling smooth collaboration during meetings. High security features such as end-to-end encryption and waiting room functionality protect privacy and prevent unauthorized participation. These features have made Zoom a popular communication tool that meets a wide range of needs for business, education, and personal use."

[0059] Subsequently, the device 100 uses the topic setting and at least one of the attributes included in the learner data or the learner's role included in the scene setting to request the generating AI model (hereinafter also referred to as the "second generating AI model") to generate teaching materials (hereinafter also referred to as the "second request") (S209), and can receive the generated teaching materials (S210). More specifically, the device 100 uses the topic setting and at least one of the attributes included in the learner data or the learner's role included in the scene setting to create an instruction to generate teaching materials (hereinafter also referred to as the "second instruction"), and makes a second request including the second instruction to the second generating AI model. Then, the device 100 transmits the received teaching materials to the learner terminal 110 (S211). Here, the learner's attributes or roles do not necessarily have to be explicitly given in the second request, as will be described later. Also, if the scene is a conversation scene, the role of a conversation partner may be given in addition to or as a substitute for the learner's role. Learner data may include, in addition to or in place of, the learner's learning status, and the second instruction may include such learning status. Examples of learning status include vocabulary acquisition status.

[0060] If the topic setting is speaker-specific utterances or summaries thereof based on audio data obtained from an online meeting tool, the device 100 may request a second generative AI model to generate materials for the learner to rephrase their utterances in a meeting in another language, redo stumbled parts, or rephrase them in different expressions, from the perspective of any speaker or participant, by selecting scene setting options such as "redo" or "practice". The learner's perspective can be determined by the device 100 based on learner data, for example, or it may be explicitly selected by the learner.

[0061] In the example topic setting shown in Figure 3, selecting the "Ask a Question" button allows the learner to send a scenario to device 100 that includes the learner's role, or a scenario in which the learner asks the AI ​​a question about the topic "OpenAI and SoftBank Group Partner." The same applies to the "Discuss with Participants" button in Figure 4. In this case, after selecting the button, the user may be able to individually select one or more participants with whom they wish to discuss.

[0062] The above explanation described an example where the scene settings are entered after the topic data is specified, but the scene settings may be entered first. Figure 6 shows a screen where the "Sales Negotiation" scene is selected from the list of scenes, and the content of the predetermined scene settings for that scene is confirmed. The user can confirm the content of the scene settings, modify them as needed, and then specify topic data or select topic settings for the topic they wish to learn under that scene setting. If the scene settings are predetermined, an identifier identifying the scene settings may be sent from the learner terminal 110 to the device 100. Also, if a pre-generated topic setting is selected after the scene setting has been determined, an identifier identifying the topic setting may be sent from the learner terminal 110 to the device 100.

[0063] Furthermore, scene settings do not necessarily have to be entered by the learner each time material is generated; they can be entered in advance. For example, if device 100 acquires the URL of a website viewed by a learner as topic data, and a scene setting that explains the content of the website viewed by the learner is already stored in device 100, then the topic setting based on that topic data and the material generated using that scene setting can be automatically delivered to the learner.

[0064] Figure 7 shows an example of instructions for generating teaching materials according to the first embodiment of the present invention. In this instruction, the instructions are for a conversation scene, with the variable {yourrole} representing the learner's role, the variable {airole} representing the role of the AI ​​conversation partner, and the variable {topic} representing the topic setting. The variables {scenesummary} and {scenedetail} represent the scene's overview and details, respectively. Furthermore, in this instruction, the variable {difficulty} can be set to the learner's English level, and the variable {region} to represent the regionality of the language. The language level can be a beginner, intermediate, or advanced level, or a level according to the CEFR, an international evaluation index, or a level corresponding to the learning history in the service provided by the device 100. The example in Figure 4 is an example of generating the first question posed by the conversation partner, but similar instructions can be prepared for subsequent statements in response to the learner's statements.

[0065] In device 100, by executing code to make a request including the instructions shown in Figure 4, the OpenAI API can be called specifying model "4o," causing the second generative AI model provided on platform 120 to generate educational materials. The OpenAI API is an example, and other APIs may be used. The code for making the second request is stored in the memory unit 103, and device 100 can retrieve it and execute code containing instructions obtained by setting values ​​for each variable in the code. Although not mentioned earlier, the necessary code for the first request to the first generative AI model can be similarly prepared.

[0066] In Figure 2, the first and second generating AI models are distinguished, but they may be the same generating AI model. In this specification, whether or not "generating AI models" are the same is determined by whether or not the type of generating AI model specified by the user is the same. For example, in the case of the OpenAI API, if the value of the variable "model" is the same, it is said that they are the same generating AI model. If the first and second AI models are not the same, they may be provided on the same platform 120.

[0067] Figure 8A shows an example of teaching materials generated under the topic and scene settings shown in Figure 5, while Figure 8B shows an example of teaching materials generated under the scene settings shown in Figure 5 but with the topic setting left blank, as a comparative example. In Figure 8B, the initial question is vague, whereas in Figure 8A, the question is specific to the context of the company's products, which is the topic the learner wants to learn about. The example in Figure 8A generates teaching materials with both audio and text, but it is also possible to generate only audio or only text.

