System

Abstract

An object of the system according to the embodiment is to translate audio of a moving image in real time and automatically generate subtitles.SOLUTION: A system according to an embodiment includes a speech recognition unit, a translation unit, and a subtitle generation unit. The voice recognition unit converts voice of the moving image into text. The translation unit translates the text converted by the speech recognition unit into a specified language. The caption generation unit generates a caption on the basis of the text translated by the translation unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional technology has the drawback of requiring the manual translation of video audio and creation of subtitles, making it difficult to quickly accommodate viewers in different language regions.

[0005] The system according to the embodiment aims to translate the audio of a video in real time and automatically generate subtitles. [Means for solving the problem]

[0006] The system according to the embodiment includes a speech recognition unit, a translation unit, and a subtitle generation unit. The speech recognition unit converts the speech of a video into text. The translation unit translates the text converted by the speech recognition unit into a specified language. The subtitle generation unit generates subtitles based on the text translated by the translation unit. [Effects of the Invention]

[0007] The system according to the embodiment can translate the audio of a video in real time and automatically generate subtitles. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE INVENTION

[0009] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), and Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A subtitle generation system according to an embodiment of the present invention is a system that translates the audio of a video in real time and automatically generates subtitles. The subtitle generation system analyzes the audio of a video in real time, and a generation AI converts the audio into text. Next, the generation AI translates the converted text into a specified language. Finally, subtitles are automatically generated based on the translated text and displayed on the video. For example, the subtitle generation system analyzes the audio of a video in real time, and a generation AI converts the audio in the video into text using speech recognition technology. Next, the generation AI translates the converted text into a specified language. For example, the generation AI translates the acquired text into a specified language using a multilingual translation model. Finally, the generation AI displays the translated text as subtitles aligned with the video timeline. For example, the generation AI automatically adjusts the timing so that the content spoken in a specific scene of the video is displayed as subtitles. This allows the subtitle generation system to eliminate the need for manual subtitle creation and enable the creation of content that is easily accessible to viewers from different language regions. This allows the subtitle generation system to eliminate the need for manual subtitle creation and enable the creation of content that is easily accessible to viewers from different language regions. For example, by automatically generating subtitles in Japanese, Chinese, etc. for a video produced in English, it is possible to provide content that is easy to understand for viewers who speak different languages.

[0029] A subtitle generation system according to an embodiment includes a speech recognition unit, a translation unit, and a subtitle generation unit. The speech recognition unit converts video audio into text. The speech recognition unit, for example, uses a generation AI to analyze the video audio in real time and acquire the resulting text data. The speech recognition unit can also support multiple languages ​​using a speech recognition algorithm. For example, the speech recognition unit recognizes speech input in multiple languages, such as English, Japanese, and French, and converts the speech input into text data. The speech recognition unit can also use noise canceling technology to remove background sounds and noise and acquire clearer audio. For example, the speech recognition unit can remove specific frequency bands and emphasize the speaker's voice. The translation unit uses a generation AI to translate the text converted by the speech recognition unit into a specified language. The translation unit, for example, uses a multilingual translation model to translate the acquired text into a specified language. For example, the translation unit translates the text using technologies such as neural machine translation (NMT) and statistical machine translation (SMT). The translation unit can also perform context analysis to generate natural-sounding translations. For example, the translation unit analyzes the context of the video and selects appropriate expressions. The subtitle generation unit generates subtitles based on the text translated by the translation unit. The subtitle generation unit, for example, uses a generation AI to display the translated text as subtitles in accordance with the timeline of the video. For example, the subtitle generation unit adjusts the timing so that the content spoken in a specific scene of the video is displayed as subtitles. Furthermore, the subtitle generation unit can customize the font and color of the subtitles to improve visibility. For example, the subtitle generation unit changes the font and color based on the content of the video. As a result, the subtitle generation system according to the embodiment can translate the audio of the video in real time and automatically generate subtitles.

[0030] The speech recognition unit can analyze the audio of a video in real time and convert it into text. For example, the speech recognition unit uses a generation AI to analyze the audio of a video in real time and acquire it as text data. For example, the speech recognition unit analyzes words spoken in a video in real time and acquires them as text data. The speech recognition unit can also support multiple languages ​​using a speech recognition algorithm. For example, the speech recognition unit recognizes voice input in multiple languages, such as English, Japanese, and French, and converts it into text data. Furthermore, the speech recognition unit can use noise canceling technology to remove background sounds and noise and acquire clear audio. For example, the speech recognition unit can remove specific frequency bands and emphasize the speaker's voice. This allows the audio of a video to be converted into text in real time. Some or all of the above-mentioned processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input the audio data of a video into the generation AI and cause the generation AI to generate text data from the audio data.

[0031] The translation unit can translate the acquired text into a specified language using a multilingual translation model. The translation unit translates the acquired text into a specified language using, for example, a generation AI. For example, the translation unit translates the text using technologies such as neural machine translation (NMT) or statistical machine translation (SMT). Furthermore, the translation unit can perform context analysis to generate natural-sounding translations. For example, the translation unit analyzes the context before and after the video and selects appropriate expressions. This allows the acquired text to be translated into multiple languages. Some or all of the above-mentioned processing in the translation unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the translation unit can input the acquired text data into a generation AI and have the generation AI perform translation from the text data.

[0032] The subtitle generation unit can display the translated text as subtitles in accordance with the timeline of the video. The subtitle generation unit, for example, uses a generation AI to display the translated text as subtitles in accordance with the timeline of the video. For example, the subtitle generation unit adjusts the timing so that the content spoken in a specific scene of the video is displayed as subtitles. Furthermore, the subtitle generation unit can customize the font and color of the subtitles to improve visibility. For example, the subtitle generation unit changes the font and color based on the content of the video. This allows the translated text to be displayed as subtitles in accordance with the timeline of the video. Some or all of the above-described processing in the subtitle generation unit may be performed using, or without, the generation AI. For example, the subtitle generation unit can input translated text data to the generation AI and cause the generation AI to generate subtitles.

[0033] The speech recognition unit can automatically remove background sounds and noise from a video during speech recognition to make the speech clearer. The speech recognition unit can automatically remove background sounds and noise from a video to make the speech clearer, for example, using a generation AI. For example, the speech recognition unit uses a generation AI to analyze background sounds from a video and filter out noise unnecessary for speech recognition. The speech recognition unit can also use a generation AI to remove specific frequency bands and emphasize the speaker's voice. The speech recognition unit can also use a generation AI to perform noise cancellation in real time to obtain clear speech. This makes it possible to remove background sounds and noise during speech recognition and obtain clear speech. Some or all of the above-described processing in the speech recognition unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the speech recognition unit can input speech data from a video to the generation AI and have the generation AI perform noise removal and speech recognition.

[0034] The speech recognition unit can analyze the characteristics of a speaker's voice during speech recognition and apply a different recognition model to each speaker. The speech recognition unit can, for example, use a generation AI to analyze the characteristics of a speaker's voice during speech recognition and apply a different recognition model to each speaker. For example, the speech recognition unit can have the generation AI analyze the pitch and tone of the speaker's voice and apply an individual recognition model. The speech recognition unit can also have the generation AI learn the speaker's pronunciation habits and improve recognition accuracy. The speech recognition unit can also have the generation AI store the speaker's voice characteristics in a database and use them for subsequent recognition. This allows for applying a different recognition model to each speaker, thereby improving speech recognition accuracy. Some or all of the above-mentioned processing in the speech recognition unit can be performed using, or without, the generation AI. For example, the speech recognition unit can input the speaker's voice data into the generation AI and have the generation AI apply the recognition model for each speaker.

