Information processing system

Abstract

This technology provides an improved efficiency in the process of translating data from a first language into a second language. [Solution] The information processing system displays a list of multiple data in a first language contained in a data file, translates one data selected by the user from the list of data into a second language using one or more generative models, and stores one translation proposal resulting from the translation of the selected data into the second language using one or more generative models, associating it with the selected data.

Description

Technical Field

[0001] This disclosure relates to an information processing system.

Background Art

[0002] An application for translating text from a first language into a second language is known.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] According to this disclosure, a technique for improving the efficiency of the operation of translating text from a first language into a second language can be provided.

Means for Solving the Problems

[0005] An information processing system according to an aspect of this disclosure includes one or more memories and one or more processors. The one or more processors display a list of a plurality of data in a first language included in a data file, translate one piece of data selected by a user from the plurality of data displayed in the list into a second language by one or more generation models, and save one translation result obtained by translating the selected one piece of data into the second language by one or more generation models in association with the selected one piece of data.

Brief Description of the Drawings

[0006] [Figure 1] It is a diagram showing an example of the flow of translation processing in an information processing system. [Figure 2] It is a diagram showing an example of a page displaying text to be translated. [Figure 3]This figure shows an example of a page where a translation suggestion is displayed. [Figure 4] This figure shows an example of a page displaying the translation result selected by the user. [Figure 5] This figure shows an example of the editing process flow in an information processing system. [Figure 6] This figure shows an example of a page that accepts editing guidelines for translation proposals from users. [Figure 7] This figure shows an example of a page where editing suggestions are displayed. [Figure 8] This figure shows an example of the refinement process flow in an information processing system. [Figure 9] This figure shows an example of a page displaying a refined version of the proposal. [Figure 10] This figure shows an example of a page that accepts definitions of labels to be given to translation proposals from the user. [Figure 11] This figure shows an example of a page displaying a translation suggestion using information associated with a label. [Figure 12] This figure shows an example of a page where translated suggestions using images are displayed. [Figure 13] This diagram shows an example of the processes performed by an information processing system and the data used in each process. [Figure 14] This figure shows an example of a project creation screen. [Figure 15] This figure shows an example of how to display a text record. [Figure 16] This figure shows an example of displaying multiple translation candidates and their back translations. [Figure 17] This figure shows examples of how terms detected from the translated text are displayed. [Figure 18] This figure shows an example of a chat-style user interface display when reviewing translation results. [Figure 19] This figure shows an example of a chat-style user interface display when reviewing the original text. [Figure 20]A block diagram showing an example of a computer hardware configuration. [Modes for carrying out the invention]

[0007] Hereinafter, embodiments of this disclosure will be described with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0008] An information processing device of the present disclosure is, as an example without limitation, a computer 7, as shown in Figure 20, which includes one or more memories (main memory 72, auxiliary memory 73) and one or more processors (processor 71). The memories are storage units for storing programs, and the processors are processing units that execute programs and perform the processing related to the present disclosure. Hereinafter, "information processing device" or "information processing system" may be read as processor or processing unit, as appropriate.

[0009] The information processing system disclosed herein comprises one or more information processing devices. The information processing system disclosed herein may be a system consisting of a single device or a system including multiple devices. As an example, the information processing system has a web browser that displays a screen, and in the web browser, it translates data (e.g., text) in a first language contained in a data file (e.g., a text file) selected by the user into data (e.g., text) in a second language, and displays the translation result in the web browser. The first and second languages ​​are, for example, Japanese and English, but are not limited to specific languages. The selection of data files in the web browser, the input of text and instructions, and the display of various processing results such as translation results are realized using external devices 9B, such as a keyboard, pointing device, display, touchscreen, microphone, etc., which are connected via a device interface 75.

[0010] In the present disclosure, "translation" may conceptually include generating data (e.g., text) in a second language based on data (e.g., text) in a first language, generating another data (e.g., text) in the second language based on the generated data in the second language, and generating data (e.g., text) in a third language based on at least one of the generated data in the second language or the other data. The data to be "translated" is not limited to text and may be voice, still image, or moving image.

[0011] In the present disclosure, the generation model may be a machine learning model that accepts a prompt and generates data. The generation model is not limited to a machine learning model of a specific architecture, such as, for example, a machine learning model, a transformer, a state space model, a combination of a transformer and a state space model, etc. Further, the generation model may be a language model that handles language via text, or a language model that handles language via text and / or other modalities (e.g., images, voice, moving images, etc.). The generation model may be, for example, a pre-trained language model such as a large language model (LLM) or a vision language model (VLM).

[0012] In the present disclosure, the generation model is described as a common generation model, but may be separate generation models. That is, each of the following embodiments may utilize one or more generation models, and for convenience, one or more generation models are referred to as generation models. Further, the location of the generation model may be inside the information processing system or an external information processing device.

[0013] The information processing system may input a prompt described in natural language to the generation model and obtain a response to the input of the prompt from the generation model. At this time, the generation model may perform processing according to the input prompt and output the processing result as a response described in natural language.

[0014] (First Embodiment) A first embodiment of the information processing system of this disclosure will be described with reference to Figures 1 to 4. In this embodiment, the information processing system includes one or more information processing devices as client devices and one or more information processing devices as server devices. The client devices and server devices communicate with each other via a communication network such as the Internet using their respective network interfaces 74. The client devices may function as terminals operated by a user. The server devices may function as translation devices that translate data received from the terminals. Hereafter, an example of translating text in a first language into a second language will be described, but the invention is not limited thereto.

[0015] The client device (one or more processors) accesses the server device to perform text translation. The server device (one or more processors) generates a page in response to the access and sends the generated page to the client device. The steps described below are examples, and their execution order may be changed as appropriate.

[0016] Figure 1 shows an example of the translation process flow in the information processing system of this embodiment.

[0017] In step S101, the client device displays the received page in a web browser and accepts the data file to be translated from the user through that page. That is, the client device accepts the data file specified by the user. For example, the client device may accept the data file when the user enters the file path of the data file on the page, or when the user drags and drops the icon of the data file onto the page.

[0018] The data file to be translated contains multiple texts to be translated. These texts are written in the primary language from which they will be translated.

[0019] In step S102, the client device sends the received data file to be translated to the server device. The server device receives the data file from the client device.

[0020] In step S103, the server device extracts and stores each of the multiple texts to be translated from the received data file. That is, the server device imports the received data file. For example, the server device may extract the first column of each record in a comma-separated values ​​(CSV) data file as the multiple texts to be translated. Alternatively, the server device may extract the columns specified by the user in each record of a CSV data file as the multiple texts to be translated. Another example is that the server device may extract each line of a text data file as the multiple data to be translated.

[0021] In step S104, the server device sends the text extraction results to the client device. The client device receives the text extraction results from the server device. The extraction results may include each text separated into records.

[0022] The series of steps for text extraction from step S102 to step S104 may be performed on the client device instead of sending the data file to the server device. For example, text extraction may be performed on the client device by splitting the given data file into columns or rows.

[0023] In step S105, the client device displays the received extraction results, for example, as a list of texts, on a page displayed in the web browser. The client device then accepts requests from the user to modify, add, or delete text to be translated via the page displayed in the web browser. For example, the client device may accept text entered by the user via keyboard or text entered by the user via voice input.

[0024] The client device may accept user requests to modify, add, or delete text on the page displaying the extracted text results. The client device may also update the text stored in the server device by sending the modified, added, or deleted text to the server device. When a user selects text to be translated from a list of texts, the client device sends a translation instruction for that text to the server device.

[0025] Figure 2 shows an example of a page displayed in the web browser of the client device in step S105. Page 20 has a translation target display area 203 that displays the text extraction results. The translation target display area 203 displays the text to be translated, with each text item separated into a line (cell). For example, the translation target display area 203 displays a portion of the Japanese text to be translated, "This is a function that allows you to travel to various worlds. You can earn travel photos and star coins as rewards," in cell 203-1.

[0026] Page 20 has a label area 202 that displays labels assigned to the text to be translated. The label area 202 displays the labels in separate cells for each piece of text to be translated. For example, the label area 202 displays the label assigned to the text to be translated displayed in cell 203-1 in cell 202-1. Labels may be metadata for the text to be translated, may be used when filtering data, or may be used as context when translating data.

[0027] Metadata is information associated with a label (e.g., text, audio, images, videos, etc.). Metadata may also include, for example, text describing the text to be translated, or text describing the person or character who speaks the text as dialogue.

[0028] Page 20 has a translation result display area 204 that displays the translation results of the text. The translation result display area 204 displays the translated text in separate rows (cells) for each text to be translated. The cell displaying the text to be translated and the cell displaying the translated text may be displayed adjacent to each other within a single record, for example, so that the relationship between the source text and the target text is easy for the user to understand. For example, the translation result display area 204 displays cell 204-1, which displays the English text that is the result of translating the Japanese text to be translated into English, next to cell 203-1 which displays the Japanese text to be translated.

[0029] Page 20 has a key area 201 that displays a key to identify each piece of text to be translated. The key area 201 may display a button 201-1 (the "add new item" button in Figure 2) for adding text to be translated. The user may add text to be translated by pressing button 201-1 and entering a key, a label, and the Japanese text to be translated.

[0030] Label area 202 may accept user operations such as selecting a cell in label area 202 to modify, add, or delete labels. Translatable text display area 203 may accept user operations such as selecting a cell in translation area 203 to modify the text to be translated displayed in that cell. Translation result display area 204 may accept user operations such as selecting a cell in translation result display area 204 to modify the translated text displayed in that cell.

[0031] Let's return to Figure 1 for explanation. In step S106, the client device sends a text translation instruction to the server device. The server device receives the text translation instruction from the client device.

