System

Abstract

An object of a system according to an embodiment is to convert audio of a television program into text in real time, convert the text into audio, and provide the audio.SOLUTION: A system according to an embodiment includes a voice acquisition unit, a text generation unit, a voice generation unit, and a subtitle providing unit. The sound acquisition unit acquires sound of a television program. The text generator generates a text from the voice acquired by the voice acquirer in real time. The speech generation unit converts the text generated by the text generation unit into speech. The caption providing unit provides the sound generated by the sound generating unit as a caption.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 technologies have not adequately converted television program audio into text in real time and then converted that text into audio for provision, and there is room for improvement.

[0005] The system according to the embodiment aims to convert the audio of a television program into text in real time, and then convert the text into audio for provision. [Means for solving the problem]

[0006] The system according to the embodiment includes an audio acquisition unit, a text conversion unit, an audio generation unit, and a subtitle provision unit. The audio acquisition unit acquires audio from a television program. The text conversion unit converts the audio acquired by the audio acquisition unit into text in real time. The audio generation unit converts the text converted into text by the text conversion unit into audio. The subtitle provision unit provides the audio generated by the audio generation unit as subtitles. [Effects of the Invention]

[0007] The system according to the embodiment can convert the audio of a television program into text in real time, and convert the text into audio for provision. [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) The TV auxiliary system according to the embodiment of the present invention is a system that provides a combination of a voice-to-text service and a text-to-voice service, thereby making television broadcasts more accessible and convenient.

[0029] The TV auxiliary system according to the embodiment includes an audio acquisition unit, a text conversion unit, an audio generation unit, and a subtitle provision unit. The audio acquisition unit acquires audio from a television program. For example, the audio acquisition unit acquires audio directly from the audio output of a television. The audio acquisition unit can also acquire streaming audio via the Internet. The audio acquisition unit can also acquire ambient audio using a microphone. For example, the audio acquisition unit acquires audio through a cable connected to the audio output terminal of a television. Audio acquisition via the Internet is achieved by downloading audio data from a streaming service. Audio acquisition using a microphone is achieved by using the microphone to pick up audio output from the television speakers. The text conversion unit converts the audio acquired by the audio acquisition unit into text in real time. For example, the text conversion unit converts audio into text using a generation AI (e.g., a generation AI using voice recognition technology). The text conversion unit can also analyze audio data using the generation AI and convert it into text data. The text conversion unit can also summarize the content of audio data and convert it into text using the generation AI. For example, the generation AI analyzes audio data using voice recognition technology and converts it into text data. The generation AI summarizes the content of the audio data, extracts important information, and converts it into text. The audio generation unit converts the text converted into text by the text conversion unit into audio. For example, the audio generation unit converts text into audio using a generation AI (e.g., a generation AI using text-to-speech technology). The audio generation unit can also analyze text data using a generation AI to generate natural audio. The audio generation unit can also generate audio according to the content of the text data using a generation AI. For example, the generation AI analyzes text data using text-to-speech technology to generate natural audio. The generation AI adjusts the tone and speed of the audio according to the content of the text data to generate more natural audio. The subtitle provision unit provides the audio generated by the audio generation unit as subtitles. For example, the subtitle provision unit displays the audio as subtitles using a generation AI (e.g., a generation AI using subtitle generation technology).The subtitle providing unit can also analyze audio data using a generation AI and generate subtitle data. The subtitle providing unit can also generate subtitles according to the content of the audio data using the generation AI. For example, the generation AI analyzes audio data using subtitle generation technology and generates subtitle data. The generation AI adjusts the display timing and format of the subtitles according to the content of the audio data to generate subtitles that are more visually understandable. This allows the TV auxiliary system according to the embodiment to convert the audio of a television program into text in real time, convert it into audio, and provide it as subtitles. For example, this can accommodate people with hearing impairments and people who have difficulty reading Japanese subtitles. Providing audio subtitles in real time improves convenience for viewers.