[0035] The speech recognition unit can apply different speech recognition algorithms to each video scene during speech recognition. For example, the speech recognition unit uses a generation AI to apply different speech recognition algorithms to each video scene during speech recognition. For example, the speech recognition unit uses a generation AI to analyze video scenes and select the optimal speech recognition algorithm for each scene. The speech recognition unit can also use the generation AI to perform appropriate noise filtering by taking into account background sounds in the scene. The speech recognition unit can also improve the recognition accuracy of technical terms and proper nouns depending on the content of the scene. This allows the application of the optimal speech recognition algorithm for each scene, thereby improving the accuracy of speech recognition. Some or all of the above-described processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input video scene data to the generation AI and have the generation AI apply the speech recognition algorithm for each scene.

[0036] The speech recognition unit can recognize regional accents and dialects by taking into account the geographical location information of the viewer of the video during speech recognition. The speech recognition unit, for example, uses a generation AI to recognize regional accents and dialects by taking into account the geographical location information of the viewer of the video during speech recognition. For example, the speech recognition unit uses a generation AI to acquire the viewer's geographical location information and recognize the regional accent. The speech recognition unit can also have the generation AI learn the viewer's dialect to improve recognition accuracy. The speech recognition unit can also have the generation AI store regional phrases and expressions in a database and use them for subsequent recognition. This allows the recognition of regional accents and dialects to improve speech recognition accuracy. Some or all of the above-described processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input the viewer's geographical location information to the generation AI and have the generation AI recognize regional accents and dialects.

[0037] The speech recognition unit can prioritize recognition of technical terms and proper nouns based on the content of the video during speech recognition. The speech recognition unit, for example, uses a generation AI to prioritize recognition of technical terms and proper nouns based on the content of the video during speech recognition. For example, the speech recognition unit uses a generation AI to analyze the content of the video and prioritize recognition of technical terms and proper nouns. The speech recognition unit can also improve recognition accuracy by having the generation AI store technical terms from a specific field in a database. The speech recognition unit can also have the generation AI learn the pronunciation of proper nouns to reduce recognition errors. This prioritizes recognition of technical terms and proper nouns, thereby improving speech recognition accuracy. Some or all of the above-described processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input video content data into the generation AI and have the generation AI recognize technical terms and proper nouns.

[0038] The speech recognition unit can improve the recognition accuracy by reflecting the user's past viewing history during speech recognition. The speech recognition unit can improve the recognition accuracy by, for example, using a generation AI to reflect the user's past viewing history during speech recognition. For example, the speech recognition unit can improve the recognition accuracy by having the generation AI analyze the user's past viewing history. The speech recognition unit can also improve the recognition accuracy by having the generation AI learn the terminology of the genre the user frequently watches. The speech recognition unit can also improve the recognition accuracy by having the generation AI learn the vocal characteristics of a specific speaker from the user's viewing history. In this way, the accuracy of speech recognition can be improved by reflecting the user's past viewing history. Some or all of the above-described processing in the speech recognition unit can be performed, for example, using the generation AI, or can be performed without using the generation AI. For example, the speech recognition unit can input the user's viewing history data into the generation AI and have the generation AI analyze the viewing history and improve the recognition accuracy.

[0039] The translation unit can apply different translation algorithms depending on the genre of the video during translation. The translation unit, for example, uses a generation AI to apply different translation algorithms depending on the genre of the video during translation. For example, the translation unit uses a generation AI to analyze the genre of the video and select the optimal translation algorithm. The translation unit can also use a generation AI to provide casual translations for entertainment videos and formal translations for business videos. The translation unit can also use a generation AI to provide detailed translations including technical terms for educational videos. This allows for applying the optimal translation algorithm depending on the genre of the video, thereby improving the accuracy of the translation. Some or all of the above-mentioned processing in the translation unit may be performed using, or without, the generation AI. For example, the translation unit can input video genre data into the generation AI and have the generation AI apply a translation algorithm for each genre.

[0040] The translation unit can generate a natural-looking translation by taking into account the context of the video during translation. The translation unit can generate a natural-looking translation by using, for example, a generation AI during translation. For example, the translation unit can have the generation AI analyze the context before and after the video to provide a natural-looking translation. The translation unit can also have the generation AI select appropriate expressions according to the context, improving translation accuracy. The translation unit can also have the generation AI perform context analysis to reduce mistranslations. In this way, a natural-looking translation can be generated by taking into account the context of the video. Some or all of the above-mentioned processing in the translation unit can be performed using, for example, the generation AI, or can be performed without using the generation AI. For example, the translation unit can input context data of the video into the generation AI and have the generation AI perform context analysis and generate a natural-looking translation.

[0041] The translation unit can perform context analysis to preserve the speaker's intention and nuances during translation. The translation unit can use, for example, a generation AI to perform context analysis to preserve the speaker's intention and nuances during translation. For example, the translation unit can have the generation AI analyze the speaker's intention and provide a translation that preserves appropriate nuances. The translation unit can also have the generation AI perform context analysis to provide a translation that reflects the speaker's intention. The translation unit can also select appropriate expressions to preserve nuances. This allows for context analysis to preserve the speaker's intention and nuances, thereby providing a more accurate translation. Some or all of the above-mentioned processing in the translation unit can be performed using, for example, the generation AI, or can be performed without using the generation AI. For example, the translation unit can input speaker's intention data into the generation AI and have the generation AI perform context analysis and preserve nuances.

[0042] The translation unit can select appropriate expressions taking into consideration the cultural background of the viewer of the video when translating. The translation unit, for example, uses a generation AI to select appropriate expressions taking into consideration the cultural background of the viewer of the video when translating. For example, the translation unit uses a generation AI to analyze the viewer's cultural background and select appropriate expressions. The translation unit can also provide a translation that takes the viewer's culture into consideration. The translation unit can also use a generation AI to select expressions that do not lead to misunderstandings based on the cultural background. This makes it possible to select appropriate expressions by taking the viewer's cultural background into consideration. Some or all of the above-mentioned processing in the translation unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the translation unit can input cultural background data of the viewer into the generation AI and cause the generation AI to analyze the cultural background and select appropriate expressions.

[0043] The translation unit can appropriately translate technical terms and proper nouns based on the content of the video during translation. The translation unit, for example, uses a generation AI to appropriately translate technical terms and proper nouns based on the content of the video during translation. For example, the translation unit has the generation AI analyze the content of the video and appropriately translate technical terms and proper nouns. The translation unit can also improve translation accuracy by having the generation AI store technical terms from a specific field in a database. The translation unit can also have the generation AI learn how to translate proper nouns and reduce mistranslations. This can improve translation accuracy by appropriately translating technical terms and proper nouns based on the content of the video. Some or all of the above-mentioned processing in the translation unit may be performed using, or without, the generation AI. For example, the translation unit can input video content data into the generation AI and have the generation AI translate the technical terms and proper nouns.

