Conversion device and conversion method

Abstract

The present disclosure makes it possible to acquire content on the basis of various types of input data without the need to select a model of an input destination. An RAG system 20 according to the present disclosure comprises: a reception unit 21 that receives a data file as input information; a determination unit 22 that determines a generative AI model to which the data file is to be input on the basis of the data type of the data file; a conversion unit 23 that converts the data file into a converted file of a data type corresponding to the generative AI model determined by the determination unit 22; and an input unit 24 that inputs the converted file into the generative AI model.

Description

Conversion device and conversion method

[0001] One aspect of the present disclosure relates to a conversion device and a conversion method.

[0002] In recent years, various types of content have been generated using generative artificial intelligence (AI) models. Various types of input data, such as text files and image files, can be input to the generative AI models along with prompts. Various types of generative AI models, such as ChatGPT (registered trademark) and PaLM2, are available.

[0003] A technique for optimally training AI is described, for example, in Patent Document 1. In this technique, a neural network learning model is trained using one learning algorithm selected from multiple learning algorithms in accordance with the problem to be solved.

[0004] Japanese Patent Application Publication No. 5-298277

[0005] Users of generative AI models desire to be able to acquire content even when various types of input data are used as input. However, there are cases where the types of input data that can be input to a single generative AI model are limited.

[0006] Therefore, an object of the present disclosure is to provide a conversion device and a conversion method that can acquire content based on various types of input data without requiring the user to select a model as the input destination.

[0007] The conversion device of the present disclosure includes a receiving unit that receives input information, a determination unit that determines one of a plurality of generative AI models based on the type of input information, a conversion unit that converts the input information into conversion information according to the generative AI model type corresponding to the generative AI model determined by the determination unit, and an input unit that inputs the conversion information to the generative AI model.

[0008] Alternatively, the conversion method of the present disclosure includes a receiving step for receiving input information, a determination step for determining one of a plurality of generative AI models based on the type of input information, a conversion step for converting the input information into conversion information according to the generative AI model type corresponding to the generative AI model determined in the determination step, and an input step for inputting the conversion information into the generative AI model.

[0009] According to one aspect of the present disclosure, content can be acquired based on various types of input data without the need to select a model to which the input data is to be sent.

[0010] FIG. 1 is a block diagram showing the configuration of a conversion system of the present disclosure. FIG. 2 is a diagram showing an example of the configuration of an input information management table stored in the table storage unit 25. FIG. 3 is a diagram showing an example of the configuration of a conversion management table stored in the table storage unit 25. FIG. 4 is a diagram showing an example of a prompt P1 processed by the RAG system 20. FIG. 5 is a diagram showing an example of response information R1 to the prompt P1. FIG. 6 is a diagram showing an example of response information R1 to the prompt P1. FIG. 7 is a diagram showing an example of another prompt P2 processed by the RAG system 20 and an example of response information R2 to the prompt P2. FIG. 8 is a diagram showing an example of a prompt P3 received from the terminal 10 in response to the response information R2 shown in FIG. 7 and an example of response information R3 to the prompt P3. FIG. 9 is a flowchart showing the procedure of data relay processing by the RAG system 20. FIG. 10 is a flowchart showing the procedure of data relay processing by the RAG system 20. FIG. 11 is a diagram showing an example of the hardware configuration of the RAG system 20 according to an embodiment of the present disclosure.

[0011] The present disclosure will be described with reference to the accompanying drawings. Whenever possible, the same parts are designated by the same reference numerals and redundant description will be omitted.

[0012] Fig. 1 is a diagram showing the device configuration of a conversion system according to this embodiment. The conversion system shown in Fig. 1 includes a terminal 10, a Retrieval-Augmented Generation (RAG) system 20, and a server device 30, which are configured to be able to communicate with each other via a network including a wireless communication network and a fixed communication network. The RAG system 20 constitutes a conversion device that converts input information received from the terminal 10 into conversion information.

[0013] The terminal 10 is a device used by a user who wishes to obtain various types of content (text, audio, images, videos, etc.) based on input information using an interactive AI model. The terminal 10 may be, for example, a personal computer, a smartphone, a tablet terminal, a feature phone, a server device, a game console, etc. Note that while only two terminals 10 are illustrated in FIG. 1 , the conversion system may include any number of terminals 10 greater than or equal to two.