[0030] The text conversion unit can use the generative AI to create a custom model that learns the voice quality and accent of a specific speaker. For example, the text conversion unit provides the generative AI with a large amount of audio data to create a custom model to learn the voice quality and accent of a specific speaker. For example, the text conversion unit can train the AI ​​with audio data from a news anchor to improve the speech recognition accuracy of news programs. When creating a custom model, the text conversion unit also learns not only the speaker's voice quality and accent, but also the speaker's speaking style and phrasing. For example, the text conversion unit can train the AI ​​with audio data including a specific regional dialect or slang. The text conversion unit also regularly updates the custom model to accommodate new speakers and accents. For example, when a new news anchor appears, the text conversion unit adds that anchor's audio data to update the model. This improves the accuracy of speech recognition by learning the voice quality and accent of a specific speaker.

[0031] The text conversion unit can add a filtering function that automatically removes background sounds and noise from audio data and converts only clear audio into text. For example, the text conversion unit incorporates a noise filtering algorithm into the generation AI to automatically remove background sounds and noise from audio data. For example, it can remove city noise and wind sounds and convert only the speaker's voice into clear text. The text conversion unit also trains the generation AI on different types of noise data to enhance its filtering function. For example, it can train the AI ​​on various environmental sounds, such as cafe noise and car engine sounds. The text conversion unit also has the generation AI perform real-time noise filtering to improve the accuracy of noise removal. For example, it can instantly remove sudden noises that occur during live broadcasts. This improves the accuracy of voice recognition by removing background sounds and noise.

[0032] The text conversion unit can simultaneously translate audio into multiple languages ​​when converting audio to text, thereby providing multilingual subtitles. For example, when converting audio to text, the text conversion unit adds the ability to simultaneously translate into multiple languages ​​to the generation AI. For example, English audio can be simultaneously translated into Japanese, Chinese, and Spanish to provide multilingual subtitles. The text conversion unit also trains the generation AI to learn technical terms and slang in each language to improve translation accuracy. For example, it trains the AI ​​on audio data containing medical and technical terms. The text conversion unit also has the generation AI perform translation in real time to provide multilingual subtitles. For example, it can simultaneously translate into multiple languages ​​during a live broadcast and provide them to viewers. This makes it possible to provide multilingual subtitles by translating into multiple languages ​​simultaneously.

[0033] The text conversion unit can add a function to analyze audio data and highlight specific keywords and phrases. For example, the text conversion unit incorporates a keyword extraction algorithm into the generation AI to analyze audio data and highlight specific keywords and phrases. For example, this is done to highlight important information in a news program. The text conversion unit also adds visual effects to the generation AI when highlighting keywords and phrases. For example, important keywords may be displayed in bold or color. The text conversion unit also trains the generation AI on large amounts of text data to improve the accuracy of keyword extraction. For example, it trains the generation AI on news articles and scripts to extract important keywords. This makes it easier to convey important information to viewers by highlighting specific keywords and phrases.

[0034] The text conversion unit allows the generation AI to learn from past broadcast data and create a model specialized for a specific program and genre in order to improve real-time speech recognition accuracy. For example, the text conversion unit has the generation AI learn from past broadcast data in order to improve real-time speech recognition accuracy. For example, the text conversion unit has the generation AI learn from audio data of news programs or dramas to create a model specialized for a specific program. In addition, when creating a model specialized for a specific program or genre, the text conversion unit has the generation AI learn the characteristics of the program and the voice quality of the speaker. For example, the text conversion unit has the generation AI learn from the live audio of a sports program. In addition, in order to improve the accuracy of the model, the text conversion unit periodically adds new broadcast data to the generation AI and has it learn from it. For example, data from a new season of a drama or news program is added. In this way, by learning from past broadcast data, a speech recognition model specialized for a specific program or genre can be created.

[0035] The text conversion unit can introduce an algorithm that optimizes the processing speed of the generation AI to minimize delays in speech recognition. For example, the text conversion unit introduces a high-speed processing algorithm into the generation AI to minimize delays in speech recognition. For example, it analyzes voice data in real time and converts it into text instantly. The text conversion unit also incorporates parallel processing technology into the generation AI to optimize processing speed. For example, it analyzes and converts multiple pieces of voice data into text simultaneously. The text conversion unit also uses hardware acceleration in the generation AI to minimize delays. For example, it uses a GPU or dedicated chip to speed up speech recognition processing. This minimizes delays in speech recognition and enables real-time speech recognition.