[0044] The translation unit can improve translation accuracy by reflecting the user's past translation history during translation. The translation unit can improve translation accuracy by using, for example, a generation AI to reflect the user's past translation history during translation. For example, the translation unit can improve translation accuracy by having the generation AI analyze the user's past translation history. The translation unit can also improve translation accuracy by having the generation AI learn expressions frequently used by the user. The translation unit can also improve translation accuracy by having the generation AI learn specific phrases and expressions from the user's translation history. In this way, translation accuracy can be improved by reflecting the user's past translation history. Some or all of the above-mentioned processing in the translation unit can be performed using, or without, the generation AI. For example, the translation unit can input the user's translation history data into the generation AI and have the generation AI analyze the translation history and improve translation accuracy.

[0045] The subtitle generation unit can apply different subtitle styles to each video scene when generating subtitles. The subtitle generation unit can apply different subtitle styles to each video scene when generating subtitles, for example, using a generation AI. For example, the subtitle generation unit analyzes video scenes using the generation AI and selects the optimal subtitle style for each scene. The subtitle generation unit can also change the font and color depending on the content of the scene. The subtitle generation unit can also adjust the display method of subtitles to match the atmosphere of the scene. This allows the optimal subtitle style to be applied to each scene, thereby improving visibility. Some or all of the above-mentioned processing in the subtitle generation unit can be performed using, or without, the generation AI. For example, the subtitle generation unit can input video scene data to the generation AI and cause the generation AI to apply a subtitle style for each scene.

[0046] The subtitle generation unit can automatically adjust the display timing of subtitles to match the timeline of the video when generating subtitles. The subtitle generation unit can automatically adjust the display timing of subtitles to match the timeline of the video when generating subtitles, for example, using a generation AI. For example, the subtitle generation unit uses a generation AI to analyze the timeline of the video and set the optimal subtitle display timing. The subtitle generation unit can also adjust the display timing of subtitles to match the start and end of audio. The subtitle generation unit can also adjust the display timing of subtitles to match scene changes. This can improve the viewing experience by adjusting the display timing of subtitles to match the timeline of the video. Some or all of the above-described processing in the subtitle generation unit can be performed using, or without, the generation AI. For example, the subtitle generation unit can input video timeline data to the generation AI and cause the generation AI to adjust the display timing of subtitles.

[0047] The subtitle generation unit can customize the font and color of subtitles based on the content of the video when generating subtitles. The subtitle generation unit, for example, uses a generation AI to customize the font and color of subtitles based on the content of the video when generating subtitles. For example, the subtitle generation unit uses the generation AI to analyze the content of the video and select a font and color that matches the scene. The subtitle generation unit can also set a font and color that is highly visible to match the atmosphere of the video. The subtitle generation unit can also apply different fonts and colors to parts that the generation AI wants to emphasize in a specific scene. In this way, visibility can be improved by customizing the font and color of subtitles based on the content of the video. Some or all of the above-mentioned processing in the subtitle generation unit may be performed using, or without, the generation AI. For example, the subtitle generation unit can input video content data into the generation AI and have the generation AI customize the font and color.

[0048] The subtitle generation unit can select the optimal subtitle display method by taking into account the device information of the viewer of the video when generating subtitles. The subtitle generation unit, for example, uses a generation AI to select the optimal subtitle display method by taking into account the device information of the viewer of the video when generating subtitles. For example, the subtitle generation unit uses the generation AI to acquire the viewer's device information and provide a subtitle display method that matches the screen size. The subtitle generation unit can also set the optimal font size according to the resolution of the viewer's device. The subtitle generation unit can also adjust the display position of the subtitles according to the display mode (portrait or landscape) of the viewer's device. In this way, the optimal subtitle display method can be provided by taking into account the viewer's device information. Some or all of the above-described processing in the subtitle generation unit may be performed using, for example, the generation AI, or may be performed without using the generation AI. For example, the subtitle generation unit can input the viewer's device information into the generation AI and have the generation AI analyze the device information and select the optimal subtitle display method.

[0049] The subtitle generation unit can adjust the length and display time of subtitles based on the content of the video when generating subtitles. The subtitle generation unit adjusts the length and display time of subtitles based on the content of the video when generating subtitles, for example, using a generation AI. For example, the subtitle generation unit analyzes the content of the video and adjusts the length of the subtitles appropriately. The subtitle generation unit can also adjust the display time of subtitles to match the speed of the audio. The subtitle generation unit can also adjust the display time of subtitles to match scene changes. In this way, by adjusting the length and display time of subtitles based on the content of the video, visibility can be improved. Some or all of the above-mentioned processing in the subtitle generation unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the subtitle generation unit can input content data of the video to the generation AI and cause the generation AI to adjust the length and display time of the subtitles.

[0050] The subtitle generation unit can customize the display method by reflecting the user's past subtitle display history when generating subtitles. The subtitle generation unit, for example, uses a generation AI to customize the display method by reflecting the user's past subtitle display history when generating subtitles. For example, the subtitle generation unit uses the generation AI to analyze the user's past subtitle display history and provide the optimal display method. The subtitle generation unit can also learn the user's preferred fonts and colors and reflect them in the subtitle display. The subtitle generation unit can also learn specific display positions and styles from the user's past viewing history and reflect them in the subtitle display. In this way, the optimal display method can be provided by reflecting the user's past subtitle display history. Some or all of the above-described processing in the subtitle generation unit may be performed using, or without, the generation AI. For example, the subtitle generation unit can input the user's subtitle display history data into the generation AI and have the generation AI customize the display method.

[0051] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0052] The subtitle generation system can further include a user gaze tracking function. The gaze tracking function can detect in real time which part of the screen the user is looking at and dynamically adjust the display position of the subtitles. For example, if the user is looking at the top of the screen, the subtitles can be displayed at the bottom of the screen to improve visibility. Alternatively, if the user is looking at the center of the screen, the subtitles can be displayed distributed to the left and right sides of the screen. Furthermore, if the user is focusing on a particular scene, the gaze tracking function can display additional information related to that scene. This allows the display position of the subtitles to be optimized based on the user's gaze, improving the viewing experience.

[0053] The translation unit can also adjust the translation expression taking into account the user's learning history. For example, if a user watches many videos related to a specific field, the translation unit can provide a translation using specialized terminology specific to that field. If the user is a beginner, the translation unit can use concise and easy-to-understand expressions. Furthermore, if the user is proficient in a specific language, the translation unit can provide a natural-sounding translation that reflects the nuances of that language. This allows the translation expression to be optimized based on the user's learning history, making it possible to provide content that is easy to understand.

[0054] The speech recognition unit can further learn the characteristics of the user's voice and generate an individual speech recognition model. For example, if the user has a particular accent or dialect, the unit can learn those characteristics and improve recognition accuracy. Also, if the user has a particular pronunciation habit, the unit can learn those habits and reduce recognition errors. Furthermore, if the user speaks multiple languages, the unit can generate a speech recognition model corresponding to each language and perform optimal recognition for each language. This allows the speech recognition model to be optimized based on the user's voice characteristics and improve recognition accuracy.