[0014] The server device 30 is a device that enables the provision of content using one of multiple generative AI models. A generative AI model is a model that, in response to a prompt containing input information, generates content according to any one or a combination of the instructions, context, question, and output format indicated by the prompt, and returns the content as response information. The prompt may also include input information, in which case the generative AI model generates response information targeted at the input information. The generative AI model may be, for example, an interactive AI model that includes a large-scale language model (LLM) and a user interface (UI) for interacting with the user, enabling text or voice chat with the user. Examples of such generative AI models include ChatGPT, GPT (registered trademark)-3.5, GPT-4V, PaLM2, etc. In this embodiment, the server device 30 is capable of providing content provision functions using multiple types of interactive AI models 31, 32, 33, etc. These interactive AI models 31, 32, 33, ... may be stored in the server device 30, or may be stored in another device connected to the server device 30 via a network, and configured to enable information exchange with a user via the server device 30. Note that while only one server device 30 is shown in FIG. 1, the conversion system may include multiple server devices 30. Also, although the above describes an example of a large-scale language model, other AI models may also be used.

[0015] RAG system 20 is configured to include, as functional components, a reception unit 21, a determination unit 22, a conversion unit 23, an input unit 24, and a table storage unit 25. RAG system 20 relays prompts including input information from terminal 10 to server device 30, and relays response information from server device 30 to the prompt to terminal 10. RAG system 20 also has a function of converting input information from terminal 10 into conversion information. The functions of each functional unit of RAG system 20 will be described in detail below.

[0016] The reception unit 21 receives a prompt containing input information from a user using the terminal 10. A prompt is information indicating an instruction or question input by a user in an interactive system such as a dialogue with an AI model or a command line interface (CLI). The prompt received by the reception unit 21 expresses, in text, for example, an instruction to be executed by the interactive AI model, a task to be executed by the interactive AI model, a background / context to be considered by the interactive AI model (e.g., a role or condition), a question to be answered by the interactive AI model, and an output format of response information from the interactive AI model. The prompt may also include input information to be used as the target of an instruction / task to be executed by the interactive AI model. Examples of such input information include data files with file names including a predetermined extension, such as text data, image data, application-related data, audio data, video data, and still image data. Application-related data is data such as document data, table data, and graph data that can be processed by a default application program. When the receiving unit 21 receives a prompt from the terminal 10 , it passes the received prompt to the determining unit 22 and the converting unit 23 .

[0017] The determination unit 22 determines one of the plurality of interactive AI models 31, 32, 33, ... as the interactive AI model to which the prompt including the input information received by the reception unit 21 is to be input. If the prompt received by the reception unit 21 is text only and does not include a data file as input information, the determination unit 22 determines one of the plurality of predetermined interactive AI models 31, 32, 33, ... selected by the user using the terminal 10, or one randomly selected interactive AI model (e.g., interactive AI model 31), as the input destination of the prompt. Specifically, the determination unit 22 identifies the type of input information, i.e., the extension of the data file indicating the data format of the input information, and determines the interactive AI model to which the prompt is to be input based on the extension by referring to the input information management table stored in the table storage unit 25.

[0018] 2 shows an example of the configuration of the input information management table stored in the table storage unit 25. In this way, the input information management table stores multiple pairs of "generation AI model type" (generation AI model specification information) data that specifies the type of interactive AI model and "input file extension" (type specification information) data that specifies the extension of data files that can be processed by the interactive AI model, in association with each other. The "input file extension" data in the input information management table includes a list of multiple data file extensions. For example, data "ChatGPT" as the "generation AI model type" is associated with data "txt, csv, json, yaml, py, js, c, cpp, html, css, xml" as the "input file extension", and data "PaLM2" as the "generation AI model type" is associated with data "txt, csv, json, yaml, py, js, c, cpp, html, css, xml, mp3, wav, ogg, jpg, png, gif, svg, mp4, avi, mov, pdf" as the "input file extension". In the above example, it is shown that a data file in text data format with the extension "txt" can be processed by both the interactive AI model "ChatGPT" and the interactive AI model "PaLM2", but a data file in audio data format with the extension "mp3" or "wav", or a data file in video data format with the extension "mp4" can be processed by the interactive AI model "PaLM2", but cannot be processed by the interactive AI model "ChatGPT".