[0032] For example, when a user selects cell 204-1 shown in Figure 2, the client device sends a translation instruction to the server device to translate the Japanese text displayed in cell 203-1, which corresponds to cell 204-1. In other words, the client device sends a translation instruction to the server device to translate one text selected by the user from among multiple texts in the first language displayed in a list.

[0033] In step S107, the server device translates the text in the first language to be translated according to the translation instructions. The server device may obtain one or more translations in the second language as a result of translating the text.

[0034] For example, a server device may generate one or more prompts, each containing at least one of the following: text in a first language to be translated, and instructions to translate that text from the first language to the second language. The server device may input these prompts into a generative model and obtain the translation results from the generative model's response. If multiple translation options (for example, N, where N is an integer greater than or equal to 2) are to be obtained, the server device may, for example, input these prompts into the generative model and obtain the translation results N times, or it may include instructions in the prompt to propose N translation options.

[0035] In step S108, the server device sends the text translation result to the client device. The client device receives the text translation result from the server device. Hereafter, the server device will be described as sending multiple translation options to the client device as translation results, as an example, but will not be limited to this.

[0036] The series of steps from step S106 to step S108, which translate the selected text to be translated from the first language to the second language, may be performed on the client device instead of sending the text and translation instructions to the server device.

[0037] In step S109, the client device displays the received translation result on the page shown in the web browser.

[0038] Figure 3 shows an example of a page displayed in the web browser of the client device in step S109. Page 21 has a translation suggestion display area 205 that displays the translation results received from the server device. The client device prompts the user to select one translation suggestion from one or more translation suggestions 205-1, 205-2, 205-3 displayed in the translation suggestion display area 205. For example, the client device displays multiple translation suggestions 205-1, 205-2, 205-3 in the translation suggestion display area 205 and accepts the user's selection of one translation suggestion. The user may select, for example, translation suggestion 205-1.

[0039] Page 21 may include a re-translate button 206. When the user selects the re-translate button 206, the series of steps S106 to S109 shown in Figure 1 are re-executed.

[0040] Let's return to Figure 1 for explanation. In step S110, the client device receives a selection of one translation result from the user. As an example, the client device determines which of several translation options the user has selected.

[0041] In step S111, the client device displays the selection results from step S110 on a page. On the page displaying the selection results, the client device may accept operations from the user to obtain suggestions for edited translations (details in the second embodiment) or refined translations (details in the third embodiment). The client device may also accept operations from the user to discard translation results or save translation results on the page displaying the selection results.

[0042] Figure 4 shows an example of a page displayed in the web browser of the client device in step S111. Page 22 displays the translation selected by the user in step S110 in cell 204-1 of the translation result display area 204 and in the selection result display area 207. As an example, the selection result display area 207 displays the English text "This feature allows Crypko to travel to various worlds. You can earn travel photos and star coins as rewards," and cell 204-1 displays a portion of that English text.

[0043] The client device may accept an operation to modify the translated text displayed in the selection result display area 207. If cell 204-1 accepts an operation to modify the translated text, the selection result display area 207 may display the same text as cell 204-1 according to the modifications made. If the selection result display area 207 accepts an operation to modify the translated text, cell 204-1 may display the same text as the selection result display area 207 according to the modifications made.

[0044] Cell 204-1, which displays the translated text, may display a cancel button 208 to allow the user to discard the translation result, and a save button 209 to allow the user to save the translation result. If the user selects the cancel button 208, the client device discards the text displayed in cell 204-1, the selection result display area 207, and the translation suggestion display area 205. If the user selects the save button 209, step S112 shown in Figure 1 is executed. If the user selects the re-translate button 206, the series of steps from step S106 to step S109 shown in Figure 1 are re-executed.

[0045] Page 22 may display a re-translate button 210, an edit button 211, and a refine button 212. The re-translate button 210 is a button for performing a new translation and obtaining a translation from the server device. The edit button 211 is a button for obtaining an edited version of the translation displayed in the selection result display area 207 from the server device. The refine button 212 is a button for obtaining a refined version of the translation displayed in the selection result display area 207 from the server device.

[0046] Let's return to Figure 1 for explanation. In step S112, the client device sends a translation result save instruction to the server device. The server device receives the translation result save instruction from the client device. The translation result save instruction is information that instructs the server to save the text in the second language, which is the target text, displayed in cell 204-1 on page 22, as the translation result of the text in the first language, which is the source text, displayed in cell 203-1. As an example, the save instruction may include information identifying cell 203-1, information identifying cell 204-1, the text in the first language which is the source text, and the text in the second language which is the target text.

[0047] In step S113, the server device saves the text in the second language displayed in cell 204-1 as the translation result of the text in the first language displayed in cell 203-1, in accordance with the save instruction received in step S112.

[0048] The above example illustrates the translation of text in the first language displayed in a single cell, 203-1, into the second language. However, translation of all text in the first language displayed in the translation target display area 203 into the second language may also be performed.

[0049] Furthermore, in the above scenario, multiple translation options in the second language were presented to the user for a single text in the first language, and the translation selected by the user from among these options was set (saved) as the translation result. However, when importing the data file, the first translation option for each text in the first language may be automatically set (saved) as the translation result in the second language as a provisional translation result. In this case, the user may later review the translation result and perform re-translation or manual correction operations as appropriate.

[0050] The client device may receive instructions from the user to export the translation results for each text in the data file received in step S101. Upon receiving the instruction to export the translation results, the client device instructs the server device to export the translation results. The server device writes each source text and each translated result text displayed in the translation target display area 203 and the translation result display area 204 to a file in a format specified by the user, such as CSV or text format, and sends that file to the client device. The export of translation results is the same in each embodiment described below.

[0051] (Modified version of the first embodiment) In step S101, if the client device receives a data file with the same filename as a previously input data file, steps S106 to S113 may be executed on the difference between the previously input data file and the currently input data file. In this case, in step S103, when the server device extracts text from the currently input data file, it extracts only the difference from the previously input data file. In step S105, the client device displays the result of the difference extraction in step S103 on the page. Then, in steps S106 to S113, the information processing system performs processing on that difference.

[0052] In step S103, when extracting differences, you may extract text that is not included in the previously entered data file but is included in the currently entered data file. In step S105, when displaying differences, you may additionally display the records at the positions corresponding to the extracted text. Also, in step S103, when extracting differences, you may extract text that is included in the previously entered data file but is not included in the currently entered data file. In step S105, when displaying differences, you may delete the records at the positions corresponding to the extracted text before displaying.

[0053] (Second embodiment) A second embodiment of the information processing system of this disclosure will be described with reference to Figures 5 to 7. In this embodiment, an example of processing when a user performs an operation to obtain a proposed edit of the translation in step S111 of the first embodiment will be described. Each step described below is an example, and the execution order may be changed as appropriate.

[0054] Figure 5 shows an example of the editing process flow in the information processing system of this embodiment. When the user selects the edit button 211 on page 22 shown in Figure 4, step S201 shown in Figure 5 is executed.

[0055] In step S201, the client device displays a page for receiving editing directions from the user for editing the translation proposal. The client device receives editing directions from the user via the page for receiving editing directions from the user.

[0056] Figure 6 shows an example of a page displayed in the web browser of the client device in step S201. Page 23 displays cell 203-1 showing the text to be translated, cell 204-1 showing the translation suggestion, selection result display area 207, and instruction area 213 for indicating the direction of editing.

[0057] The instruction area 213 may display buttons for performing pre-defined instructions. For example, the instruction area 213 may display a "Shorter" button 213-1, a "Formal" button 213-2, a "Casual" button 213-3, and a "Polite" button 213-4. The "Shorter" button 213-1 is a button for requesting an edited version of the proposed translation that is shorter. The "Formal" button 213-2 is a button for requesting an edited version of the proposed translation that is more formally expressed. The "Casual" button 213-3 is a button for requesting an edited version of the proposed translation that is more informal. The "Polite" button 213-4 is a button for requesting an edited version of the proposed translation that is more politely expressed. By selecting one of these pre-defined buttons, the user can easily instruct the server device on the direction of editing the proposed translation.

[0058] The instruction area 213 may display a description area 213-5 that allows the user to directly describe the editing direction in text. The client device can accept text input to the description area 213-5 and then provide detailed instructions to the server device regarding the editing direction of the translation draft. In this embodiment, for example, an example in which the user selects the "Shorter" button 213-1 is described, but the embodiment is not limited to this.

[0059] Let's return to Figure 5 for explanation. In step S202, the client device sends an instruction to the server device to edit the translation draft (editing instruction). The server device receives the editing instruction for the translation draft from the client device.

[0060] For example, the editing instructions may include the source text in the first language, displayed in cell 203-1; the (provisional) proposed translation text in the second language, displayed in cell 204-1; and the direction of editing (or information indicating it) requested by the user in step S201.

[0061] In step S203, the server device generates text that will be the edited version of the translation in response to the editing instructions received in step S202. The generation of the edited version can be rephrased as the task of translating the text in the first language, which is the source text, into the second language based on the provisional translated version text in the second language and the direction of editing. The server device may obtain one or more edited versions as the edited version.

[0062] As an example, a server device may generate one or more prompts, including a text in a first language which is the source text, a text in a second language which is a provisional translation of that text, a text explaining the direction of editing, and an instruction to "re-translate the text in the first language into the second language, taking into account the direction of editing, by referring to the translation from the text in the first language to the text in the second language," input these prompts into a generation model, and obtain the translation result (i.e., the editing proposal) from the generation model's response. Since the generation model generates output while updating its internal state in response to the input prompts, even if it is not explicitly instructed in the prompts to refer to specific information, if the specific information that should be referenced is included in the prompts, it will generate output taking that specific information into account. Furthermore, since the generation model generates output while updating its internal state in response to the input prompts, if the prompts contain information that implicitly indicates the situation, it will generate output taking that situation into account. This characteristic of the generation model is the same in other embodiments as well.