[0055] The subtitle generation unit can further customize the subtitle style by taking into account the user's viewing history. For example, if a user prefers a specific font or color, subtitles that reflect that style can be provided. Also, if a user prefers a specific display position, subtitles can be displayed at that position. Furthermore, if a user prefers a specific subtitle display speed, subtitles can be displayed at that speed. This allows the subtitle style to be optimized based on the user's viewing history, improving the viewing experience.

[0056] The translation unit can further adjust the translation expression taking into account the user's cultural background. For example, if the user belongs to a specific cultural sphere, it can use expressions that take that culture into consideration. Also, if the user belongs to a different cultural sphere, it can provide a translation appropriate for that culture. Furthermore, if the user has a multicultural background, it can provide translations that correspond to each culture. This allows the translation expression to be optimized based on the user's cultural background, avoiding misunderstandings.

[0057] The processing flow of the first embodiment will be briefly explained below.

[0058] Step 1: The speech recognition unit converts the video audio into text. The speech recognition unit uses generative AI to analyze the video audio in real time and capture it as text data. The speech recognition unit can also support multiple languages ​​using speech recognition algorithms. For example, it can recognize voice input in English, Japanese, French, etc. and convert it into text data. Furthermore, the speech recognition unit can also use noise canceling technology to remove background sounds and noise to obtain clear audio. For example, it can remove specific frequency bands and emphasize the speaker's voice. Step 2: The translation unit uses generative AI to translate the text converted by the speech recognition unit into the specified language. The translation unit uses a multilingual translation model to translate the acquired text into the specified language. For example, it translates the text using techniques such as neural machine translation (NMT) or statistical machine translation (SMT). In addition, the translation unit can perform context analysis to generate natural-sounding translations. For example, it analyzes the context before and after the video and selects appropriate expressions. Step 3: The subtitle generator generates subtitles based on the text translated by the translation unit. The subtitle generator uses generative AI to display the translated text as subtitles along the video timeline. For example, it adjusts the timing so that the subtitles display the content spoken in a specific scene of the video. In addition, the subtitle generator can customize the font and color of the subtitles to improve visibility. For example, it can change the font and color based on the content of the video.

[0059] (Example 2) A subtitle generation system according to an embodiment of the present invention is a system that translates the audio of a video in real time and automatically generates subtitles. The subtitle generation system analyzes the audio of a video in real time, and a generation AI converts the audio into text. Next, the generation AI translates the converted text into a specified language. Finally, subtitles are automatically generated based on the translated text and displayed on the video. For example, the subtitle generation system analyzes the audio of a video in real time, and a generation AI converts the audio in the video into text using speech recognition technology. Next, the generation AI translates the converted text into a specified language. For example, the generation AI translates the acquired text into a specified language using a multilingual translation model. Finally, the generation AI displays the translated text as subtitles aligned with the video timeline. For example, the generation AI automatically adjusts the timing so that the content spoken in a specific scene of the video is displayed as subtitles. This allows the subtitle generation system to eliminate the need for manual subtitle creation and enable the creation of content that is easily accessible to viewers from different language regions. This allows the subtitle generation system to eliminate the need for manual subtitle creation and enable the creation of content that is easily accessible to viewers from different language regions. For example, by automatically generating subtitles in Japanese, Chinese, etc. for a video produced in English, it is possible to provide content that is easy to understand for viewers who speak different languages.

[0060] A subtitle generation system according to an embodiment includes a speech recognition unit, a translation unit, and a subtitle generation unit. The speech recognition unit converts video audio into text. The speech recognition unit, for example, uses a generation AI to analyze the video audio in real time and acquire the resulting text data. The speech recognition unit can also support multiple languages ​​using a speech recognition algorithm. For example, the speech recognition unit recognizes speech input in multiple languages, such as English, Japanese, and French, and converts the speech input into text data. The speech recognition unit can also use noise canceling technology to remove background sounds and noise and acquire clearer audio. For example, the speech recognition unit can remove specific frequency bands and emphasize the speaker's voice. The translation unit uses a generation AI to translate the text converted by the speech recognition unit into a specified language. The translation unit, for example, uses a multilingual translation model to translate the acquired text into a specified language. For example, the translation unit translates the text using technologies such as neural machine translation (NMT) and statistical machine translation (SMT). The translation unit can also perform context analysis to generate natural-sounding translations. For example, the translation unit analyzes the context of the video and selects appropriate expressions. The subtitle generation unit generates subtitles based on the text translated by the translation unit. The subtitle generation unit, for example, uses a generation AI to display the translated text as subtitles in accordance with the timeline of the video. For example, the subtitle generation unit adjusts the timing so that the content spoken in a specific scene of the video is displayed as subtitles. Furthermore, the subtitle generation unit can customize the font and color of the subtitles to improve visibility. For example, the subtitle generation unit changes the font and color based on the content of the video. As a result, the subtitle generation system according to the embodiment can translate the audio of the video in real time and automatically generate subtitles.

[0061] The speech recognition unit can analyze the audio of a video in real time and convert it into text. For example, the speech recognition unit uses a generation AI to analyze the audio of a video in real time and acquire it as text data. For example, the speech recognition unit analyzes words spoken in a video in real time and acquires them as text data. The speech recognition unit can also support multiple languages ​​using a speech recognition algorithm. For example, the speech recognition unit recognizes voice input in multiple languages, such as English, Japanese, and French, and converts it into text data. Furthermore, the speech recognition unit can use noise canceling technology to remove background sounds and noise and acquire clear audio. For example, the speech recognition unit can remove specific frequency bands and emphasize the speaker's voice. This allows the audio of a video to be converted into text in real time. Some or all of the above-mentioned processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input the audio data of a video into the generation AI and cause the generation AI to generate text data from the audio data.

[0062] The translation unit can translate the acquired text into a specified language using a multilingual translation model. The translation unit translates the acquired text into a specified language using, for example, a generation AI. For example, the translation unit translates the text using technologies such as neural machine translation (NMT) or statistical machine translation (SMT). Furthermore, the translation unit can perform context analysis to generate natural-sounding translations. For example, the translation unit analyzes the context before and after the video and selects appropriate expressions. This allows the acquired text to be translated into multiple languages. Some or all of the above-mentioned processing in the translation unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the translation unit can input the acquired text data into a generation AI and have the generation AI perform translation from the text data.

[0063] The subtitle generation unit can display the translated text as subtitles in accordance with the timeline of the video. The subtitle generation unit, for example, uses a generation AI to display the translated text as subtitles in accordance with the timeline of the video. For example, the subtitle generation unit adjusts the timing so that the content spoken in a specific scene of the video is displayed as subtitles. Furthermore, the subtitle generation unit can customize the font and color of the subtitles to improve visibility. For example, the subtitle generation unit changes the font and color based on the content of the video. This allows the translated text to be displayed as subtitles in accordance with the timeline of the video. Some or all of the above-described processing in the subtitle generation unit may be performed using, or without, the generation AI. For example, the subtitle generation unit can input translated text data to the generation AI and cause the generation AI to generate subtitles.