[0068] While the learner's role included in the scene setting may be included in the scene overview, scenario, or other scene objectives, it is preferable to explicitly include at least one of the learner's role and the conversation partner's role in the scene setting, as shown in Figure 5, in order to stabilize the quality of the teaching materials output by the generating AI model.

[0069] In the above description, it was assumed that the device 100 receives both learner data and scene settings, but it is also possible to substitute the learner's role included in the scene settings with attributes included in the learner data. Furthermore, if at least one of the learner's role or attributes and the role of the conversation partner is input and can be explicitly set in the variables in the second instruction, the scene outline, purpose, etc., do not necessarily have to be explicitly included in the scene settings.

[0070] As described above, according to one embodiment of the present invention, by making requests to the generating AI model using topic settings based on topic data related to the learning topic, in addition to the learner's attributes or role, it becomes possible to generate practical teaching materials that are suitable for specific scenes that the learner may face.

[0071] (Second embodiment) In the first embodiment, an example was given in which one or more scene settings are created in advance and one of them is selected from a pull-down menu, but scene settings can also be generated using a generation AI model.

[0072] Specifically, device 100 can use topic settings based on topic data and attributes included in learner data to request a generative AI model (hereinafter also referred to as the "third generative AI model") to generate a scene setting (hereinafter also referred to as the "third request"), and receive the generated scene setting. More specifically, device 100 can create an instruction (hereinafter also referred to as the "third instruction") to generate a scene setting using topic settings based on topic data and attributes included in learner data, and then make a third request including the third instruction to the third generative AI model. As an example, the third request can be executed when the learner selects the "AI Recommend" button shown in Figure 5.

[0073] For example, if the topic is a business in which the learner works, the scene is a conversation scene, and the occupation included in the learner's attributes is sales, a scene setting can be generated that is useful for learning conversation scenes for a sales professional. The third instruction may include the format of the scene setting, preferably including at least one of the following: a scene overview, scene details or purpose, the learner's role, and the role of the conversation partner.

[0074] In this way, by generating and making selectable scene settings that correspond to the topic setting and learner attributes, it becomes possible for learners to study in scenes that are more relevant to them than pre-created options.

[0075] The first and third generation AI models may be the same generation AI model. If the first and third generation AI models are not the same, they may be provided on the same platform 120. If the first and third generation AI models are the same generation AI model, these requests may be made by a single API call. Alternatively, the first request to the first generation AI model may be divided into multiple requests and implemented by multiple API calls, and the first request may include one or more processes performed by the device 100 other than the request to the first generation AI model. The same applies to the second and third generation AI models.

[0076] (Third embodiment) The first instruction for generating a topic setting using the topic data described in the first embodiment may also be an instruction to acquire additional data related to the topic data and to generate a topic setting based on the topic data and the additional data. Examples of such additional data include technical terms and the latest information.

[0077] Furthermore, in the embodiments described above, unless the word "only" is used, such as "based only," "depending only," "in the case of only," or "referencing only," it is assumed in this specification that additional information may also be considered. Also, as an example, the statement "if a, then b" does not necessarily mean "always b in the case of a" or "b immediately after a," unless explicitly stated otherwise. In addition, the statement "each a constituting A" does not necessarily mean that A is composed of multiple components, but includes the possibility that the component is singular.

[0078] Furthermore, it should be noted that the embodiments of the present invention described above are included in the disclosure herein, in any way that they are not inconsistent with each other.

[0079] Furthermore, for the sake of clarity, even if there are aspects of operation in some method, program, terminal, device, server, or system (hereinafter referred to as "method, etc.") that differ from the operation described herein, each aspect of the present invention is intended to cover the same operation as any of the operations described herein, and the existence of operation different from the operation described herein does not mean that such method, etc. is outside the scope of each aspect of the present invention.

[0080] Furthermore, while the above explanation refers to multiple generative AI models, in the second embodiment, for example, a request is made to the third generative AI model before making a request to the second generative AI model. In such cases, for the sake of readability, the third generative AI model may be called "AI model l" and the second generative AI model may be called "AI model m". The first generative AI model to which a request is made before the third generative AI model may be called "generative AI model k", and the other generative AI models may be referred to as appropriate. [Explanation of Symbols]

[0081] 100 devices 101 Communications Department 102 Processing Unit 103 Storage section 104 Databases 300 Topic Settings Viewing Screen 400 Topic Settings Viewing Screen 500 Scene Settings Screen 510 Scene Settings Pull-down Menu 511 Input field for learner's role 512 Input field for the role of the conversation partner 513 Input field for scene details 600 Scene setting confirmation screen

Claims

1. A method for supporting foreign language learning, The steps include receiving a specification of topic data related to a learning topic from a learner's terminal or an organization terminal used by the organization to which the learner belongs, The steps include receiving from the learner terminal at least one of learner data representing the learner's attributes and a scene setting that is the setting of the scene the learner wants to learn, or an identifier thereof, A step of making a request (m) to the generating AI model m to generate teaching materials using the topic setting based on the topic data and the learner's role included in the learner data or the scene setting, The steps include: transmitting the generated learning materials to the learner's terminal; Includes.