[0063] As another example, the server device may generate one or more prompts, including a source text in a first language, a provisional translation of that text in a second language, text explaining the direction of editing, and an instruction to "re-translate the text in the first language into the second language," input these prompts into a generative model, and obtain the translation result (i.e., the edited version) from the generative model's response.

[0064] As another example, the server device may generate one or more prompts, each containing text in a first language, information indicating that the text in the first language is the source text, text in a second language, information indicating that the text in the second language is a proposed translation, text describing the direction of editing, and an instruction to "re-translate the text in the first language into the second language," input these prompts into a generative model, and obtain the translation result (i.e., the proposed edit) from the generative model's response.

[0065] When obtaining multiple (for example, N) editing suggestions, the server device may, for instance, input the generated prompt into the generation model to obtain one editing suggestion, and repeat this process N times. Alternatively, the generated prompt may include instructions to suggest N editing suggestions.

[0066] In step S204, the server device sends the editing proposal to the client device. The client device receives the editing proposal from the server device. Hereafter, the server device will be described as sending multiple editing proposals to the client device as an example, but will not be limited to this.

[0067] The series of steps from step S202 to step S204 for obtaining suggested edits may be performed on the client device instead of sending edit instructions to the server device.

[0068] In step S205, the client device displays the received proposal on the page shown in the web browser.

[0069] Figure 7 shows an example of a page displayed in the web browser of the client device in step S205. Page 24 displays the editing direction 214-1 instructed in step S201 and the suggestions 214-2 and 214-3 received from the server device in step S204 in the suggestion display area 214, and prompts the user to select one of the one or more suggestions 214-2 and 214-3 displayed in the suggestion display area 214. Page 24 displays, as an example, the suggestions when the "Shorter" button 213-1 is selected by the user. For example, the client device displays the editing direction 214-1 and multiple suggestions 214-2 and 214-3 in the suggestion display area 214 and accepts the user's selection of one suggestion.

[0070] Page 24 may include a re-translate button 206. When the user selects the re-translate button 206, the series of steps from step S202 to step S205 are re-executed without changing the direction of editing.

[0071] Let's return to Figure 5 for explanation. In step S206, the client device accepts one of the proposed edits as the new translation result from the user. For example, the client device determines which of the multiple edits the user has selected.

[0072] In step S207, the client device displays the selection result from step S206 on the page. As an example, the client device displays the editing suggestion selected by the user in step S206 in cell 204-1 and the selection result display area 207.

[0073] The subsequent processing is the same as the processing described with reference to Figure 4 in the first embodiment. That is, on page 24, the client device may accept an operation to perform a translation again and obtain a translation proposal (see the first embodiment) by selecting the re-translate button 210. Also, on page 24, the client device may accept an operation to obtain a refined translation proposal (detailed in the third embodiment) by selecting the refine button 212. Also, on page 24, the client device may accept an operation from the user to discard the translation result or to save the translation result, as in the first embodiment. That is, steps S208 and S209 are the same as steps S112 and S113 shown in Figure 1.

[0074] (Third embodiment) A third embodiment of the information processing system of this disclosure will be described with reference to Figures 8 and 9. In this embodiment, an example of processing when the user performs an operation to obtain a refined version of the translation in step S111 of the first embodiment will be described. Each step described below is an example, and the execution order may be changed as appropriate.

[0075] Figure 8 shows an example of the refinement process flow in the information processing system of this embodiment. When the refinement button 212 is selected by the user on page 22 shown in Figure 4, step S301 shown in Figure 8 is executed.

[0076] In step S301, the client device sends an instruction to the server device to refine the translation draft (refinement instruction). The server device receives the translation draft refinement instruction from the client device.

[0077] For example, the refinement instructions may include the source text in the first language, displayed in cell 203-1, and the (provisional) proposed translation in the second language, displayed in cell 204-1.

[0078] In step S302, the server device generates suggestions for improvement to refine the translation in response to the refinement instruction received in step S301. The server device may generate one or more suggestions for improvement. The suggestions for improvement may be expressed in text. As an example, the server device may generate one or more prompts including a text in a first language which is the source text, a text in a second language which is a provisional translation of that text, and an instruction to "suggest suggestions for improvement to refine the text in the second language by referring to the translation from the text in the first language to the text in the second language," input these prompts into the generation model, and obtain suggestions for improvement from the generation model's response.

[0079] As another example, the server device may generate one or more prompts, including a source text in a first language, a provisional translation of that text in a second language, and instructions to "suggest improvements to refine the translation," input these prompts into a generative model, and obtain improvement suggestions from the generative model's response.

[0080] As another example, the server device may generate one or more prompts, each including text in a first language, information indicating that the text in the first language is the source text, text in a second language, information indicating that the text in the second language is a proposed translation, and "instructions to suggest improvements to refine the proposed translation," input these prompts into a generative model, and obtain improvement suggestions from the generative model's response.

[0081] The server device may explicitly state in the "Instructions to suggest improvements" in this prompt that it is possible to suggest multiple (for example, N) or fewer improvements.

[0082] In step S303, the server device may generate a refined text (refined version) of the translation based on one or more improvement suggestions generated in step S302. The server device may generate one or more refined versions.

[0083] As an example, a server device may generate one or more prompts, each containing a source text in a first language, a provisional translation of that text in a second language, one or more suggestions for improvement, and an instruction to "propose a refined version of the translation by referring to the translation from the first language text to the second language text and the suggestions for improvement," input these prompts into a generative model, and obtain a refined version from the generative model's response.

[0084] As another example, the server device may generate one or more prompts, each containing a source text in a first language, a provisional translation of that text in a second language, one or more suggestions for improvement, and an instruction to "propose a refined translation," input these prompts into a generative model, and obtain a refined translation from the generative model's response.

[0085] As another example, the server device may generate one or more prompts, each containing text in a first language, information indicating that the text in the first language is the source text, text in a second language, information indicating that the text in the second language is a proposed translation, one or more suggestions for improvement, and an instruction to "propose a refined translation," input these prompts into a generative model, and obtain a refined translation from the generative model's response.

[0086] In step S304, the server device sends the improvement suggestions generated in step S302 and the refined proposals generated in step S303 to the client device as a proposal to refine the translation. The client device receives the proposal to refine the translation from the server device.

[0087] The series of steps from step S301 to step S304 for obtaining suggestions to refine the translation draft may be performed on the client device instead of sending refinement instructions to the server device.

[0088] In step S305, the client device displays the received proposal on the page shown in the web browser.

[0089] Figure 9 shows an example of a page displayed in the web browser of the client device in step S305. Page 25 displays the suggestions received from the server device in the refined proposal display area 215 and allows the user to select one refined proposal from the one or more refined proposals displayed in the refined proposal display area 215. For example, the client device displays one or more refined proposals 215-1 and one or more improvement suggestions 215-2 in the refined proposal display area 215 and accepts the user's selection of one refined proposal.

[0090] Page 25 may include a re-translate button 206. When the user selects the re-translate button 206, the series of steps from step S301 to step S305 are re-executed.

[0091] Let's return to Figure 8 for explanation. In step S306, the client device receives one of the proposed refinements as a new translation result from the user. For example, the client device determines which of the one or more refinements the user has selected.

[0092] In step S307, the client device displays the selection result from step S306 on the page. For example, the client device displays the refined proposal selected by the user in step S306 in cell 204-1 and the selection result display area 207.

[0093] The subsequent processing is the same as the processing described with reference to Figure 4 in the first embodiment. That is, on page 25, the client device may accept an operation to obtain a translation proposal by performing a translation again (see the first embodiment) by selecting the re-translate button 210. Also, on page 25, the client device may accept an operation to obtain a proposal for an edited translation (see the second embodiment) by selecting the edit button 211. Also, on page 25, the client device may accept an operation from the user to discard the translation result or to save the translation result, as in the first embodiment. That is, steps S308 and S309 are the same as steps S112 and S113 shown in Figure 1.

[0094] (Fourth embodiment) A fourth embodiment of the information processing system of this disclosure will be described with reference to Figures 10 to 11. Examples of translation processing, editing processing, and refinement processing have been described in each of the embodiments described above. In this embodiment, an example of performing at least one of translation processing, editing processing, or refinement processing will be described, using the label displayed in the label area 202 shown in Figure 2, or the information associated with that label (e.g., text, audio, image, video, etc.) as context. That is, in at least one of the steps S107, S203, S302, and S303 described above, the server device generates one or more prompts that further include "an instruction to refer to the information associated with the label as context," and inputs the prompts into the generation model.

[0095] When translating text from a first language (the source language) into a second language by referring to the information associated with the label as context, the process from steps S101 to S113 described in the first embodiment can be the same as in step S107, with only a part of the process in step S107 being different.

[0096] As an example, in step S107, the server device may generate one or more prompts, each containing at least one of the following: a text in the first language to be translated; information associated with that text as a label; and an instruction to translate the text from the first language to the second language in a way that fits the context, referencing the information associated with the label as context. The server device may then input these prompts into the generative model and obtain the translation result from the generative model's response.

[0097] As another example, in step S107, the server device may generate one or more prompts, each containing at least one of the following: text in the first language to be translated, information associated with that text as a label, and an instruction to translate the text from the first language to the second language; input these prompts into the generative model; and obtain the translation result from the generative model's response.