[0019] The determination unit 22 determines the interactive AI model to which the prompt is to be input by searching the input information management table having the above configuration based on the extension of the identified data file. That is, the determination unit 22 extracts data on the "generated AI model type" associated with the "input file extension" corresponding to the extension of the data file in the input information management table, and determines the interactive AI model corresponding to the extracted "generated AI model type" as the interactive AI model to which the prompt is to be input. For example, if the extension of the identified data file is "png," the determination unit 22 extracts "GPT-4V" and "PaLM2" as the "generated AI model type" associated with the "input file extension" that matches the extension, and determines one interactive AI model to which the prompt is to be input from among the two interactive AI models corresponding to the extracted "generated AI model type." In this case, if data of multiple "generated AI model types" is extracted, the determination unit 22 may determine one interactive AI model from among the multiple interactive AI models in accordance with the selection of the user of the terminal 10, or may determine one interactive AI model randomly selected from among the multiple interactive AI models.

[0020] Furthermore, if there is no data for "generated AI model type" associated with the "input file extension" corresponding to the extension of the data file in the input information management table, the determination unit 22 repeats the extraction of the data for the above-mentioned "generated AI model type" using the extension of the converted file, which is the type of conversion information determined by the conversion unit 23. Then, the determination unit 22 repeats extraction using other conversion file extensions determined by the conversion unit 23 until data for the "generated AI model type" can be extracted, and determines the interactive AI model corresponding to the successfully extracted "generated AI model type" as the interactive AI model to which the prompt is to be input.

[0021] If, in the determination process by the determination unit 22, there is no data for "generated AI model type" associated with the "input file extension" corresponding to the extension of the data file, that is, if there is no interactive AI model capable of processing the data file among the multiple interactive AI models 31, 32, 33, ..., the conversion unit 23 determines the data type of the conversion file, which is the conversion information to which the data file is converted, based on the extension of the data file identified by the determination unit 22. Specifically, the conversion unit 23 identifies the extension of the data file that indicates the data format of the input information, and determines the data format of the conversion file based on that extension by referring to the conversion management table stored in the table storage unit 25.

[0022] 3 shows an example of the configuration of the conversion management table stored in the table storage unit 25. Thus, the conversion management table stores multiple sets of associated data: "input file extension" data specifying the extension (data type) of the input data file; "extended file extension" data specifying the extension of the converted file after conversion from the data file; "conversion method information" data specifying the conversion method used in the RAG system 20 when converting from the data file to the converted file; and "priority" data indicating the priority of the conversion method. The "conversion file extension" data in the input information management table may include multiple conversion file extensions. Furthermore, multiple sets of "conversion file extension," "conversion method information," and "priority" data are associated with one "input file extension" data in the input information management table. For example, data with an "input file extension" of "mp4" is associated with a combination of "mp3, jpg" data as a "conversion file extension," "conversion method information" of "read an mp4 file using tool A, set the mp3 bit rate to 128 kbps...," and "priority" of "1," and a combination of "mp3, jpg" data as a "conversion file extension," "conversion method information" of "read an mp4 file using tool A, set the mp3 bit rate to 192 kbps...," and "priority" of "2." In the above example, for one "input file extension" data, the data combination of "conversion file extension" and "conversion method information" associated with a higher priority is preferentially extracted during conversion processing by the conversion unit 23. Furthermore, the conversion processing by the conversion unit 23 is performed by executing the procedure indicated by the extracted "conversion method information."

[0023] The conversion unit 23 determines the data type of the converted file to be converted and the conversion method by searching the conversion management table having the above configuration based on the extension of the data file. That is, the conversion unit 23 extracts a combination of "conversion file extension" data and "conversion method information" data associated with the "input file extension" corresponding to the extension of the data file in the conversion management table, determines the data format corresponding to the extracted "conversion file extension" as the data format of the converted file to be converted, and determines the conversion method for the converted file based on the extracted "conversion method information." At this time, the conversion unit 23 extracts a combination with a relatively high "priority" from among multiple combinations of "conversion file extension" data and "conversion method information" data associated with the "input file extension" corresponding to the extension of one data file in the conversion management table. For example, if the extension of the identified data file is "xlsx," the conversion unit 23 extracts "png" and "Start software A, read the input file..." as a combination of the "conversion file extension" and "conversion method information" associated with the "input file extension" that matches the extension, determines the data format of the image data corresponding to the extracted "conversion file extension" as the data format of the converted file, and determines the method of conversion processing to the converted file by referring to the extracted "conversion method information." At this time, if the "conversion file extension" data includes multiple extensions, the conversion unit 23 may determine one data format from among the data formats indicated by the multiple extensions in accordance with a selection made by the user of the terminal 10, or may determine one data format randomly selected from among the data formats indicated by the multiple extensions.