[0036] The text conversion unit can add a function to display the text converted in real time in a visually easy-to-understand format. For example, the text conversion unit adds a chart or graph generation function to the generation AI to display the text converted in real time in a visually easy-to-understand format. For example, it displays important points of news in a graph. The text conversion unit also incorporates a data summary or highlight function into the generation AI when visually displaying the text converted data. For example, it highlights important keywords. The text conversion unit also causes the generation AI to generate interactive graphs and charts to enhance the visual display. For example, it allows users to access detailed information by clicking or zooming. This makes it easier to convey information to viewers by displaying it in a visually easy-to-understand format.

[0037] The text conversion unit can add a function to synchronize the transcribed data with other devices (smartphones and tablets) in real time. For example, the text conversion unit incorporates a data synchronization algorithm into the generation AI to synchronize the transcribed data with other devices in real time. For example, the text data is instantly sent to smartphones and tablets. The text conversion unit also introduces network optimization technology into the generation AI to improve the accuracy of synchronization between devices. For example, it uses a protocol for transmitting data with low latency. The text conversion unit also uses cloud-based data storage in the generation AI to achieve real-time data synchronization. For example, it stores text data on a cloud server and makes it accessible from each device. This allows viewers to view information on multiple devices by synchronizing the transcribed data with other devices in real time.

[0038] The speech generation unit can add a function that allows the user to customize the speaker's voice quality and accent when converting text to speech. For example, the speech generation unit adds a function that allows the user to select the speaker's voice quality and accent when converting text to speech. For example, it can allow the user to select a male voice, a female voice, or an accent from a specific region. In addition, to enhance the customization function, the speech generation unit trains the generation AI with data on multiple voice qualities and accents. For example, it can train voices of various ages and genders. In addition, the speech generation unit adds a function that allows the user to record their own voice and generate speech based on that voice quality and accent. For example, it can train the user's voice and convert text into speech using that voice. In this way, the speaker's voice quality and accent can be customized to provide speech that suits the user's preferences.

[0039] The speech generation unit can incorporate an algorithm that automatically adjusts the tone and speed of the speech according to the content of the text. For example, the speech generation unit incorporates an algorithm into the generation AI that automatically adjusts the tone and speed of the speech according to the content of the text. For example, the tone may be emphasized and the speed may be slowed down when conveying important information. The speech generation unit also has the generation AI analyze the context and emotion of the text in order to adjust the tone and speed of the speech. For example, the tone may be raised and the speed increased in emotionally charged parts. The speech generation unit also adds a function that allows the user to customize the tone and speed of the speech. For example, the user can set their preferred tone and speed and generate speech based on that. This makes it possible to provide a more natural speech by adjusting the tone and speed of the speech according to the content of the text.

[0040] The speech generation unit can incorporate a function to automatically add background music and sound effects when converting text to speech. For example, the speech generation unit incorporates a function to automatically add background music and sound effects into the generation AI when converting text to speech. For example, music suitable for the background of news is added. The speech generation unit also has the generation AI analyze the content and context of the text to select background music and sound effects. For example, moving music is added for moving scenes. The speech generation unit also incorporates a function that allows the user to customize background music and sound effects. For example, the user can select their favorite music and sound effects and generate speech based on them. This allows for a richer audio experience by adding background music and sound effects.

[0041] The speech generation unit can simulate the voices of multiple speakers when converting text to speech and generate conversational speech. For example, the speech generation unit adds a function to the generation AI to simulate the voices of multiple speakers when converting text to speech. For example, it converts conversational text into speech using multiple voices. In addition, the speech generation unit trains the generation AI with data on various voice qualities and accents to simulate the voices of multiple speakers. For example, it trains voices of different ages and genders. In addition, the speech generation unit adds a function that allows the user to customize the speaker voice. For example, the user can select their favorite voice and generate conversational speech based on that. In this way, conversational speech can be provided by simulating the voices of multiple speakers.