[0064] The speech recognition unit can estimate the user's emotions and adjust the accuracy of speech recognition based on the estimated user emotions. The speech recognition unit can estimate the user's emotions using, for example, a generation AI and adjust the accuracy of speech recognition based on the estimated user emotions. For example, when the user is nervous, the generation AI can increase the accuracy of speech recognition and reduce misrecognition. When the user is relaxed, the speech recognition unit can maintain the accuracy of speech recognition at normal levels and recognize natural conversations. When the user is in a hurry, the generation AI can prioritize the speed of speech recognition and quickly convert the speech to text. This allows the accuracy of speech recognition to be adjusted based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the speech recognition unit can be performed using, for example, a generation AI, or without a generation AI. For example, the voice recognition unit can input the user's voice data into the generation AI and have the generation AI estimate emotions and adjust the accuracy of voice recognition.

[0065] The speech recognition unit can automatically remove background sounds and noise from a video during speech recognition to make the speech clearer. The speech recognition unit can automatically remove background sounds and noise from a video to make the speech clearer, for example, using a generation AI. For example, the speech recognition unit uses a generation AI to analyze background sounds from a video and filter out noise unnecessary for speech recognition. The speech recognition unit can also use a generation AI to remove specific frequency bands and emphasize the speaker's voice. The speech recognition unit can also use a generation AI to perform noise cancellation in real time to obtain clear speech. This makes it possible to remove background sounds and noise during speech recognition and obtain clear speech. Some or all of the above-described processing in the speech recognition unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the speech recognition unit can input speech data from a video to the generation AI and have the generation AI perform noise removal and speech recognition.

[0066] The speech recognition unit can analyze the characteristics of a speaker's voice during speech recognition and apply a different recognition model to each speaker. The speech recognition unit can, for example, use a generation AI to analyze the characteristics of a speaker's voice during speech recognition and apply a different recognition model to each speaker. For example, the speech recognition unit can have the generation AI analyze the pitch and tone of the speaker's voice and apply an individual recognition model. The speech recognition unit can also have the generation AI learn the speaker's pronunciation habits and improve recognition accuracy. The speech recognition unit can also have the generation AI store the speaker's voice characteristics in a database and use them for subsequent recognition. This allows for applying a different recognition model to each speaker, thereby improving speech recognition accuracy. Some or all of the above-mentioned processing in the speech recognition unit can be performed using, or without, the generation AI. For example, the speech recognition unit can input the speaker's voice data into the generation AI and have the generation AI apply the recognition model for each speaker.

[0067] The speech recognition unit can apply different speech recognition algorithms to each video scene during speech recognition. For example, the speech recognition unit uses a generation AI to apply different speech recognition algorithms to each video scene during speech recognition. For example, the speech recognition unit uses a generation AI to analyze video scenes and select the optimal speech recognition algorithm for each scene. The speech recognition unit can also use the generation AI to perform appropriate noise filtering by taking into account background sounds in the scene. The speech recognition unit can also improve the recognition accuracy of technical terms and proper nouns depending on the content of the scene. This allows the application of the optimal speech recognition algorithm for each scene, thereby improving the accuracy of speech recognition. Some or all of the above-described processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input video scene data to the generation AI and have the generation AI apply the speech recognition algorithm for each scene.

[0068] The speech recognition unit can estimate the user's emotion and adjust the timing of speech recognition based on the estimated user's emotion. The speech recognition unit can estimate the user's emotion using, for example, a generation AI and adjust the timing of speech recognition based on the estimated user's emotion. For example, if the user is nervous, the speech recognition unit can cause the generation AI to delay the timing of speech recognition to reduce recognition errors. Furthermore, if the user is relaxed, the speech recognition unit can cause the generation AI to maintain the timing of speech recognition normally. Furthermore, if the user is in a hurry, the speech recognition unit can cause the generation AI to advance the timing of speech recognition and quickly convert the speech to text. This allows the timing of speech recognition to be adjusted based on the user's emotion. Emotion estimation is achieved using an emotion estimation function, for example, using an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the speech recognition unit can be performed using, for example, the generation AI, or without the generation AI. For example, the voice recognition unit can input the user's voice data into the generation AI and have the generation AI estimate emotions and adjust the timing of voice recognition.

[0069] The speech recognition unit can recognize regional accents and dialects by taking into account the geographical location information of the viewer of the video during speech recognition. The speech recognition unit, for example, uses a generation AI to recognize regional accents and dialects by taking into account the geographical location information of the viewer of the video during speech recognition. For example, the speech recognition unit uses a generation AI to acquire the viewer's geographical location information and recognize the regional accent. The speech recognition unit can also have the generation AI learn the viewer's dialect to improve recognition accuracy. The speech recognition unit can also have the generation AI store regional phrases and expressions in a database and use them for subsequent recognition. This allows the recognition of regional accents and dialects to improve speech recognition accuracy. Some or all of the above-described processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input the viewer's geographical location information to the generation AI and have the generation AI recognize regional accents and dialects.

[0070] The speech recognition unit can prioritize recognition of technical terms and proper nouns based on the content of the video during speech recognition. The speech recognition unit, for example, uses a generation AI to prioritize recognition of technical terms and proper nouns based on the content of the video during speech recognition. For example, the speech recognition unit uses a generation AI to analyze the content of the video and prioritize recognition of technical terms and proper nouns. The speech recognition unit can also improve recognition accuracy by having the generation AI store technical terms from a specific field in a database. The speech recognition unit can also have the generation AI learn the pronunciation of proper nouns to reduce recognition errors. This prioritizes recognition of technical terms and proper nouns, thereby improving speech recognition accuracy. Some or all of the above-described processing in the speech recognition unit may be performed using, or without, the generation AI. For example, the speech recognition unit can input video content data into the generation AI and have the generation AI recognize technical terms and proper nouns.

[0071] The speech recognition unit can improve the recognition accuracy by reflecting the user's past viewing history during speech recognition. The speech recognition unit can improve the recognition accuracy by, for example, using a generation AI to reflect the user's past viewing history during speech recognition. For example, the speech recognition unit can improve the recognition accuracy by having the generation AI analyze the user's past viewing history. The speech recognition unit can also improve the recognition accuracy by having the generation AI learn the terminology of the genre the user frequently watches. The speech recognition unit can also improve the recognition accuracy by having the generation AI learn the vocal characteristics of a specific speaker from the user's viewing history. In this way, the accuracy of speech recognition can be improved by reflecting the user's past viewing history. Some or all of the above-described processing in the speech recognition unit can be performed, for example, using the generation AI, or can be performed without using the generation AI. For example, the speech recognition unit can input the user's viewing history data into the generation AI and have the generation AI analyze the viewing history and improve the recognition accuracy.

[0072] The translation unit can estimate the user's emotions and adjust the translation expression based on the estimated user emotions. The translation unit can estimate the user's emotions using, for example, a generation AI and adjust the translation expression based on the estimated user emotions. For example, if the user is nervous, the translation unit can have the generation AI provide a concise and easy-to-understand translation. Also, if the user is relaxed, the translation unit can have the generation AI provide a detailed and natural translation. Also, if the user is in a hurry, the translation unit can have the generation AI perform a quick translation and provide an expression that gets to the point. This allows the translation expression to be adjusted based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-mentioned processing in the translation unit can be performed using, for example, the generation AI, or without the generation AI. For example, the translation unit can input the user's emotion data into the generation AI and have the generation AI estimate the emotion and adjust the translation expression.