[0098] As another example, in step S107, the server device may generate one or more prompts including at least one of the following: text in a first language, information indicating that the text in the first language is to be translated, information associated with the text as a label, information indicating that the information is related to the text in the first language, and an instruction to translate the text from the first language to the second language. The server device may then input these prompts into a generative model and obtain the translation result from the generative model's response.

[0099] When editing the text in the second language, which is the (provisional) translation result, by referring to the information associated with the label as context, the processing from steps S201 to S209 described in the second embodiment can be the same, with only a part of the processing in step S203 being different.

[0100] As an example, in step S203, the server device may generate one or more prompts including the source text in the first language, information associated with that text as a label, a provisional translation of that text in the second language, text explaining the direction of editing, and an instruction to re-translate the text in the first language, taking into account the direction of editing, by referring to the information associated with the label as context and making it fit that context, and by referring to the translation from the text in the first language to the text in the second language. The server device may input these prompts into the generative model and obtain the translation result (i.e., the proposed edit) from the generative model's response.

[0101] As another example, in step S203, the server device may generate one or more prompts including a text in the first language which is the source text, information associated with that text as a label, a text in the second language which is a provisional translation of that text, text explaining the direction of editing, and "an instruction to re-translate the text in the first language into the second language", input these prompts into the generative model, and obtain the translation result (i.e., the editing proposal) from the generative model's response.

[0102] As another example, in step S203, the server device may generate one or more prompts including text in a first language, information indicating that the text in the first language is the source text, information associated with the text as a label, information indicating that the information is related to the text in the first language, text in a second language, information indicating that the text in the second language is a translation proposal, text describing the direction of editing, and an instruction to re-translate the text in the first language into the second language, input these prompts into the generative model, and obtain the translation result (i.e., the editing proposal) from the generative model's response.

[0103] When refining the (provisional) translation result in the second language text by referring to the information associated with the label as context, the processes from steps S301 to S309 described in the third embodiment may be the same, with only a part of the processes in step S302 and / or step S303 being different.

[0104] As an example, in step S302, the server device may generate one or more prompts including the source text in the first language, information associated with that text as a label, a provisional translation of that text in the second language, and "an instruction to suggest improvements to refine the text in the second language, referring to the information associated with the label as context and to make it fit that context, and referring to the translation from the text in the first language to the text in the second language," input these prompts into the generative model, and obtain improvement suggestions from the generative model's response.

[0105] As another example, in step S302, the server device may generate one or more prompts including a text in the first language which is the source text, information associated with that text as a label, a text in the second language which is a provisional translation of that text, and an instruction to "suggest improvements to refine the text in the second language", input these prompts into the generation model, and obtain improvement suggestions from the generation model's response.

[0106] As another example, in step S302, the server device may generate one or more prompts including text in a first language, information indicating that the text in the first language is the source text, information associated with the text as a label, information indicating that the information is related to the text in the first language, text in a second language, information indicating that the text in the second language is a proposed translation, and an instruction to "suggest improvements to refine the proposed translation," input these prompts into the generation model, and obtain improvement suggestions from the generation model's response.

[0107] As an example, in step S303, the server device may generate one or more prompts including a text in the first language which is the source of translation, information associated with that text as a label, a text in the second language which is a provisional translation proposal of that text, one or more suggestions for improvement, and an instruction to propose a refined version of this translation, referring to the information associated with the label as context and to the translation from the text in the first language to the text in the second language and the suggestions for improvement, and input these prompts into the generative model and obtain a refined version from the generative model's response.

[0108] As another example, in step S303, the server device may generate one or more prompts including the source text in the first language, information associated with that text as a label, a provisional translation of that text in the second language, one or more suggestions for improvement, and an instruction to "propose a refined version of this translation," input these prompts into the generation model, and obtain a refined version from the generation model's response.

[0109] As another example, in step S303, the server device may generate one or more prompts including text in a first language, information indicating that the text in the first language is the source of translation, information associated with the text as a label, information indicating that the information is related to the text in the first language, text in a second language, information indicating that the text in the second language is a proposed translation, one or more improvement suggestions, and an instruction to "propose a refined version of this translation," input these prompts into the generation model, and obtain a refined version from the generation model's response.

[0110] Figure 10 shows an example of a page displayed in the web browser of a client device, similar to the page in Figure 2. As an example, page 26 shown in Figure 10 has two additional texts to be translated ("I don't dislike it." and "Well, tomorrow...I'm free.") compared to page 20 shown in Figure 2. Also, on page 26, cell 202-2 is selected by the user to assign a label to the text to be translated displayed in cell 203-2. Here, the user has selected the text in cell 203-2 and assigned the label "CharaA" to that text. Thus, cell 203-2 displays the text in the first language to be translated, and cell 202-2 displays the label assigned to that text.

[0111] The label definition area 216 is an area for defining the label (metadata) to be assigned to the selected text. The client device receives the label to be assigned to the text and the definition of that label from the user via the label definition area 216. For example, the label definition area 216 displays the label name area 216-1 and the label content area 216-2. For example, the label name area 216-1 accepts the input of the label name to be defined from the user and displays that label name. The label content area 216-2 accepts the input of the content of the label being defined from the user and displays that content. The information that the user inputs into the label content area 216-2 may be information associated with the label, or it may be metadata of the text displayed in cell 203-2.

[0112] Figure 10 shows the label name "CharaA" and the content of this label "A curt character's line." When this label "CharaA" is given to text, the text is translated, edited, or refined to fit the context of "A curt character's line." Label "CharaA" is just one example, and the label definition area 216 can also accept definitions of other labels from the user.

[0113] Figure 11 shows an example of a page displayed in the web browser of a client device when the text in the first language, which is the source of the translation, is translated into the second language, referring to the information associated with the labels as context. Page 27 displays multiple translation suggestions 205-1, 205-2 in the second language for the text in the first language, which is the source of the translation, in the translation suggestion display area 205, and displays the translation suggestion 205-2 selected by the user as the translation result in cell 204-2 and the selection result display area 207. In other words, the client device allows the user to define the labels to be given to the text in the first language displayed in cell 203-2, which is the source of the translation.

[0114] The server then uses the generative model to consider the information (metadata) associated with the label as context, translating the text in the first language to text in the second language that fits that context. For example, the Japanese text "Well, I'm free tomorrow...but whatever." is translated as "a line spoken by a curt character" into the English text "Well, I'm free tomorrow...but whatever."

[0115] (Fifth embodiment) A fifth embodiment of the information processing system of this disclosure will be described with reference to Figure 12. In the fourth embodiment, when translating text in a first language to a second language, the generation model is made to consider the information (metadata) associated with the label given to the text as context. In this embodiment, an example will be described in which, in addition to labels, media (images, videos, etc.) is considered as context and the generation model is made to translate text in a first language to a second language. That is, similar to the fourth embodiment, in at least one of the steps S107, S203, S302, and S303 described above, the server device generates one or more prompts that further include an "instruction to refer to media as context" and inputs the prompt to the generation model.

[0116] One or more prompts that further include instructions referencing media as context may be generated in several ways.

[0117] When translating text from a first language (the source language) into a second language by referring to media as context, the process from steps S101 to S113 described in the first embodiment can be the same as in step S107, with only a part of the process in step S107 being different.

[0118] As an example, in step S107, the server device may generate one or more prompts, each including at least one of the following: a text in a first language to be translated, media referenced for the translation of that text, and an instruction to translate the text from the first language to the second language in a way that fits the context, referencing the media as a context; input these prompts into the generative model; and obtain the translation result from the generative model's response.

[0119] Here, the instruction to translate text from a first language to a second language, referencing the media as a context and fitting it to that context, may be an instruction that directly or indirectly references the media as a context. For example, if the media is an image, the server device may generate one or more prompts containing the instruction to translate text in a first language to a second language, referencing the image or fitting it to the context indicated by the image. Alternatively, the server device may generate one or more prompts containing the instruction to list some features that can be seen in the image, and input this prompt into the generative model to have the generative model extract the features of the image. Furthermore, the server device may generate one or more prompts containing the extracted features and the instruction to translate text in a first language to a second language, referencing the extracted features or fitting it to the context indicated by the features.

[0120] As another example, the server device may, without inputting the source text of the first language into the generation model, cause the generation model to generate text in the second language corresponding to the text of the first language based on the media associated with the text of the first language. For example, if the media is an image, the server device may generate one or more prompts including the image and "instructions for expressing this image in the most appropriate second language text for use in the second region," and input these prompts into the generation model to obtain the second language text as a proposed translation of the first language text from the generation model. In this case, the second region may be a region that uses at least one of the media or the proposed second language translation of the first language text. In this way, the information processing system may perform translation in localization.

[0121] Alternatively, for example, a server device may cause a generative model to generate text in the first language in a similar manner. That is, the server device may generate one or more prompts, each containing an image and "an instruction to express this image in the most appropriate text in the first language for use in the first region," and input these prompts into the generative model to obtain text in the first language from the generative model. The first region may be the region where the media is used. "An instruction to express this image in the most appropriate text in the first language for use in the first region" may include information indicating what part of the image is being represented (e.g., hairstyle, scene, etc.). This could be, for example, "an instruction to express the hairstyle (or scene) of this image in the most appropriate text in the first language for use in the first region."

[0122] In this embodiment as well, prompts as described in each of the embodiments above may be used.

[0123] When editing the text in the second language, which is a (provisional) translation result, by referring to the media as context, the processing in the fourth embodiment may be carried out by substituting the information associated with the label with the media.

[0124] When refining the (provisional) translation result in the second language text by referring to the media as context, the processing in the fourth embodiment may be carried out by substituting the information associated with the label with the media.