[0042] The speech generation unit can add a function to automatically adjust the intonation and emphasis of the speech according to the content of the text data. For example, the speech generation unit adds a function to the generation AI that automatically adjusts the intonation and emphasis of the speech according to the content of the text data. For example, the intonation can be emphasized when conveying important information. The speech generation unit also has the generation AI analyze the context and emotions of the text to adjust the intonation and emphasis of the speech. For example, the intonation can be raised in emotionally charged parts. The speech generation unit also adds a function that allows the user to customize the intonation and emphasis of the speech. For example, the user can set their preferred intonation and emphasis and generate speech based on that. This makes it possible to provide a more natural speech by adjusting the intonation and emphasis of the speech according to the content of the text data.

[0043] The speech generation unit can analyze text data and generate speech that emphasizes specific keywords and phrases. For example, the speech generation unit analyzes text data and incorporates a keyword extraction algorithm into the generation AI to emphasize specific keywords and phrases. For example, it generates speech that emphasizes important points of news. The speech generation unit also adds visual effects to the generation AI when emphasizing keywords and phrases. For example, it displays important keywords in bold or color. The speech generation unit also trains the generation AI on large amounts of text data to improve the accuracy of keyword extraction. For example, it trains the AI ​​on news articles and scripts to extract important keywords. This makes it easier to convey important information to viewers by emphasizing specific keywords and phrases.

[0044] The speech generation unit can add a function to simultaneously generate speech in different languages ​​when converting text data into speech. For example, the speech generation unit adds a function to the generation AI to simultaneously generate speech in different languages ​​when converting text data into speech. For example, English text can be simultaneously converted into speech in Japanese, Chinese, and Spanish. The speech generation unit also trains the generation AI to learn technical terms and slang in each language to improve translation accuracy. For example, it trains the AI ​​on text data containing medical and technical terms. The speech generation unit also has the generation AI perform translation in real time to provide multilingual speech. For example, it can translate into multiple languages ​​simultaneously during a live broadcast and provide the results to viewers. This allows multilingual speech to be provided by simultaneously generating speech in different languages.

[0045] The speech generation unit can incorporate a function that allows customization of the speech speed and pitch when converting text data into speech. For example, the speech generation unit incorporates a function that allows the generation AI to customize the speech speed and pitch when converting text data into speech. For example, a user can set their preferred speed and pitch, and speech is generated based on that. The speech generation unit also has the generation AI analyze the context and emotion of the text to adjust the speech speed and pitch. For example, the speed can be increased and the pitch can be raised in emotionally charged parts. The speech generation unit also adds a function that allows the user to adjust the speech speed and pitch in real time. For example, the user can change the speed and pitch during playback. This allows the speech speed and pitch to be customized, thereby providing speech that meets the user's preferences.

[0046] The subtitle providing unit allows the generation AI to learn past subtitle data and create a model specialized for a specific program and genre in order to improve the accuracy of providing audio subtitles in real time. For example, the subtitle providing unit has the generation AI learn past subtitle data in order to improve the accuracy of providing audio subtitles in real time. For example, the subtitle providing unit has the generation AI learn subtitle data from news programs or dramas to create a model specialized for a specific program. In addition, when creating a model specialized for a specific program or genre, the subtitle providing unit has the generation AI learn the characteristics of the program and the voice quality of the speaker. For example, the subtitle providing unit has the generation AI learn the live audio of a sports program. In addition, in order to improve the accuracy of the model, the subtitle providing unit periodically adds new subtitle data to the generation AI and has it learn from it. For example, data from a new season of a drama or news program is added. In this way, by learning from past subtitle data, an audio subtitle providing model specialized for a specific program or genre can be created.

[0047] The subtitle providing unit can introduce an algorithm that improves the processing speed of the generation AI to optimize the timing of displaying audio subtitles. For example, the subtitle providing unit introduces a high-speed processing algorithm into the generation AI to optimize the timing of displaying audio subtitles. For example, it analyzes audio data in real time and displays subtitles instantly. The subtitle providing unit also incorporates parallel processing technology into the generation AI to optimize the processing speed. For example, it analyzes multiple pieces of audio data simultaneously and generates subtitles. The subtitle providing unit also uses hardware acceleration in the generation AI to optimize the display timing. For example, it uses a GPU or dedicated chip to speed up speech recognition processing. This makes it possible to provide subtitles in real time by optimizing the timing of displaying audio subtitles.