2. The method according to claim 1, The aforementioned designation is the URL of the web page containing the aforementioned topic data.

3. The method according to claim 1, The aforementioned specification refers to one or more files.

4. The method according to claim 1, The aforementioned designation is the learner's input of the topic data.

5. The method according to claim 1, The aforementioned specification specifies the application to be integrated with.

6. The method according to claim 5, The aforementioned application is an online meeting tool.

7. The method according to claim 6, The aforementioned topic setting includes speaker-specific utterances during the meeting.

8. The method according to claim 1, The aforementioned specification includes the selection of one of several topic settings.

9. A method according to any one of claims 1 to 5, The aforementioned topics are the business in which the learner is engaged or the foreign language lessons that the learner takes.

10. A method according to any one of claims 1 to 9, The process further includes the step of making a request (k) to the generating AI model k to generate the topic settings using the topic data.

11. A method according to any one of claims 1 to 10, The aforementioned scene setting includes the roles of the learners.

12. The method according to claim 11, The learner's role includes any of the two or three options selected regarding the content of the topic setting.

13. A method according to any one of claims 1 to 12, The aforementioned scene is a conversation scene, The aforementioned scene setting includes the role of the conversation partner.

14. A method according to any one of claims 1 to 12, The aforementioned scene is a conversation scene, The aforementioned scene setting includes the scenario of the aforementioned conversation.

15. A method according to any one of claims 1 to 12, The aforementioned scene is a speaking, writing, or reading scene.

16. The method according to claim 15, The aforementioned scene setting includes the purpose of speaking, writing, or reading.

17. A method according to any one of claims 1 to 16, The process further includes the step of making a request (l) to the generating AI model l to generate the scene setting using the aforementioned topic setting.

18. A method according to any one of claims 1 to 16, The specification of the topic data is received after the scene setting or its identifier is received.

19. A program for causing a computer to perform a method for supporting foreign language learning, wherein the method is The steps include receiving a specification of topic data related to the topic the learner wishes to learn from a learner's terminal or an organization terminal used by the organization to which the learner belongs, The steps include receiving from the learner terminal at least one of learner data representing the learner's attributes and a scene setting that is the setting of the scene the learner wants to learn, or an identifier thereof, A step of making a request (m) to the generating AI model m to generate teaching materials using the topic setting based on the topic data and the learner's role included in the learner data or the scene setting, The steps include: transmitting the generated learning materials to the learner's terminal; Includes.

20. A device for supporting foreign language learning, The system receives a specification of topic data related to the topic the learner wishes to study from the learner's terminal or the organization's terminal used by the organization to which the learner belongs. The system receives from the learner terminal at least one of learner data representing the learner's attributes and scene settings or their identifiers, which are the settings for the scene the learner wants to learn. Using the topic setting based on the aforementioned topic data and the attribute included in the learner data or the learner's role included in the scene setting, a request (m) is made to the generating AI model m to generate teaching materials. The system is configured to transmit the generated learning materials to the learner's terminal.

21. A method for supporting foreign language learning, The steps include receiving the selection of the topic setting that the learner wishes to study from the learner's terminal or the organization's terminal used by the organization to which the learner belongs, The steps include receiving from the learner terminal at least one of either a scene setting, which is the setting of the scene the learner wants to learn, or its identifier, The steps include: making a request (m) to the generating AI model m to generate educational materials using the learner roles included in the topic setting and the scene setting; The steps include: transmitting the generated learning materials to the learner's terminal; Includes.

22. The method according to claim 21, The learner's role is to take a stance on the description of the topic setting.

23. The method according to claim 21, The learner's role is to ask questions, explain, or discuss opinions regarding the description of the topic setting.

24. A program for causing a computer to perform a method for supporting foreign language learning, wherein the method is The steps include receiving the selection of the topic setting that the learner wishes to study from the learner's terminal or the organization's terminal used by the organization to which the learner belongs, The steps include receiving from the learner terminal at least one of either a scene setting, which is the setting of the scene the learner wants to learn, or its identifier, The steps include: making a request (m) to the generating AI model m to generate educational materials using the learner roles included in the topic setting and the scene setting; The steps include: transmitting the generated learning materials to the learner's terminal; Includes.

25. A device for supporting foreign language learning, The system receives the selection of the topic setting that the learner wishes to study from the learner's terminal or the organization's terminal used by the organization to which the learner belongs. The learner's terminal receives at least one of either a scene setting, which is the setting of the scene the learner wants to learn, or its identifier. Using the learner roles included in the aforementioned topic settings and scene settings, a request (m) is made to the generative AI model m to generate educational materials. The system is configured to transmit the generated learning materials to the learner's terminal.

Images