[0125] Figure 12 shows an example of a page displayed in the web browser of a client device when text in a first language (source) is translated into a second language, referencing media as context. Page 28, for example, displays the text in the first language (source) in cell 203-3 and the media referenced when translating that text into the second language in cell 202-3. Page 28 displays multiple translation suggestions 205-1, 205-2, and 205-3 in the second language for the text in the first language (source), and displays the translation suggestion 205-1 selected by the user as the translation result in cell 204-3 and the selection result display area 207. In other words, the client device allows the user to select the media to correspond to the text in the first language displayed in cell 203-3 (source).

[0126] The server then allows the generative model to consider the media as context, translating the text in the first source language into text in the second language that fits that context. For example, when the generative model translates the Japanese text "half twin" into English, it references an image of a character with a hairstyle called "half twin." The generative model probabilistically considers the relationship between translating the Japanese text "half twin" into English and the image (or its features), and translates the Japanese text "half twin" as the hairstyle of the character in the image into English. In this way, the generative model translates the Japanese text "half twin" into the English text "two side up."

[0127] (Sixth embodiment) A sixth embodiment of the information processing system of this disclosure will be described with reference to Figures 13 to 19. As an example, the information processing system according to this embodiment includes a web browser that displays a screen, and displays the results of analyzing multiple texts acquired via the web browser on the display device (web browser). For example, the information processing system translates text in a source language into a target language and displays the translated text in the target language on the display device. In translation, the information processing system may detect terms from the text and translate them into the target language, and then translate multiple texts in a batch using those terms and their translations. Alternatively, in translation, the information processing system may use pairs of the translated first text in the source language and the first translated text in the target language to translate a second text in the source language into a second translated text in the target language. The information processing system may review the translated text and display an explanation of its content. The information processing system may accept user questions and / or consultations regarding text translation and provide answers, including proposed translations of the text into the target language, through a chat-style user interface. Note that the source language is an example of a first language. The target language is an example of a second language.

[0128] The information processing system acquires source data containing multiple texts to be translated (multiple source texts), automatically extracts translation reference information from the source data, and uses this reference information to automatically translate the multiple texts from the language in which the texts are written (source language) to the target language, and / or, interactively translates them into the target language through interaction with the user via a chat-style user interface, and outputs the translation result. The chat-style user interface is a user interface for interaction between the user and the device. The chat-style user interface may display the user's statements and the device's statements in the order they are made.

[0129] Furthermore, the information processing system may use reference information to review and / or revise the translation results and output the final translation results. The source data may include information about the translation project (project information) in addition to the multiple texts to be translated. The project information may include at least one of the following: the project name, the source language, one or more target languages, and a description of the project. The information processing system may automatically refer to the project information when translating multiple texts in the same project. The information processing system does not have to automatically refer to the project information of other projects when translating a translation in one project. The reference information may include at least one of the following: terms and their translations automatically extracted from the source data, automatically generated glossaries, manually generated glossaries, information about labels given to the texts to be translated, and the context of the translation. If parts of multiple texts have already been translated, the reference information may include the translation results for those parts (existing translations). The reference information may be extracted from the source data within a project and may differ from project to project.

[0130] The information processing system may perform translation of text or terms by machine translation. Machine translation may include rule-based translation. The information processing system may use a generative model when extracting and / or translating reference information. The generative model according to this embodiment may be the same as the generative models according to each of the embodiments described above.

[0131] The information processing system may implement a chat-style user interface using a so-called AI (Artificial Intelligence) agent. The information processing system may provide chat areas as a chat-style user interface on the right side of page 20, such as areas 205, 213, 214, 215, and 216. For example, the information processing system may acquire user questions and / or consultations via the chat-style user interface, generate one or more prompts containing the acquired statements, and input these prompts into a generative model to generate statements from the device that respond to the user's statements. The generative model for generating the device's statements may be the same generative model as the generative model for translation, or it may be a different generative model.

[0132] An information processing system may have an AI agent automatically perform translations in response to user statements obtained through a chat-style user interface. For example, the information processing system may have the AI ​​agent generate a processing procedure for performing translations in response to user statements, and then instruct the AI ​​agent to execute the generated processing procedure in order, thereby translating one or more texts to be translated. One or more processing steps in the processing procedure generated by the AI ​​agent may perform processing using a generative model. For example, if the acquired user statement instructs the translation of a specific expression in the source language to a specified expression in the target language, the AI ​​agent may generate a processing procedure that includes the steps of searching for one or more texts containing the specific expression from multiple texts to be translated in the source language, and translating the one or more texts found using the specified expression in the target language, and then execute each step in order according to the generated processing procedure.

[0133] Figure 13 shows an example of the processes performed by an information processing system and the data used in each process. The information processing system may display a project creation screen and obtain project information from the user through the project creation screen. The information processing system may obtain project information by accepting a project configuration file that contains the project information.

[0134] Figure 14 shows an example of a project creation screen. The project creation screen 400 may, for example, accept input of project information including project name 401, source language 402, target language 403, and project description 404. In the example shown in Figure 14, the project name is "OmegaCrafter", the source language is "Japanese", the target languages ​​are "English (American)" and "Chinese (Simplified)", and the project description is "'Omega Crafter' is an open-world survival crafting game set in a game world where development is stalled due to a mysterious sabotage program."

[0135] Let's return to Figure 13 for explanation. In step S401, the information processing system acquires multiple texts (source data) that are to be translated in the project. In this disclosure, "text" may refer to a single sentence or a collection of sentences. The method for acquiring multiple texts is not limited to a specific method; for example, the information processing system may acquire multiple texts by having a user manually input multiple texts into the information processing system, or by having a user select a file containing multiple texts and upload it to the information processing system.

[0136] For example, if the uploaded file is a comma-separated (CSV) file, the information processing system may retrieve the text between commas as a single text. If the uploaded file is a tab-separated (TSV) file, the system may retrieve the text between tabs as a single text. If the uploaded file is a text file, the system may retrieve the text between line breaks as a single text.

[0137] The information processing system registers and displays the acquired source data in each record. The information processing system may display multiple texts to be translated, with one text per record.

[0138] Figure 15 shows an example of a text record display. A record may include an id cell 411 displaying the record's serial number (id), a key cell 412 displaying the record's key, a label cell 413 displaying the label of the source text, a context cell 414 displaying the context of the source text, a source text cell 415 displaying the source text, and a translation result cell 416 (416-1, 416-2) displaying the text translated into the target language.

[0139] Note that each cell in the record corresponds to each cell on page 20 shown in Figure 2. For example, key cell 412 corresponds to key area 201. label cell 413 corresponds to label area 202. Source cell 415 corresponds to translation target display area 203. Translation result cells 416-1 and 416-2 correspond to translation result display area 204.

[0140] Label cell 413 may contain label information that is commonly used for translations to each target language. In addition to the commonly used label information, label cell 413 may also contain label information specific to each target language. Label information may be registered by the user for each record.

[0141] Label information may be a sentence that specifies how the source text registered in the record should be translated, and may differ for each target language. Label information may also be a sentence that specifies, for example, the length or politeness of the translation. If the text to be translated is a character's line of dialogue, the label information may be a sentence that describes the situation in which the line is spoken, or a sentence that describes the character who speaks the line. The information processing system can obtain a translation that reflects the label information from the generative model by including the label information as reference information in the prompts input to the generative model.

[0142] Translation result cells 416 may exist for each target language. The information processing system may translate the source text in the corresponding source cell 415 into the target language corresponding to the translation result cell for each translation result cell selected by the user. The information processing system may also automatically translate each source text in all source cells 415 into their respective target languages ​​in response to a batch translation operation by the user, without requiring the user to select any translation result cells 416.

[0143] Let's return to Figure 13 for explanation. In step S402, the information processing system generates reference information. The reference information may include, for example, a glossary, existing translations, and context.

[0144] An information processing system may generate a glossary. For example, an information processing system may obtain the expression of a given term in the source language and the expression of that term in one or more target languages, and generate a glossary that links the source language expression with the expression in each target language. An information processing system may also obtain a descriptive text that explains the content of a term to be registered in the glossary, and generate a glossary that links the term with the descriptive text.

[0145] In text translation, the information processing system may detect terms in the source language registered in the glossary from the text to be translated, and translate the terms included in the text to be translated into the expressions in the target language registered in the glossary. For example, the information processing system may input a prompt including the text to be translated, a glossary that associates expressions in the source language with expressions in the target language, and an instruction to translate the text to be translated into the target language by referring to the glossary into the generation model, and obtain a translated text in the target language as its response described in natural language, thereby executing the translation.

[0146] Translation by referring to the glossary realizes highly consistent translation. Highly consistent translation, or translation that maintains consistency, includes uniformly translating the same terms (expressions) in the source language into the same expressions in the target language. For example, highly consistent translation includes uniformly translating Japanese expressions such as "strengthened" that appear multiple times in the text to be translated into English expressions such as "upgraded" based on the context of the translation.

[0147] In step S403, the information processing system translates the text to be translated from the source language into the target language. The information processing system registers (associates) the translation result in a record and displays the translation result in the translation result cell in the record.

[0148] The information processing system may receive an operation from the user to select a translation result cell. If the selected translation result cell is blank (untranslated state), the information processing system may translate the original text corresponding to the translation result cell into the target language corresponding to the translation result cell, and display the translated text in the translation result cell. Further, the information processing system may back-translate the translated text obtained by translating the original text into the target language from the target language into the source language, and display the result of the back-translation simultaneously. When the information processing system receives a registration operation from the user to register the displayed translated text as a translation result, the information processing system may register the translated text in the record of the original text.

[0149] The information processing system may receive a user's instruction to translate the source text of the current project in bulk, translate multiple source texts into their respective target languages, and register the translation results in each record. The information processing system may determine whether each translation result cell is blank or not, and for translation result cells determined to be blank, it may translate the source text corresponding to that translation result cell into the target language corresponding to that translation result cell, and register the translation results in the source text record.