[0048] The subtitle providing unit can add a function to display real-time audio subtitles in a visually easy-to-understand format. For example, the subtitle providing unit adds color coding and font change functions to the generation AI to display real-time audio subtitles in a visually easy-to-understand format. For example, important information may be displayed in red. The subtitle providing unit also adds interactive effects to the generation AI to enhance the visual display of the subtitles. For example, the subtitle providing unit allows the user to access detailed information by clicking or zooming. The subtitle providing unit also adds a function to customize the visual display. For example, the user may select their preferred color or font and display subtitles based on that. This makes it easier to convey information to viewers by displaying it in a visually easy-to-understand format.

[0049] The subtitle provider can add a function to synchronize audio subtitles with other devices (smartphones and tablets) in real time. For example, the subtitle provider incorporates a data synchronization algorithm into the generation AI to synchronize audio subtitles with other devices in real time. For example, the subtitle provider instantly sends subtitle data to smartphones and tablets. The subtitle provider also introduces network optimization technology into the generation AI to improve synchronization accuracy between devices. For example, the subtitle provider uses a protocol for transmitting data with low latency. The subtitle provider also uses cloud-based data storage in the generation AI to achieve real-time data synchronization. For example, the subtitle data is stored on a cloud server and made accessible from each device. This allows viewers to view information on multiple devices by synchronizing audio subtitles with other devices in real time.

[0050] When the system converts speech to text in real time and converts the text to speech to provide subtitles, the generation AI can learn from past data to create a model specialized for a specific program and genre in order to improve the accuracy of real-time speech recognition and speech generation. For example, the system has the generation AI learn from past data to improve the accuracy of real-time speech recognition and speech generation. For example, the system has the generation AI learn from audio data of news programs or dramas to create a model specialized for a specific program. In addition, when creating a model specialized for a specific program or genre, the system has the generation AI learn the characteristics of the program and the voice quality of the speaker. For example, the system has the generation AI learn from the commentary audio of a sports program. In addition, the system periodically adds new data to the generation AI to improve the accuracy of the model. For example, data from a new season of a drama or news program is added. In this way, by learning from past data, the accuracy of speech recognition and speech generation specialized for a specific program or genre is improved.

[0051] The system can introduce an algorithm that optimizes the processing speed of the generative AI to minimize delays in speech recognition and speech generation. For example, the system introduces a high-speed processing algorithm into the generative AI to minimize delays in speech recognition and speech generation. For example, the system analyzes speech data in real time and instantly converts it into text and speech. The system also incorporates parallel processing technology into the generative AI to optimize processing speed. For example, the system simultaneously analyzes multiple pieces of speech data, converts them into text, and converts them into speech. The system also uses hardware acceleration in the generative AI to minimize delays. For example, the system uses a GPU or dedicated chip to speed up speech recognition and speech generation processing. This minimizes delays in speech recognition and speech generation, enabling real-time speech recognition and speech generation.

[0052] The system can add a function to display real-time text data and audio in a visually easy-to-understand format. For example, the system adds a chart or graph generation function to the generation AI to display real-time text data and audio in a visually easy-to-understand format. For example, the system displays important points of news in a graph. The system also incorporates a data summary or highlight function into the generation AI when visually displaying text data and audio. For example, the system highlights important keywords. The system also causes the generation AI to generate interactive graphs and charts to enhance the visual display. For example, the system allows users to access more detailed information by clicking or zooming. This makes it easier to convey information to viewers by displaying it in a visually easy-to-understand format.