[0024] Furthermore, based on the data format of the conversion file determined by the above-mentioned function and the method of conversion processing to the conversion file, if the determination unit 22 determines that data for the "generated AI model type" associated with the "input file extension" corresponding to the extension of the conversion file exists, and if a conversion file extension corresponding to the "generated AI model type" corresponding to the generated AI model determined by the determination unit 22 exists, i.e., if an interactive AI model capable of processing the data type of the conversion file exists, the conversion unit 23 converts the input data file into a conversion file. For example, if the data shown in Figures 2 and 3 is referenced and the data format of the conversion file is determined to be the "png" extension, the conversion unit 23 converts the data file into a conversion file of an image data format with the extension "png" because the determination unit 22 determines "PaLM2" as the interactive AI model capable of processing the data format of the conversion file. Then, the conversion unit 23 passes the conversion file converted from the data file to the input unit 24.

[0025] If the determination process by the determination unit 22 finds that data for the "generation AI model type" associated with the "input file extension" corresponding to the extension of the data file exists, the input unit 24 inputs a prompt containing the data file as input information to an interactive AI model belonging to the type determined by the determination unit 22 from among the interactive AI models 31, 32, 33, .... The input of the prompt to the interactive AI model is performed by transmitting the prompt to the server device 30. Furthermore, if the determination process by the determination unit 22 finds that data for the "generation AI model type" associated with the "input file extension" corresponding to the extension of the data file does not exist, the input unit 24 replaces the data file included in the prompt with the converted file delivered from the conversion unit 23, and inputs the prompt to an interactive AI model belonging to the type determined by the determination unit 22 based on the data type of the converted file. In addition, the input unit 24 receives, from the server device 30, response information returned from the interactive AI model in response to the input of the prompt, and transmits the received response information to the terminal 10 that input the prompt.

[0026] Next, examples of prompts input to the conversational AI model from the RAG system 20 and response information returned to the terminal 10 by the conversational AI model in response to the prompts will be shown.

[0027] FIG. 4 shows an example of a prompt P1 processed by the RAG system 20, and FIGS. 5 and 6 show examples of response information R1 to the prompt P1. The prompt P1 includes text indicating an instruction received from the terminal 10: "The next image to be input is an image of an Excel spreadsheet. Please read the contents from the image and summarize them concisely." The prompt P1 also includes a conversion file F1 in the data format of image data, which has been converted from a data file in the data format of table data by the RAG system 20. The response information R1 to the prompt P1 includes text data processed for the prompt P1 by an interactive AI model determined by the RAG system 20. The response information R1 shown in the example includes "image content" data indicating the meaning of the numerical values ​​of the graph shown in the conversion file F1, "details" data providing a detailed analysis of the trend of fluctuations in the graph, and "discussion" data providing a summary of the data, including a discussion.

[0028] Figure 7 shows an example of another prompt P2 processed by the RAG system 20 and an example of reply information R2 to the prompt P2, and Figure 8 shows an example of a prompt P3 received from the terminal 10 in response to the reply information R2 shown in Figure 7 and an example of reply information R3 to the prompt P3.

[0029] 7, prompt P2 includes text indicating the command received from terminal 10, such as "The information to be entered next is in CSV format. Please read the contents and answer the questions that follow," as well as a CSV-format conversion file F2 of text data converted by RAG system 20 from a data file in tabular data format. Response information R2 to prompt P2 includes text data and image data processed for prompt P2 by an interactive AI model determined by RAG system 20. The illustrated response information R2 includes data analyzing the meaning of the arrangement of the CSV-format data shown in conversion file F2, image data displaying the CSV-format data in a tabular format, and data explaining the meaning of each column shown in the image data.

[0030] 8, the prompt P3 received from the terminal 10 in response to the response information R2 includes text data expressing a question following the prompt P2 for the interactive AI model that generated the response information R2. The response information R3 to this prompt P3 includes text data indicating an answer processed for the prompt P3 by the interactive AI model that generated the response information R2.