[0150] The information processing system may analyze multiple texts to be translated in a batch and automatically detect terms from the multiple texts that should be associated with their expressions in the source language and the target language. The detected terms may include, for example, terms that appear more than a predetermined number of times (frequently occurring terms) throughout the multiple texts, keywords such as technical terms and specialized terms, and terms specified by the generative model from the multiple texts.

[0151] The information processing system may obtain pairs of terms and their translations from the source texts as reference information by translating terms detected from multiple texts into the target language. In this case, the information processing system may translate the detected terms using either project information or reference information. The information processing system may use the detected terms and their translations to translate each text to be translated in batch from the source language to the target language, thereby unifying the same terms and expressions appearing in each text to be translated into the same expression in the target language.

[0152] The information processing system may use a glossary preferentially over detected term-and-translate pairs when performing translations. The information processing system may also preferentially use translations registered in the glossary for the same term. The information processing system may automatically or manually register detected term-and-translate pairs in the glossary. This ensures consistent translations across multiple texts. The information processing system may automatically register detected term-and-translate pairs in the glossary.

[0153] The information processing system may generate multiple translation candidates (multiple translations) by translating a single source text into a target language in multiple variations, display each translation candidate, and register the selected translation candidate (translation) as the translation result by accepting a selection operation from the user as a registration operation. When displaying each translation candidate, the information processing system may back-translate each translation candidate and display the result of the back-translation.

[0154] The information processing system may input a prompt into a generative model that includes a term obtained from the source text, a source language, a target language, and an instruction to translate the term from the source language to the target language, and obtain the translation of the term from the generative model as a response written in natural language. The prompt may include either project information or reference information, and the instructions included in the prompt may include instructions to translate the term by referring to them.

[0155] An information processing system may input a prompt to a generative model that includes source text, a source language, a target language, and instructions to translate the source text from the source language to the target language, and obtain a translation of the source text from the generative model as a response written in natural language. The prompt may include either project information or reference information, and the instructions included in the prompt may include instructions to translate the source text by referring to them.

[0156] The information processing system may input a prompt to a generative model that includes a translated text in the target language, the source language, the target language, and an instruction to translate the translated text from the target language to the source language (a back-translation instruction). The system may then obtain from the generative model, in response written in natural language, the back-translated text of the translated text. The prompt may include either project information or reference information, and the back-translation instruction included in the prompt may include an instruction to translate the translated text by referring to that information.

[0157] Hereafter, translation or back-translation of text or terminology using generative models may be performed in the same manner.

[0158] Figure 16 shows an example of displaying multiple translation candidates and their back-translation results. In Figure 16, two translations (translation candidates) 421 (421-1, 421-2) obtained by translating the text of the first record of the source data shown in Figure 15 from the source language Japanese to the target language English, and their back-translations 422 (422-1, 422-2) are displayed. For example, the back-translation 422 may be displayed adjacent to the corresponding translation candidate 421 directly below it. In Figure 16, back-translation 422-1 is the result of the back-translation corresponding to translation candidate 421-1. Similarly, back-translation 422-2 is the result of the back-translation corresponding to translation candidate 421-2.

[0159] When the translation result for the source text is registered in a record, the information processing system may automatically detect terms from the translation result (translated text) and display the detected terms. The information processing system may also automatically detect expressions corresponding to the terms detected from the translated text in the source text and display the expressions detected from the source text and the terms (expressions) detected from the translated text in association.

[0160] When the source text is registered in a record, or when the translation result for the source text is registered in a record, the information processing system may automatically detect terms from the source text and display the detected terms. The information processing system may also automatically detect expressions corresponding to the terms detected from the source text in the translation and display the terms (expressions) detected from the source text and the expressions detected from the translation in association.

[0161] The information processing system may analyze multiple texts to be translated and automatically detect from the multiple texts the "terms that associate expressions in the source language with expressions in the target language" as described above. The information processing system may translate the terms detected from the multiple texts into the target language and display the term and / or its translation if the translation result is included in the registered translation results. The information processing system may display the term and / or its translation if the terms detected from the multiple texts are included in the source text corresponding to the registered translation results.

[0162] Figure 17 shows an example of how terms detected from the translated text are displayed. In the example shown in Figure 17, "sabotage" is detected from the translated text in the target language, English, and the corresponding expression in the original text, "interference," is displayed in association with "sabotage."

[0163] The information processing system may register detected terms (expressions in the source language) and their corresponding translations (expressions in the target language) in a glossary. This allows the information processing system to enrich its glossary during the translation process of multiple texts without requiring manual management. Furthermore, by utilizing a comprehensive glossary, the information processing system can achieve highly consistent translations.

[0164] When an information processing system has a registered translation result for the target language (an existing translation of the target language) in its records, it may automatically refer to the existing translation when translating source text from the same project into the target language. In this case, the information processing system may automatically refer not only to the existing translation but also to the source text of that existing translation. That is, the information processing system may translate another source text based on the existing translation and / or the source text of that existing translation.

[0165] The information processing system may automatically search for a translated source text related to the source text to be translated from among multiple translated source texts, and automatically refer to the registered translation and / or the source text for the found source text. For example, the information processing system may use a text embedding model such as sentence2vec, doc2vec, or PLaMo-Embedding-1B to calculate the embedding vectors of multiple source texts in the project, and then search for the source text closest to the source text to be translated based on the distance in the embedding space, and refer to the translation result and / or the source text for the found source text. This ensures that translations maintain consistency with existing translations.

[0166] Let's return to Figure 13 for explanation. In step S404, after the translation result for the source text is registered, the information processing system may perform an automatic review of the translation result, suggest a revised translation (revised version), or answer user questions and / or inquiries regarding the translation result. In step S405, the information processing system outputs the translation result after the review or revision is completed as the final translation result. The information processing system may perform these reviews, suggestion of revised versions, or answer user questions and / or inquiries via a chat-style user interface.

[0167] Figure 18 shows an example of a chat-style user interface display when reviewing translation results. Figure 18 shows a chat where the information processing system presents reviews and suggested revisions to the translation results, the user asks questions, and the information processing system responds to the user's questions by presenting multiple expressions and their explanations.

[0168] The information processing system may display a review of the translation result, along with suggested revisions, based on the corresponding source text. The review and suggested revisions may include at least one aspect of the translation result's expression, such as spelling, grammar, terminology consistency, or cultural elements. For example, the information processing system may input a prompt into a generative model that includes a pair of the translation result and the source text, along with instructions to "review whether the translation result is appropriate as a translation of the source text and suggest revisions," and retrieve the review and suggested revisions of the translation result from the generative model as a response written in natural language. The instructions included in the prompt may include instructions to review the translation result's expression from one of the following perspectives and suggest revisions: spelling, grammar, terminology consistency, or cultural elements. One example of the cultural elements perspective is the impression the translation result's expression would give to a reader and / or listener if it were written and / or spoken in a country where the target language is used as an official and / or semi-official language.

[0169] The information processing system may highlight and display parts of the original text and / or translation results, depending on the content of the review. The information processing system may display a text explaining the content of the review in natural language (a review explanation). Prompts may include either project information or reference information, and instructions included in the prompts may include instructions to refer to them, review the information, and propose revisions.

[0170] An information processing system may accept user questions and / or inquiries about translation results in natural language and display answers thereto. For example, an information processing system may input a prompt into a generative model that includes pairs of translation results and source texts, user questions and / or inquiries, and instructions to answer them, and retrieve answers to the user questions and / or inquiries from the generative model as a response written in natural language. The prompt may further include either project information or reference information, and the instructions included in the prompt may include instructions to answer by referring to that information.

[0171] If the translation result for the source text has not yet been registered, the information processing system may provide a proposed translation of the source text and answer user questions and / or inquiries regarding the proposed translation. The information processing system may provide these proposed translations and answer user questions and / or inquiries via a chat-style user interface.

[0172] Figure 19 shows an example of a chat-style user interface display when reviewing a source text. Figure 19 shows a chat where the information processing system presents a review and translation proposal for the source text, the user consults with it, and the information processing system presents multiple expressions as a response to the user's inquiry. In the chat shown in Figure 19, the user responds with an option selected from the multiple expression choices presented by the information processing system, and the information processing system asks whether the user is okay with finalizing the translation result based on that response.

[0173] An information processing system may display a proposed translation of a source text, along with a review of the source text. For example, an information processing system may input a prompt into a generative model that includes the source text and an instruction to "review the source text and propose a proposed translation into the target language," and retrieve a review of the source text and a proposed translation from the generative model as a response written in natural language. The prompt may further include either project information or reference information, and the instructions included in the prompt may include instructions to propose a translation by referring to that information.

[0174] An information processing system may accept user questions and / or inquiries regarding translation proposals in natural language and display the answers thereto. For example, the information processing system may input a prompt into a generative model that includes pairs of source text and translation proposals, user questions and / or inquiries, and instructions to answer them, and retrieve the answers to the user questions and / or inquiries from the generative model as a response written in natural language. The prompt may further include either project information or reference information, and the instructions included in the prompt may include instructions to answer by referring to that information.

[0175] (Other embodiments) In the embodiments described above, some or all of the devices (client devices, server devices, information processing devices, information processing systems) may be composed of hardware, or they may be composed of information processing by software (programs) executed by a CPU (Central Processing Unit), GPU (Graphics Processing Unit), etc. If the information processing is composed of software, the software that realizes at least some of the functions of each device in the embodiments described above may be stored on a non-temporary storage medium (non-temporary computer-readable medium) such as a CD-ROM (Compact Disc-Read Only Memory) or USB (Universal Serial Bus) memory, and the software information processing may be executed by loading it into a computer. Alternatively, the software may be downloaded via a communication network. Furthermore, all or part of the software processing may be implemented in a circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), so that the information processing by the software is executed by hardware.