[0053] The system can add a function for synchronizing the transcribed data and audio with other devices (smartphones and tablets) in real time. For example, the system incorporates a data synchronization algorithm into the generation AI to synchronize the transcribed data and audio with other devices in real time. For example, the system instantly sends text and audio data to smartphones and tablets. The system also introduces network optimization technology into the generation AI to improve the accuracy of synchronization between devices. For example, the system uses a protocol for transmitting data with low latency. The system also uses cloud-based data storage in the generation AI to achieve real-time data synchronization. For example, the system stores text and audio data on a cloud server and makes it accessible from each device. This allows viewers to view information on multiple devices by synchronizing the transcribed data and audio with other devices in real time.

[0054] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0055] The TV assistant system can also include a recommendation unit that analyzes the user's viewing history and recommends programs based on the viewer's preferences. For example, it can analyze the genre and content of previously viewed programs and recommend similar programs. The recommendation unit can also recommend programs at optimal times, taking into account the viewer's viewing time and frequency. Furthermore, the recommendation unit can collect viewer feedback and use it to improve the recommendation algorithm. This enables program recommendations based on the viewer's preferences, improving the viewing experience.

[0056] The TV auxiliary system can also be equipped with a health management unit that monitors the viewer's health and encourages appropriate breaks. For example, sensors can detect the viewer's heart rate and posture and suggest breaks to reduce the health risks associated with long viewing periods. The health management unit can also automatically adjust the screen brightness and contrast to reduce eye strain. Furthermore, the health management unit can collect the viewer's health data and provide personalized health advice. This allows for a comfortable viewing experience while maintaining the viewer's health.

[0057] The TV auxiliary system can also include a recording unit that analyzes the viewer's viewing history and automatically records programs the viewer missed. For example, it can analyze the genre and content of programs the viewer has previously watched and automatically record programs that the viewer missed. The recording unit can also record programs at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the recording unit can collect viewer feedback and use it to improve the recording algorithm. This improves the viewing experience by automatically recording programs the viewer missed.

[0058] The TV auxiliary system can also include a trailer player that analyzes the viewer's viewing history and automatically plays trailers for programs that the viewer might be interested in. For example, it can analyze the genre and content of programs previously viewed and play trailers for similar programs. The trailer player can also play trailers at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the trailer player can collect viewer feedback and use it to improve the trailer playback algorithm. This improves the viewing experience by automatically playing trailers for programs that the viewer might be interested in.

[0059] The TV auxiliary system can further include an information provider that analyzes the viewer's viewing history and provides related information about programs that may be of interest to the viewer. For example, it can analyze the genre and content of previously viewed programs and provide related news articles and interviews. The information provider can also provide related information at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the information provider can collect viewer feedback and use it to improve the information provision algorithm. This improves the viewing experience by providing related information about programs that may be of interest to the viewer.

[0060] The TV assistance system can further include a feature creation unit that analyzes the viewer's viewing history and automatically creates special features of programs that the viewer may be interested in. For example, it can analyze the genre and content of programs viewed in the past and create special features by collecting similar programs. The feature creation unit can also provide special features at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the feature creation unit can collect viewer feedback and use it to improve the feature creation algorithm. This improves the viewing experience by automatically creating special features of programs that the viewer may be interested in.

[0061] The processing flow of the first embodiment will be briefly explained below.

[0062] Step 1: The audio capture unit captures audio from a television program. For example, the audio capture unit captures audio directly from the audio output of a television. The audio capture unit can also capture streaming audio via the Internet. Furthermore, the audio capture unit can also capture ambient audio using a microphone. Step 2: The text conversion unit converts the speech captured by the speech capture unit into text in real time. For example, the text conversion unit converts speech into text using a generation AI (e.g., a generation AI using speech recognition technology). The text conversion unit can also use the generation AI to analyze speech data and convert it into text data. Furthermore, the text conversion unit can also use the generation AI to summarize the content of the speech data and convert it into text. Step 3: The speech generation unit converts the text converted by the text conversion unit into speech. For example, the speech generation unit converts the text into speech using a generation AI (e.g., a generation AI using text-to-speech technology). The speech generation unit can also analyze the text data using the generation AI and generate natural speech. Furthermore, the speech generation unit can also generate speech according to the content of the text data using the generation AI. Step 4: The subtitle providing unit provides the audio generated by the audio generating unit as subtitles. For example, the subtitle providing unit displays the audio as subtitles using a generation AI (e.g., a generation AI using subtitle generation technology). The subtitle providing unit can also analyze audio data using the generation AI and generate subtitle data. Furthermore, the subtitle providing unit can also generate subtitles according to the content of the audio data using the generation AI.