[0031] The procedure of the content providing process by the RAG system 20 configured as above, that is, the flow of the conversion method according to this embodiment, will be described below. Figures 9 and 10 are flowcharts showing the procedure of the content providing process by the RAG system 20.

[0032] In the content provision process, first, the reception unit 21 of the RAG system 20 receives a prompt including input information from the terminal 10 (step S1). The reception unit 21 then determines whether the received prompt includes a data file as input information (step S2). If the determination result shows that a data file is not included (step S2; No), the input unit 24 of the RAG system 20 outputs a prompt to one interactive AI model selected by the user from among the multiple interactive AI models 31, 32, 33, ..., or to a randomly selected interactive AI model (step S3). In response, the input unit 24 relays response information returned from the interactive AI model to the terminal 10 (step S4), and the content provision process is terminated.

[0033] On the other hand, if the prompt includes a data file (step S2; Yes), the determination unit 22 of the RAG system 20 acquires the extension of the data file (step S5). Next, the determination unit 22 determines whether or not "generated AI model type" data associated with the data file extension exists in the input information management table (step S6). If the determination result indicates that "generated AI model type" data associated with the extension does not exist (step S6; No), the conversion unit 23 of the RAG system 20 acquires "conversion file extension" data and "conversion method information" data corresponding to the data file extension in the conversion management table, and determines the data format of the converted file based on the "conversion file extension" data (step S7). The determination of step S6 is then repeated using the acquired conversion file extension.

[0034] If it is determined in step S6 that data for "generated AI model type" associated with the extension exists (step S6; Yes), the conversion unit 23 determines whether the data format of the converted file has been determined in the processing of the immediately preceding step S7 (step S8). If the determination result indicates that the data format has been determined (step S8; Yes), the conversion unit 23 uses the procedure indicated in the data for "conversion method information" acquired in the immediately preceding step S7 to perform a conversion process from the data file to a converted file (step S9), and the processing proceeds to step S10. On the other hand, if the data format has not been determined (step S8; No), the processing proceeds to step S10 without converting the data file.

[0035] Thereafter, the input unit 24 of the RAG system 20 outputs a prompt including a data file or a prompt including a conversion file to the interactive AI model corresponding to the "generation AI model type" acquired by the determination unit 22 (step S10). Next, the input unit 24 relays the response information returned from the interactive AI model to the terminal 10 (step S11). Furthermore, if the reception unit 21 receives an additional prompt in response to the response information from the terminal 10, the additional prompt is output to the same interactive AI model, and the response information thereto is relayed to the terminal 10. This completes the content provision process.

[0036] Next, the effects of the conversion device of the present disclosure will be described. According to the RAG system 20 of the present disclosure, an interactive AI model is determined from among multiple types of interactive AI models 31, 32, 33, etc. based on the data type of the data file accepted by the user, the data file is converted into a conversion file of a data type corresponding to the determined interactive AI model, and the converted conversion file is input to the generation AI model. As a result, when a user prepares a data file of any type, the data file can be automatically converted into a conversion file suitable for the interactive AI model and input to the generation AI model without the user having to select the destination interactive AI model. Therefore, content can be efficiently acquired based on various types of input data without the hassle of selecting a generation AI model.

[0037] In the RAG system 20 of the present disclosure, the determination unit 22 selects and determines an interactive AI model corresponding to the data format from among a plurality of predetermined interactive AI models 31, 32, 33, ... depending on the data format of the data file or the data format of the converted file. A generating AI model may be able to generate content in response to input of a data file in one data format, but may not be able to handle input of a data file in another data format. Even in such cases, the data file or the converted file can be input to a generating AI model corresponding to the data format of the data file or the converted file, thereby enabling content to be acquired more efficiently.

[0038] In the RAG system 20 of the present disclosure, the determination unit 22 identifies the data format of a data file based on the extension of the data file. In this case, the data format of the input information can be quickly identified, and the content can be quickly acquired.

[0039] The RAG system 20 of the present disclosure further includes a table storage unit 25 that stores an input information management table. The determination unit 22 determines a generating AI model by extracting, from the input information management table, a generating AI model type associated with an extension corresponding to the data type of the data file or the data type of the conversion file. This makes it easy to identify a generating AI model corresponding to the data type of the data file or the data type of the conversion file. As a result, the input information or conversion information can be easily input into an appropriate generating AI model, making it easy to obtain appropriate content.