[0176] The storage medium for the software may be a removable medium such as an optical disc, or a fixed storage medium such as a hard disk or memory. Furthermore, the storage medium may be located inside the computer (main memory, auxiliary storage, etc.) or outside the computer.

[0177] Figure 20 is a block diagram showing an example of the hardware configuration of each device (client device, server device, information processing device, information processing system) in the embodiment described above. Each device may be implemented as a computer 7, for example, comprising a processor 71, main memory 72 (memory), auxiliary memory 73 (memory), network interface 74, and device interface 75, which are connected via a bus 76.

[0178] The computer 7 in Figure 20 has one of each component, but it may have multiple identical components. Also, although Figure 20 shows one computer 7, the software may be installed on multiple computers, and each of these multiple computers may execute the same or different parts of the software's processing. In this case, it may be a distributed computing configuration in which each computer communicates via a network interface 74 or the like to execute processing. In other words, each device (client device, server device, information processing device, information processing system) in the above-described embodiment may be configured as a system that realizes its function by one or more computers executing instructions stored in one or more storage devices. Alternatively, it may be configured so that information transmitted from a terminal is processed by one or more computers located on the cloud, and the processing results are transmitted to the terminal.

[0179] The various calculations performed by each device (client device, server device, information processing device, information processing system) in the embodiments described above may be performed in parallel using one or more processors, or using multiple computers via a network. Alternatively, the various calculations may be distributed to multiple processing cores within a processor and performed in parallel. Furthermore, some or all of the processing and means of this disclosure may be implemented by at least one of a processor and a storage device located on a cloud that can communicate with computer 7 via a network. Thus, each device in the embodiments described above may be in the form of parallel computing using one or more computers.

[0180] The processor 71 may be an electronic circuit (processing circuit, processing circuitry, CPU, GPU, FPGA, ASIC, etc.) that performs either control or calculations of a computer. The processor 71 may also be a general-purpose processor, a dedicated processing circuit designed to perform specific calculations, or a semiconductor device that includes both a general-purpose processor and a dedicated processing circuit. Furthermore, the processor 71 may include optical circuits or quantum computing-based calculation functions.

[0181] The processor 71 may perform calculations based on data and software input from various devices within the computer 7, and may output calculation results and control signals to these devices. The processor 71 may also control the various components of the computer 7 by executing the computer 7's OS (Operating System) or applications.

[0182] Each of the devices (client device, server device, information processing device, information processing system) in the above-described embodiment may be implemented by one or more processors 71. Here, the processor 71 may refer to one or more electronic circuits arranged on one chip, or one or more electronic circuits arranged on two or more chips or two or more devices. When multiple electronic circuits are used, each electronic circuit may communicate by wire or wireless.

[0183] The main memory 72 may store instructions executed by the processor 71 and various data, and the information stored in the main memory 72 may be read by the processor 71. The auxiliary memory 73 is a memory device other than the main memory 72. These memory devices refer to any electronic component capable of storing electronic information, and may be semiconductor memory. The semiconductor memory may be either volatile memory or non-volatile memory. In each of the devices (client device, server device, information processing device, information processing system) in the above-described embodiment, the memory device for storing various data may be implemented by the main memory 72 or the auxiliary memory 73, or by the built-in memory of the processor 71. For example, each storage unit in the above-described embodiment may be implemented by the main memory 72 or the auxiliary memory 73.

[0184] In the embodiments described above, if each device (client device, server device, information processing device, information processing system) consists of at least one storage device (memory) and at least one processor connected to (coupled with) this at least one storage device, then at least one processor may be connected to one storage device. Also, at least one storage device may be connected to one processor. Furthermore, the configuration may include at least one processor among a plurality of processors being connected to at least one storage device among a plurality of storage devices. This configuration may also be realized by storage devices and processors included in a plurality of computers. Moreover, the configuration may include a storage device integrated with a processor (for example, a cache memory including an L1 cache and an L2 cache).

[0185] The network interface 74 is an interface for connecting to the communication network 8 wirelessly or via a wired connection. The network interface 74 can be any appropriate interface, such as one conforming to existing communication standards. Information may be exchanged between the computer 7 and an external device 9A connected via the communication network 8 through the network interface 74. The communication network 8 may be a WAN (Wide Area Network), LAN (Local Area Network), PAN (Personal Area Network), or a combination thereof, as long as information is exchanged between the computer 7 and the external device 9A. An example of a WAN is the Internet, an example of a LAN is IEEE 802.11 or Ethernet (registered trademark), and an example of a PAN is Bluetooth (registered trademark) or NFC (Near Field Communication).

[0186] The device interface 75 is an interface such as USB that connects directly to the external device 9B.

[0187] External device 9A is a device connected to computer 7 via a network. External device 9B is a device directly connected to computer 7.

[0188] External device 9A or external device 9B may, for example, be an input device. The input device may be a camera, microphone, motion capture device, various sensors, keyboard, mouse, touch panel, etc., and provides the acquired information to the computer 7. Alternatively, it may be a device equipped with an input unit, memory, and processor, such as a personal computer, tablet terminal, or smartphone.

[0189] Furthermore, external device 9A or external device 9B may, for example, be an output device. The output device may be a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) panel, or a speaker that outputs sound, etc. It may also be a device equipped with an output unit, memory, and a processor, such as a personal computer, tablet terminal, or smartphone.

[0190] Furthermore, external devices 9A and 9B may be storage devices (memory). For example, external device 9A may be network storage, and external device 9B may be storage such as an HDD.

[0191] Furthermore, the external device 9A or external device 9B may be a device that has some of the functions of the components of each device (client device, server device, information processing device, information processing system) in the embodiments described above. In other words, the computer 7 may transmit some or all of the processing results to the external device 9A or external device 9B, or may receive some or all of the processing results from the external device 9A or external device 9B.

[0192] In this specification (including the claims), when the expression "at least one of a, b, and c" or "at least one of a, b, or c" (including similar expressions) is used, it includes any of a, b, c, ab, ac, bc, or abc. Furthermore, any element may have multiple instances, such as aa, abb, aabbcc, etc. In addition, it is also possible to add other elements other than the enumerated elements (a, b, and c), such as abcd which has d.

[0193] In this specification (including the claims), when expressions such as "using data as input / based on data / according to / in accordance with data" (including similar expressions) are used, unless otherwise specified, this includes using the data itself or using data that has been processed in some way (e.g., data with added noise, normalized data, features extracted from the data, intermediate representations of the data, etc.). Furthermore, when it is stated that some result is obtained "using data as input / based on data / according to / in accordance with data" (including similar expressions), unless otherwise specified, this includes cases where the result is obtained based solely on the data in question or where the result is influenced by other data, factors, conditions, and / or states other than the data in question. Furthermore, when it is stated that "data is output" (including similar expressions), unless otherwise specified, this includes cases where the data itself is used as output or where data that has been processed in some way (e.g., data with added noise, normalized data, features extracted from the data, intermediate representations of various types of data, etc.) is used as output.

[0194] In this specification (including the claims), the terms “connected” and “coupled” are intended to be non-restrictive terms that include any direct connection / coupling, indirect connection / coupling, electrical connection / coupling, communicative connection / coupling, operational connection / coupling, physical connection / coupling, etc. The terms should be interpreted as appropriate in the context in which they are used, but any form of connection / coupling that is not intentionally or naturally excluded should be interpreted non-restrictively as being included in the terms.

[0195] In this specification (including the claims), when the expression "A configured to B" is used, it may include that the physical structure of element A has a configuration capable of performing operation B, and that the permanent or temporary setting / configuration of element A is configured to actually perform operation B. For example, if element A is a general-purpose processor, it is sufficient that the processor has a hardware configuration capable of performing operation B, and that it is configured to actually perform operation B by the setting of a permanent or temporary program (instruction). Furthermore, if element A is a dedicated processor, dedicated arithmetic circuit, etc., it is sufficient that the circuit structure of the processor is implemented to actually perform operation B, regardless of whether control instructions and data are actually attached.

[0196] Wherever terms meaning "comprising" or "possessing" (e.g., "comprising / including," "having," etc.) are used herein, they are intended to be open-ended terms, including cases where the subject matter of such terms is not the object of the term. Where the object of such terms meaning "comprising" or "possessing" is an expression that does not specify a quantity or suggests a singular number (an expression with the article "a" or "an"), such expression should be interpreted as not being limited to a specific number.

[0197] In this specification (including the claims), even if expressions such as "one or more" or "at least one" are used in some places, and expressions that do not specify a quantity or suggest a singularity (expressions using the articles a or an) are used in other places, the latter expressions are not intended to mean "one." In general, expressions that do not specify a quantity or suggest a singularity (expressions using the articles a or an) should not necessarily be interpreted as not being limited to a specific number.

[0198] In this specification, if a particular configuration of an embodiment is described as having a specific advantage or result, it should be understood, unless otherwise stated, that the same advantage or result can also be obtained from one or more other embodiments having that configuration. However, it should be understood that the presence or absence of such advantage or result generally depends on various factors, conditions, and / or states, and that the configuration does not necessarily guarantee that the advantage or result can be obtained. The advantage or result can only be obtained from the configuration described in the embodiment when various factors, conditions, and / or states are met, and the advantage or result cannot necessarily be obtained in the invention claimed to define that configuration or a similar configuration.