[0073] The translation unit can apply different translation algorithms depending on the genre of the video during translation. The translation unit, for example, uses a generation AI to apply different translation algorithms depending on the genre of the video during translation. For example, the translation unit uses a generation AI to analyze the genre of the video and select the optimal translation algorithm. The translation unit can also use a generation AI to provide casual translations for entertainment videos and formal translations for business videos. The translation unit can also use a generation AI to provide detailed translations including technical terms for educational videos. This allows for applying the optimal translation algorithm depending on the genre of the video, thereby improving the accuracy of the translation. Some or all of the above-mentioned processing in the translation unit may be performed using, or without, the generation AI. For example, the translation unit can input video genre data into the generation AI and have the generation AI apply a translation algorithm for each genre.

[0074] The translation unit can generate a natural-looking translation by taking into account the context of the video during translation. The translation unit can generate a natural-looking translation by using, for example, a generation AI during translation. For example, the translation unit can have the generation AI analyze the context before and after the video to provide a natural-looking translation. The translation unit can also have the generation AI select appropriate expressions according to the context, improving translation accuracy. The translation unit can also have the generation AI perform context analysis to reduce mistranslations. In this way, a natural-looking translation can be generated by taking into account the context of the video. Some or all of the above-mentioned processing in the translation unit can be performed using, for example, the generation AI, or can be performed without using the generation AI. For example, the translation unit can input context data of the video into the generation AI and have the generation AI perform context analysis and generate a natural-looking translation.

[0075] The translation unit can perform context analysis to preserve the speaker's intention and nuances during translation. The translation unit can use, for example, a generation AI to perform context analysis to preserve the speaker's intention and nuances during translation. For example, the translation unit can have the generation AI analyze the speaker's intention and provide a translation that preserves appropriate nuances. The translation unit can also have the generation AI perform context analysis to provide a translation that reflects the speaker's intention. The translation unit can also select appropriate expressions to preserve nuances. This allows for context analysis to preserve the speaker's intention and nuances, thereby providing a more accurate translation. Some or all of the above-mentioned processing in the translation unit can be performed using, for example, the generation AI, or can be performed without using the generation AI. For example, the translation unit can input speaker's intention data into the generation AI and have the generation AI perform context analysis and preserve nuances.

[0076] The translation unit can estimate the user's emotions and adjust the length of the translation based on the estimated user emotions. The translation unit can estimate the user's emotions using, for example, a generation AI and adjust the length of the translation based on the estimated user emotions. For example, if the user is in a hurry, the generation AI can provide a short, to-the-point translation. If the user is relaxed, the translation unit can provide a longer translation with detailed explanations. If the user is excited, the translation unit can provide a translation with visually stimulating effects. This allows the length of the translation to be adjusted based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the translation unit can be performed using, for example, the generation AI, or without the generation AI. For example, the translation unit can input user emotion data into the generation AI and have the generation AI estimate the emotion and adjust the length of the translation.

[0077] The translation unit can select appropriate expressions taking into consideration the cultural background of the viewer of the video when translating. The translation unit, for example, uses a generation AI to select appropriate expressions taking into consideration the cultural background of the viewer of the video when translating. For example, the translation unit uses a generation AI to analyze the viewer's cultural background and select appropriate expressions. The translation unit can also provide a translation that takes the viewer's culture into consideration. The translation unit can also use a generation AI to select expressions that do not lead to misunderstandings based on the cultural background. This makes it possible to select appropriate expressions by taking the viewer's cultural background into consideration. Some or all of the above-mentioned processing in the translation unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the translation unit can input cultural background data of the viewer into the generation AI and cause the generation AI to analyze the cultural background and select appropriate expressions.

[0078] The translation unit can appropriately translate technical terms and proper nouns based on the content of the video during translation. The translation unit, for example, uses a generation AI to appropriately translate technical terms and proper nouns based on the content of the video during translation. For example, the translation unit has the generation AI analyze the content of the video and appropriately translate technical terms and proper nouns. The translation unit can also improve translation accuracy by having the generation AI store technical terms from a specific field in a database. The translation unit can also have the generation AI learn how to translate proper nouns and reduce mistranslations. This can improve translation accuracy by appropriately translating technical terms and proper nouns based on the content of the video. Some or all of the above-mentioned processing in the translation unit may be performed using, or without, the generation AI. For example, the translation unit can input video content data into the generation AI and have the generation AI translate the technical terms and proper nouns.

[0079] The translation unit can improve translation accuracy by reflecting the user's past translation history during translation. The translation unit can improve translation accuracy by using, for example, a generation AI to reflect the user's past translation history during translation. For example, the translation unit can improve translation accuracy by having the generation AI analyze the user's past translation history. The translation unit can also improve translation accuracy by having the generation AI learn expressions frequently used by the user. The translation unit can also improve translation accuracy by having the generation AI learn specific phrases and expressions from the user's translation history. In this way, translation accuracy can be improved by reflecting the user's past translation history. Some or all of the above-mentioned processing in the translation unit can be performed using, or without, the generation AI. For example, the translation unit can input the user's translation history data into the generation AI and have the generation AI analyze the translation history and improve translation accuracy.

[0080] The subtitle generation unit can estimate the user's emotions and adjust the display method of the subtitles based on the estimated user emotions. The subtitle generation unit can estimate the user's emotions using, for example, a generation AI and adjust the display method of the subtitles based on the estimated user emotions. For example, if the user is nervous, the generation AI can provide simple, highly visible subtitles. If the user is relaxed, the generation AI can provide subtitles with detailed information. If the user is in a hurry, the generation AI can provide subtitles that focus on the main points. This allows the display method of the subtitles to be adjusted based on the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-mentioned processing in the subtitle generation unit can be performed using, for example, the generation AI, or without the generation AI. For example, the subtitle generation unit can input user emotion data into the generation AI and cause the generation AI to estimate emotions and adjust the display method of the subtitles.

[0081] The subtitle generation unit can apply different subtitle styles to each video scene when generating subtitles. The subtitle generation unit can apply different subtitle styles to each video scene when generating subtitles, for example, using a generation AI. For example, the subtitle generation unit analyzes video scenes using the generation AI and selects the optimal subtitle style for each scene. The subtitle generation unit can also change the font and color depending on the content of the scene. The subtitle generation unit can also adjust the display method of subtitles to match the atmosphere of the scene. This allows the optimal subtitle style to be applied to each scene, thereby improving visibility. Some or all of the above-mentioned processing in the subtitle generation unit can be performed using, or without, the generation AI. For example, the subtitle generation unit can input video scene data to the generation AI and cause the generation AI to apply a subtitle style for each scene.

[0082] The subtitle generation unit can automatically adjust the display timing of subtitles to match the timeline of the video when generating subtitles. The subtitle generation unit can automatically adjust the display timing of subtitles to match the timeline of the video when generating subtitles, for example, using a generation AI. For example, the subtitle generation unit uses a generation AI to analyze the timeline of the video and set the optimal subtitle display timing. The subtitle generation unit can also adjust the display timing of subtitles to match the start and end of audio. The subtitle generation unit can also adjust the display timing of subtitles to match scene changes. This can improve the viewing experience by adjusting the display timing of subtitles to match the timeline of the video. Some or all of the above-described processing in the subtitle generation unit can be performed using, or without, the generation AI. For example, the subtitle generation unit can input video timeline data to the generation AI and cause the generation AI to adjust the display timing of subtitles.