[0040] In the RAG system 20 of the present disclosure, the table storage unit 25 further stores a conversion management table, and the conversion unit 23 determines the data type of the converted file by extracting, from the conversion management table, type information of the converted file associated with the extension corresponding to the data type of the data file. This makes it easy to determine the type of conversion information to be converted from the input information and to easily identify the generation AI model corresponding to the type of conversion information. As a result, the conversion information can be easily input into the appropriate generation AI model, making it easy to obtain appropriate content.

[0041] Furthermore, in the RAG system 20 of the present disclosure, the conversion unit 23 determines the data type of the converted file by extracting, from the conversion management table, information with a relatively high priority from among multiple types of type information associated with extensions corresponding to the data type of the data file. This makes it possible to uniquely determine the type of conversion information to be converted from the input information, and to easily identify the generation AI model corresponding to the type of conversion information.

[0042] In the RAG system 20 of the present disclosure, the conversion unit 23 extracts conversion method information associated with the extension corresponding to the data type of the data file from the conversion management table, and converts the input information into conversion information by referring to the conversion method information. In this case, the conversion process can be efficiently performed when converting the input information into conversion information.

[0043] The conversion device and conversion method of the present disclosure have the following configuration.

[0044] [1] A conversion device comprising: a receiving unit that receives input information; a determination unit that determines one of a plurality of generative AI models based on the type of the input information; a conversion unit that converts the input information into conversion information according to the generative AI model type corresponding to the generative AI model determined by the determination unit; and an input unit that inputs the conversion information to the generative AI model.

[0045] [2] The conversion device according to [1] above, wherein the determination unit selects and determines a generative AI model corresponding to the data format from among a plurality of predetermined generative AI models, depending on the data format of the input information or the data format of the conversion information.

[0046] [3] The conversion device according to [2], wherein the determination unit identifies the data format of the input information based on a file extension of the input information.

[0047] [4] The conversion device according to any of [1] to [3] above, further comprising a table storage unit that stores an input information management table that associates generative AI model identification information that identifies a generative AI model with type identification information that identifies a type of input information that can be processed by the generative AI model, wherein the determination unit determines the generative AI model by extracting from the input information management table the generative AI model identification information that is associated with the type of the input information accepted by the acceptance unit or the type identification information that corresponds to the type of the conversion information.

[0048] [5] The conversion device described in [4] above, wherein the table storage unit further stores a conversion management table that associates input type information that identifies a type of input information with conversion type information that identifies a type of conversion information converted from the input information, and the conversion unit determines the type of the conversion information by extracting from the conversion management table the conversion type information that is associated with the input type information that corresponds to the type of the input information accepted by the acceptance unit.

[0049] [6] The conversion device described in [5] above, wherein the conversion management table is a table that associates the input type information, a plurality of pieces of conversion type information, and priorities of the plurality of pieces of conversion type information, and the conversion unit determines the type of the conversion information by extracting from the conversion management table information with a relatively high priority from the plurality of pieces of conversion type information associated with the input type information corresponding to the type of the input information accepted by the acceptance unit.

[0050] [7] The conversion device described in [5] or [6] above, wherein the conversion management table is a table that associates the input type information, the conversion type information, and conversion method information that indicates a method of converting the input information into the conversion information, and the conversion unit extracts from the conversion management table the conversion method information that is associated with the input type information that corresponds to the type of the input information accepted by the acceptance unit, and converts the input information into the conversion information by referring to the conversion method information.

[0051] [8] A conversion method comprising: a receiving step of receiving input information; a determination step of determining one of a plurality of generative AI models based on the type of the input information; a conversion step of converting the input information into conversion information of a type corresponding to the generative AI model determined in the determination step; and an input step of inputting the conversion information into the generative AI model.

[0052] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are connected directly or indirectly (e.g., via wire, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0053] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0054] For example, the RAG system 20 constituting the conversion system according to an embodiment of the present disclosure may function as a computer that performs processing of the control method of the present disclosure. FIG. 11 is a diagram illustrating an example of the hardware configuration of the RAG system 20 according to an embodiment of the present disclosure. The RAG system 20 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage device 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like. The RAG system 20 may be configured as a computer device including at least one processor, such as a CPU or GPU, or may be configured as a computer device including multiple processors or may include multiple computer devices. The terminal 10 and the server device 30 may also have a similar hardware configuration.