[0199] In this specification (including the claims), when multiple hardware components perform a predetermined process, each component may cooperate to perform the predetermined process, or some components may perform all of the predetermined process. Alternatively, some components may perform part of the predetermined process, while other components perform the remainder. In this specification (including the claims), when expressions such as "one or more hardware components perform a first process, and the one or more hardware components perform a second process" (including similar expressions) are used, the hardware component performing the first process and the hardware component performing the second process may be the same or different. In other words, it is sufficient that the hardware component performing the first process and the hardware component performing the second process are included in the one or more hardware components. Hardware may include electronic circuits, devices containing electronic circuits, etc.

[0200] In this specification (including the claims), when multiple memory devices store data, each of the multiple memory devices may store only a portion of the data or the entire data. Furthermore, a configuration in which some of the multiple memory devices store data is also included.

[0201] In this specification (including the claims), terms such as “first,” “second,” etc., are used merely as a way of distinguishing between two or more elements and are not necessarily intended to impose technical meanings such as temporal, spatial, order, or quantity on the subject. Therefore, for example, references to a first element and a second element do not necessarily mean that only two elements can be employed therein, that the first element must precede the second element, or that the first element must exist for the second element to exist.

[0202] While embodiments of this disclosure have been described in detail above, this disclosure is not limited to the individual embodiments described above. Various additions, modifications, substitutions, and partial deletions are possible, provided that they do not depart from the conceptual idea and spirit of the present invention derived from the claims and their equivalents. For example, where numerical values ​​or mathematical formulas are used in the description of the embodiments described above, these are provided for illustrative purposes only and do not limit the scope of this disclosure. Similarly, the sequence of operations shown in the embodiments is also illustrative and does not limit the scope of this disclosure.

[0203] Furthermore, the following forms are possible for disclosure technology.

[0204] (Note 1) One or more memory devices, It has one or more processors, The one or more processors described above are: Display a list of multiple data points in the first language contained in the data file. From the multiple data displayed in the aforementioned list, one data selected by the user is translated into a second language using one or more generative models. One translation proposal obtained by translating the selected data into the second language using one or more generative models is stored in association with the selected data. Information processing system.

[0205] (Note 2) The one or more processors described above are: Multiple translation options are generated by translating the selected data into the second language using one or more generative models. The selected data is displayed as one of several translations translated into the second language by one or more generative models. One translation selected by the user from the aforementioned multiple translation options is saved in association with the selected data. The information processing system described in Appendix 1.

[0206] (Note 3) The one or more processors described above are: We will accept feedback from the user regarding the direction of editing for the aforementioned translation proposal. Based on the aforementioned editing direction, one or more edited versions are generated by editing the one translation draft using the one or more generative models. The one translation proposal displays the one or more translation proposals edited by the one or more generative models, One editing proposal selected by the user from the one or more editing proposals is saved in association with the selected data. The information processing system described in Appendix 1 or 2.

[0207] (Note 4) The one or more processors described above are: We received instructions from the user to refine the aforementioned one translation proposal. By refining one translation proposal using one or more generative models based on the aforementioned refinement instructions, one or more refined proposals are generated. The aforementioned one translation proposal displays the aforementioned one or more refined proposals refined by the aforementioned one or more generative models, The refinement proposal selected by the user from the one or more refinement proposals is associated with the selected data and saved. An information processing system as described in any of the appendices 1 to 3.

[0208] (Note 5) The one or more processors described above are: The user provides a definition of the label to be given to the selected data. By referencing the information associated with the label as context to one or more generative models and translating the selected data into the second language, the multiple translation options are generated. The selected data is translated into the second language according to the context by one or more generative models, and the resulting multiple translation options are displayed. The information processing system described in Appendix 2.

[0209] (Note 6) The one or more processors described above are: The system receives a medium from the user that corresponds to the selected data. The media is used as context to reference one or more generative models, and the selected data is translated into the second language to generate the multiple translation options. The information processing system described in Appendix 5.

[0210] (Note 7) The one or more processors described above are: Detect terms from multiple data in the first language, The detected term is translated into the translation in the second language. Using the detected term and the translated term, multiple data in the first language are translated into data in the second language in a single operation. An information processing system as described in any of the appendices 1 to 6.

[0211] (Note 8) The one or more processors described above are: Translate the first data in the first language into the first data in the second language. Using the first data in the first language and the first data in the second language, the second data in the first language, which is included in the same project as the first data in the first language, is translated into the second data in the second language. An information processing system as described in any of the appendices 1 to 7.

[0212] (Note 9) The one or more processors described above are: Using project information for a project that translates multiple data in the first language, the multiple data in the first language is translated into multiple data in the second language. An information processing system as described in any of the appendices 1 to 8.

[0213] (Note 10) The one or more processors described above are: The data in the first language is translated into the data in the second language. Using the data for the first language and the data for the second language, the data for the second language is reviewed. Display the description of the aforementioned review. An information processing system as described in any of the appendices 1 through 9.

[0214] (Note 11) The one or more processors described above are: Display the data for the first language mentioned above. The system accepts user questions or inquiries regarding the translation of data in the first language into the second language via a chat-based user interface. In response to the received question or consultation, a response is generated that includes a proposed translation of the data in the first language into the second language. The response, including the proposed translation, is output to the chat-style user interface. An information processing system as described in any of the appendices 1 through 10.

[0215] (Note 12) The one or more processors described above are: The AI ​​agent is instructed to generate the answers to questions or inquiries received via the chat-style user interface. The information processing system described in Appendix 11.

[0216] (Note 13) The one or more processors described above are: The AI ​​agent is instructed to generate a processing procedure for performing a translation in response to the aforementioned question or consultation, and to sequentially execute the generated processing procedure. The information processing system described in Appendix 12.

[0217] This application claims priority to Provisional Application No. 63 / 705,194 filed with the United States Patent and Trademark Office on 9 October 2024, and Provisional Application No. 63 / 792,553 filed with the United States Patent and Trademark Office on 22 April 2025, which are incorporated herein by reference to their entire contents. [Explanation of symbols]

[0218] 7: Computer 71: Processor 72: Main memory 73:Auxiliary storage device 74: Network Interface 75: Device Interface 8: Communication Network 9A,9B: External device

Claims

1. One or more memory devices, It has one or more processors, The one or more processors described above are: Display a list of multiple data points in the first language contained in the data file. From the multiple data displayed in the aforementioned list, one data selected by the user is translated into a second language using one or more generative models. One translation proposal obtained by translating the selected data into the second language using one or more generative models is stored in association with the selected data. Information processing system.

2. The one or more processors described above are: Multiple translation options are generated by translating the selected data into the second language using one or more generative models. The selected data is displayed as one of several translation options translated into the second language by one or more generative models. One translation selected by the user from the aforementioned multiple translation options is saved in association with the selected data. The information processing system according to claim 1.

3. The one or more processors described above are: We will accept feedback from the user regarding the direction of editing for the aforementioned translation proposal. Based on the aforementioned editing direction, one or more edited versions are generated by editing one of the aforementioned translation proposals using one or more generative models. The one translation proposal displays the one or more translation proposals edited by the one or more generative models, One editing proposal selected by the user from the one or more editing proposals is saved in association with the selected data. The information processing system according to claim 1.

4. The one or more processors described above are: We received instructions from the user to refine the aforementioned translation proposal. Based on the instructions to refine, one translation proposal is refined by one or more generative models to generate one or more refined proposals. The aforementioned one translation proposal displays the aforementioned one or more refined proposals refined by the aforementioned one or more generative models, The refinement proposal selected by the user from the one or more refinement proposals is saved in association with the selected data. The information processing system according to claim 1.

5. The one or more processors described above are: The user provides the definition of the label to be given to the selected data. The information associated with the label is used as context for one or more generative models, and the selected data is translated into the second language to generate the multiple translation options. The selected data is translated into the second language according to the context by one or more generative models, and the resulting multiple translation options are displayed. The information processing system according to claim 2.

6. The one or more processors described above are: The system receives a medium from the user that corresponds to the selected data. The media is used as context to reference one or more generative models, and the selected data is translated into the second language to generate the multiple translation options. The information processing system according to claim 5.

7. The one or more processors described above are: Detect terms from multiple data in the first language, The detected term is translated into the translation in the second language. Using the detected term and the translated term, multiple data in the first language are translated into data in the second language in a single operation. An information processing system according to any one of claims 1 to 6.

8. The one or more processors described above are: The first data in the first language is translated into the first data in the second language. Using the first data in the first language and the first data in the second language, the second data in the first language, which is included in the same project as the first data in the first language, is translated into the second data in the second language. An information processing system according to any one of claims 1 to 6.

9. The one or more processors described above are: Using project information for a project that translates multiple data in the first language, the multiple data in the first language is translated into multiple data in the second language. An information processing system according to any one of claims 1 to 6.

10. The one or more processors described above are: The data in the first language is translated into the data in the second language. Using the data of the first language and the data of the second language, the data of the second language is reviewed. Display the description of the aforementioned review. An information processing system according to any one of claims 1 to 6.

11. The one or more processors described above are: Display the data for the first language, The system accepts user questions or inquiries regarding the translation of data in the first language into the second language via a chat-based user interface. In response to the received question or consultation, a response is generated that includes a proposed translation of the data in the first language into the second language. The response, including the proposed translation, is output to the chat-style user interface. An information processing system according to any one of claims 1 to 6.

12. The one or more processors described above are: The AI ​​agent is instructed to generate the answers to questions or inquiries received through the chat-style user interface. The information processing system according to claim 11.

13. The one or more processors described above are: The AI ​​agent is instructed to generate a processing procedure for performing a translation in response to the aforementioned question or consultation, and to sequentially execute the generated processing procedure. The information processing system according to claim 12.

Images