[0083] The subtitle generation unit can customize the font and color of subtitles based on the content of the video when generating subtitles. The subtitle generation unit, for example, uses a generation AI to customize the font and color of subtitles based on the content of the video when generating subtitles. For example, the subtitle generation unit uses the generation AI to analyze the content of the video and select a font and color that matches the scene. The subtitle generation unit can also set a font and color that is highly visible to match the atmosphere of the video. The subtitle generation unit can also apply different fonts and colors to parts that the generation AI wants to emphasize in a specific scene. In this way, visibility can be improved by customizing the font and color of subtitles based on the content of the video. Some or all of the above-mentioned processing in the subtitle generation unit may be performed using, or without, the generation AI. For example, the subtitle generation unit can input video content data into the generation AI and have the generation AI customize the font and color.

[0084] The subtitle generation unit can estimate a user's emotions and adjust the display position of the subtitles based on the estimated user emotions. The subtitle generation unit can estimate a user's emotions using, for example, a generation AI and adjust the display position of the subtitles based on the estimated user emotions. For example, if the user is nervous, the subtitle generation unit can display the subtitles in the center of the screen to improve visibility. Alternatively, if the user is relaxed, the subtitle generation unit can display the subtitles at the bottom of the screen to avoid interfering with viewing. Alternatively, if the user is in a hurry, the subtitle generation unit can display the subtitles at the top of the screen to quickly provide information. This improves visibility by adjusting the display position of the subtitles based on the user's emotions. Emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the subtitle generation unit can be performed using, for example, a generation AI, or without a generation AI. For example, the subtitle generation unit can input the user's emotion data into the generation AI and have the generation AI estimate the emotion and adjust the display position of the subtitles.

[0085] The subtitle generation unit can select the optimal subtitle display method by taking into account the device information of the viewer of the video when generating subtitles. The subtitle generation unit, for example, uses a generation AI to select the optimal subtitle display method by taking into account the device information of the viewer of the video when generating subtitles. For example, the subtitle generation unit uses the generation AI to acquire the viewer's device information and provide a subtitle display method that matches the screen size. The subtitle generation unit can also set the optimal font size according to the resolution of the viewer's device. The subtitle generation unit can also adjust the display position of the subtitles according to the display mode (portrait or landscape) of the viewer's device. In this way, the optimal subtitle display method can be provided by taking into account the viewer's device information. Some or all of the above-described processing in the subtitle generation unit may be performed using, for example, the generation AI, or may be performed without using the generation AI. For example, the subtitle generation unit can input the viewer's device information into the generation AI and have the generation AI analyze the device information and select the optimal subtitle display method.

[0086] The subtitle generation unit can adjust the length and display time of subtitles based on the content of the video when generating subtitles. The subtitle generation unit adjusts the length and display time of subtitles based on the content of the video when generating subtitles, for example, using a generation AI. For example, the subtitle generation unit analyzes the content of the video and adjusts the length of the subtitles appropriately. The subtitle generation unit can also adjust the display time of subtitles to match the speed of the audio. The subtitle generation unit can also adjust the display time of subtitles to match scene changes. In this way, by adjusting the length and display time of subtitles based on the content of the video, visibility can be improved. Some or all of the above-mentioned processing in the subtitle generation unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the subtitle generation unit can input content data of the video to the generation AI and cause the generation AI to adjust the length and display time of the subtitles.

[0087] The subtitle generation unit can customize the display method by reflecting the user's past subtitle display history when generating subtitles. The subtitle generation unit, for example, uses a generation AI to customize the display method by reflecting the user's past subtitle display history when generating subtitles. For example, the subtitle generation unit uses the generation AI to analyze the user's past subtitle display history and provide the optimal display method. The subtitle generation unit can also learn the user's preferred fonts and colors and reflect them in the subtitle display. The subtitle generation unit can also learn specific display positions and styles from the user's past viewing history and reflect them in the subtitle display. In this way, the optimal display method can be provided by reflecting the user's past subtitle display history. Some or all of the above-described processing in the subtitle generation unit may be performed using, or without, the generation AI. For example, the subtitle generation unit can input the user's subtitle display history data into the generation AI and have the generation AI customize the display method. === Hard Collateral 1-1 === Each of the multiple elements including the above-mentioned speech recognition unit, translation unit, and subtitle generation unit is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the speech recognition unit acquires audio of a video using the camera 42 or microphone 38B of the smart device 14, and converts the audio into text by the specific processing unit 290 of the data processing device 12. The translation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and translates the acquired text into a specified language. The subtitle generation unit is realized, for example, by the control unit 46A of the smart device 14, and displays the translated text as subtitles in accordance with the timeline of the video. === Hard Collateral 1-2 === Each of the multiple elements including the above-mentioned speech recognition unit, translation unit, and subtitle generation unit is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the speech recognition unit acquires audio of a video using the camera 42 or microphone 238 of the smart glasses 214, and converts the audio into text by the specific processing unit 290 of the data processing device 12. The translation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and translates the acquired text into a specified language. The subtitle generation unit is realized, for example, by the control unit 46A of the smart glasses 214, and displays the translated text as subtitles in accordance with the timeline of the video. === Hard Collateral 1-3 === Each of the multiple elements including the above-mentioned speech recognition unit, translation unit, and subtitle generation unit is realized, for example, by at least one of the headset type terminal 314 and the data processing device 12. For example, the speech recognition unit acquires the audio of a video using the camera 42 or microphone 238 of the headset type terminal 314, and converts the audio into text by the specific processing unit 290 of the data processing device 12. The translation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and translates the acquired text into a specified language. The subtitle generation unit is realized, for example, by the control unit 46A of the headset type terminal 314, and displays the translated text as subtitles in accordance with the timeline of the video. === Hard Collateral 1-4 === Each of the multiple elements including the above-mentioned speech recognition unit, translation unit, and subtitle generation unit is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the speech recognition unit acquires audio from a video using the camera 42 or microphone 238 of the robot 414, and converts the audio into text by the specific processing unit 290 of the data processing device 12. The translation unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and translates the acquired text into a specified language. The subtitle generation unit is realized, for example, by the control unit 46A of the robot 414, and displays the translated text as subtitles in accordance with the timeline of the video.

[0088] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0089] The subtitle generation system can further include a user gaze tracking function. The gaze tracking function can detect in real time which part of the screen the user is looking at and dynamically adjust the display position of the subtitles. For example, if the user is looking at the top of the screen, the subtitles can be displayed at the bottom of the screen to improve visibility. Alternatively, if the user is looking at the center of the screen, the subtitles can be displayed distributed to the left and right sides of the screen. Furthermore, if the user is focusing on a particular scene, the gaze tracking function can display additional information related to that scene. This allows the display position of the subtitles to be optimized based on the user's gaze, improving the viewing experience.