[0055] In the following description, the term "apparatus" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the RAG system 20 may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.

[0056] Each function in the RAG system 20 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0057] The processor 1001 controls the entire computer by running, for example, an operating system. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, the above-mentioned reception unit 21, determination unit 22, conversion unit 23, input unit 24, etc. may be realized by the processor 1001.

[0058] The processor 1001 also reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these programs. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, the reception unit 21, the determination unit 22, the conversion unit 23, and the input unit 24 may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be used for other functional blocks. While the above-described various processes have been described as being executed by a single processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0059] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing a control method according to an embodiment of the present disclosure.

[0060] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0061] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as a network device, network controller, network card, communication module, etc. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned reception unit 21, input unit 24, etc. may be realized by the communication device 1004.

[0062] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. Note that the input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0063] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0064] Furthermore, the RAG system 20 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0065] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0066] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0067] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0068] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0069] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0070] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0071] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0072] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0073] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0074] Note that terms described in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of a channel and a symbol may be a signal (signaling). Furthermore, a signal may be a message. Furthermore, a component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.

[0075] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information. For example, a radio resource may be indicated by an index.

[0076] The names used for the above-described parameters are not intended to be limiting in any way. Furthermore, the mathematical expressions using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.

[0077] In this disclosure, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," "terminal," and the like may be used interchangeably.

[0078] A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0079] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0080] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0081] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0082] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0083] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0084] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0085] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0086] 10...Terminal, 20...RAG system, 30...Server device, 31, 32, 33...Interactive AI model, 21...Reception unit, 22...Decision unit, 23...Conversion unit, 24...Input unit, 25...Table storage unit, F1, F2...Conversion file, P1, P2, P3...Prompt, R1, R2, R3...Response information.

Claims

1. A conversion device comprising: a reception unit that receives input information; a determination unit that determines one of a plurality of generative AI models based on the type of the input information; a conversion unit that converts the input information into conversion information according to the generative AI model type corresponding to the generative AI model determined by the determination unit; and an input unit that inputs the conversion information to the generative AI model.

2. The conversion device according to claim 1, wherein the determination unit selects and determines a generative AI model corresponding to the data format of the input information or the data format of the conversion information from among a plurality of predetermined generative AI models.

3. The conversion device according to claim 2, wherein the determination unit identifies the data format of the input information based on a file extension of the input information.

4. The conversion device of claim 1, further comprising a table storage unit that stores an input information management table that associates generative AI model identification information that identifies a generative AI model with type identification information that identifies the type of input information that can be processed by the generative AI model, and the determination unit determines the generative AI model by extracting from the input information management table the generative AI model identification information that is associated with the type of input information accepted by the acceptance unit or the type identification information that corresponds to the type of the conversion information.

5. The conversion device according to claim 4, wherein the table storage unit further stores a conversion management table that associates input type information that specifies the type of input information with conversion type information that specifies the type of conversion information converted from the input information, and the conversion unit determines the type of the conversion information by extracting from the conversion management table the conversion type information that is associated with the input type information that corresponds to the type of the input information accepted by the acceptance unit.

6. The conversion device according to claim 5, wherein the conversion management table is a table that associates the input type information, a plurality of pieces of conversion type information, and priorities of the plurality of pieces of conversion type information, and the conversion unit determines the type of the conversion information by extracting from the conversion management table information with a relatively high priority from the plurality of pieces of conversion type information associated with the input type information corresponding to the type of the input information accepted by the acceptance unit.

7. The conversion device described in claim 5, wherein the conversion management table is a table that associates the input type information, the conversion type information, and conversion method information that indicates a method of converting the input information into the conversion information, and the conversion unit extracts from the conversion management table the conversion method information that is associated with the input type information that corresponds to the type of the input information accepted by the acceptance unit, and converts the input information into the conversion information by referring to the conversion method information.

8. A conversion method comprising: a receiving step for receiving input information; a determination step for determining one of a plurality of generative AI models based on the type of the input information; a conversion step for converting the input information into conversion information of a type corresponding to the generative AI model determined in the determination step; and an input step for inputting the conversion information into the generative AI model.