[0090] The speech recognition unit can further estimate the speaker's emotions and dynamically change the speech recognition algorithm based on the estimated emotions. For example, if the speaker is excited, the speech recognition unit can analyze the speech at high speed and quickly convert it into text. On the other hand, if the speaker is calm, the speech recognition unit can prioritize accuracy and perform detailed analysis. Furthermore, if the speaker is nervous, the speech recognition unit can strengthen noise cancellation to obtain clearer speech. This allows the speech recognition algorithm to be optimized according to the speaker's emotions and improve recognition accuracy.

[0091] The translation unit can also adjust the translation expression taking into account the user's learning history. For example, if a user watches many videos related to a specific field, the translation unit can provide a translation using specialized terminology specific to that field. If the user is a beginner, the translation unit can use concise and easy-to-understand expressions. Furthermore, if the user is proficient in a specific language, the translation unit can provide a natural-sounding translation that reflects the nuances of that language. This allows the translation expression to be optimized based on the user's learning history, making it possible to provide content that is easy to understand.

[0092] The subtitle generation unit can further estimate the user's emotions and adjust the subtitle display speed based on the estimated emotions. For example, if the user is in a hurry, the subtitle display speed can be increased to provide information quickly. Alternatively, if the user is relaxed, the subtitle display speed can be kept normal to provide information at a natural pace. Furthermore, if the user is concentrating, the subtitle display speed can be decreased to provide more detailed information. This allows the subtitle display speed to be optimized based on the user's emotions, improving the viewing experience.

[0093] The speech recognition unit can further learn the characteristics of the user's voice and generate an individual speech recognition model. For example, if the user has a particular accent or dialect, the unit can learn those characteristics and improve recognition accuracy. Also, if the user has a particular pronunciation habit, the unit can learn those habits and reduce recognition errors. Furthermore, if the user speaks multiple languages, the unit can generate a speech recognition model corresponding to each language and perform optimal recognition for each language. This allows the speech recognition model to be optimized based on the user's voice characteristics and improve recognition accuracy.

[0094] The translation unit can further estimate the user's emotions and adjust the tone of the translation based on the estimated emotions. For example, if the user is sad, the generation AI can provide a translation with a gentle tone. If the user is happy, the generation AI can provide a translation with a bright tone. If the user is angry, the generation AI can provide a translation with a calm and collected tone. This allows the tone of the translation to be adjusted based on the user's emotions, enabling more appropriate communication.

[0095] The subtitle generation unit can further customize the subtitle style by taking into account the user's viewing history. For example, if a user prefers a specific font or color, subtitles that reflect that style can be provided. Also, if a user prefers a specific display position, subtitles can be displayed at that position. Furthermore, if a user prefers a specific subtitle display speed, subtitles can be displayed at that speed. This allows the subtitle style to be optimized based on the user's viewing history, improving the viewing experience.

[0096] The speech recognition unit can further estimate the user's emotions and provide speech recognition feedback based on the estimated emotions. For example, if the user is nervous, the generation AI can display an encouraging message to help them relax. Alternatively, if the user is relaxed, the generation AI can provide normal feedback. Alternatively, if the user is in a hurry, the generation AI can provide quick feedback to help them work efficiently. This allows the speech recognition feedback to be optimized based on the user's emotions, improving user satisfaction.

[0097] The translation unit can further adjust the translation expression taking into account the user's cultural background. For example, if the user belongs to a specific cultural sphere, it can use expressions that take that culture into consideration. Also, if the user belongs to a different cultural sphere, it can provide a translation appropriate for that culture. Furthermore, if the user has a multicultural background, it can provide translations that correspond to each culture. This allows the translation expression to be optimized based on the user's cultural background, avoiding misunderstandings.

[0098] The subtitle generation unit can further estimate the user's emotions and adjust the font size of the subtitles based on the estimated emotions. For example, if the user is nervous, the generation AI can provide subtitles with a large font size to improve visibility. If the user is relaxed, the generation AI can provide subtitles with a normal font size. If the user is in a hurry, the generation AI can provide subtitles with a small font size to provide information quickly. This allows the subtitle font size to be optimized based on the user's emotions, improving the viewing experience.

[0099] The processing flow of the second embodiment will be briefly explained below.

[0100] Step 1: The speech recognition unit converts the video audio into text. The speech recognition unit uses generative AI to analyze the video audio in real time and capture it as text data. The speech recognition unit can also support multiple languages ​​using speech recognition algorithms. For example, it can recognize voice input in English, Japanese, French, etc. and convert it into text data. Furthermore, the speech recognition unit can also use noise canceling technology to remove background sounds and noise to obtain clear audio. For example, it can remove specific frequency bands and emphasize the speaker's voice. Step 2: The translation unit uses generative AI to translate the text converted by the speech recognition unit into the specified language. The translation unit uses a multilingual translation model to translate the acquired text into the specified language. For example, it translates the text using techniques such as neural machine translation (NMT) or statistical machine translation (SMT). In addition, the translation unit can perform context analysis to generate natural-sounding translations. For example, it analyzes the context before and after the video and selects appropriate expressions. Step 3: The subtitle generator generates subtitles based on the text translated by the translation unit. The subtitle generator uses generative AI to display the translated text as subtitles along the video timeline. For example, it adjusts the timing so that the subtitles display the content spoken in a specific scene of the video. In addition, the subtitle generator can customize the font and color of the subtitles to improve visibility. For example, it can change the font and color based on the content of the video.

[0101] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0102] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0103] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0104] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0105] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0106] 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0107] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0108] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0109] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0110] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0111] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0112] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0113] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0114] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0115] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0116] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0117] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0118] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0119] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0120] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0121] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0122] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0123] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0124] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0125] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0126] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0127] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0128] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0129] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0130] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0131] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0132] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0133] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0134] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0135] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0136] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0137] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0138] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0139] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0140] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0141] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0142] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0143] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0144] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0145] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0146] The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0147] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0148] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0149] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0150] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0151] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AI other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0152] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0153] The correspondence between each part and the device or control part is not limited to the above example, and various modifications are possible.

[0154] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0155] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion encompasses both emotions and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0156] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0157] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[0158] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0159] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[0160] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[0161] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0162] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0163] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0164] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0165] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. A processor also includes a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0166] The hardware resource that executes the specific process may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0167] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0168] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0169] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0170] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0171] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0172] [Explanation of symbols]

[0173] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. A speech recognition unit that converts the audio of the video into text; a translation unit that translates the text converted by the speech recognition unit into a specified language; a subtitle generation unit that generates subtitles based on the text translated by the translation unit; Equipped with A system characterized by:

2. The voice recognition unit Analyzes video audio in real time and converts it into text 2. The system of claim 1.

3. The translation unit Translates the retrieved text into the specified language using a multilingual translation model.

2. The system of claim 1.

4. The subtitle generation unit Display translated text as subtitles aligned with the video timeline 2. The system of claim 1.

5. The voice recognition unit Estimate the user's emotions and adjust the accuracy of speech recognition based on the estimated user emotions.

2. The system of claim 1.

6. The voice recognition unit During speech recognition, background sounds and noise from videos are automatically removed to make the audio clearer.

2. The system of claim 1.

7. The voice recognition unit During speech recognition, the speaker's voice characteristics are analyzed and a different recognition model is applied to each speaker.

2. The system of claim 1.

8. The voice recognition unit When recognizing speech, different speech recognition algorithms are applied to each scene in the video.

2. The system of claim 1.

Images