[0063] (Example 2) The TV auxiliary system according to the embodiment of the present invention is a system that provides a combination of a voice-to-text service and a text-to-voice service, thereby making television broadcasts more accessible and convenient.

[0064] The TV auxiliary system according to the embodiment includes an audio acquisition unit, a text conversion unit, an audio generation unit, and a subtitle provision unit. The audio acquisition unit acquires audio from a television program. For example, the audio acquisition unit acquires audio directly from the audio output of a television. The audio acquisition unit can also acquire streaming audio via the Internet. The audio acquisition unit can also acquire ambient audio using a microphone. For example, the audio acquisition unit acquires audio through a cable connected to the audio output terminal of a television. Audio acquisition via the Internet is achieved by downloading audio data from a streaming service. Audio acquisition using a microphone is achieved by using the microphone to pick up audio output from the television speakers. The text conversion unit converts the audio acquired by the audio acquisition unit into text in real time. For example, the text conversion unit converts audio into text using a generation AI (e.g., a generation AI using voice recognition technology). The text conversion unit can also analyze audio data using the generation AI and convert it into text data. The text conversion unit can also summarize the content of audio data and convert it into text using the generation AI. For example, the generation AI analyzes audio data using voice recognition technology and converts it into text data. The generation AI summarizes the content of the audio data, extracts important information, and converts it into text. The audio generation unit converts the text converted into text by the text conversion unit into audio. For example, the audio generation unit converts text into audio using a generation AI (e.g., a generation AI using text-to-speech technology). The audio generation unit can also analyze text data using a generation AI to generate natural audio. The audio generation unit can also generate audio according to the content of the text data using a generation AI. For example, the generation AI analyzes text data using text-to-speech technology to generate natural audio. The generation AI adjusts the tone and speed of the audio according to the content of the text data to generate more natural audio. The subtitle provision unit provides the audio generated by the audio generation unit as subtitles. For example, the subtitle provision unit displays the audio as subtitles using a generation AI (e.g., a generation AI using subtitle generation technology).The subtitle providing unit can also analyze audio data using a generation AI and generate subtitle data. The subtitle providing unit can also generate subtitles according to the content of the audio data using the generation AI. For example, the generation AI analyzes audio data using subtitle generation technology and generates subtitle data. The generation AI adjusts the display timing and format of the subtitles according to the content of the audio data to generate subtitles that are more visually understandable. This allows the TV auxiliary system according to the embodiment to convert the audio of a television program into text in real time, convert it into audio, and provide it as subtitles. For example, this can accommodate people with hearing impairments and people who have difficulty reading Japanese subtitles. Providing audio subtitles in real time improves convenience for viewers.

[0065] The text conversion unit can use the generative AI to create a custom model that learns the voice quality and accent of a specific speaker. For example, the text conversion unit provides the generative AI with a large amount of audio data to create a custom model to learn the voice quality and accent of a specific speaker. For example, the text conversion unit can train the AI ​​with audio data from a news anchor to improve the speech recognition accuracy of news programs. When creating a custom model, the text conversion unit also learns not only the speaker's voice quality and accent, but also the speaker's speaking style and phrasing. For example, the text conversion unit can train the AI ​​with audio data including a specific regional dialect or slang. The text conversion unit also regularly updates the custom model to accommodate new speakers and accents. For example, when a new news anchor appears, the text conversion unit adds that anchor's audio data to update the model. This improves the accuracy of speech recognition by learning the voice quality and accent of a specific speaker.

[0066] The text conversion unit can add a filtering function that automatically removes background sounds and noise from audio data and converts only clear audio into text. For example, the text conversion unit incorporates a noise filtering algorithm into the generation AI to automatically remove background sounds and noise from audio data. For example, it can remove city noise and wind sounds and convert only the speaker's voice into clear text. The text conversion unit also trains the generation AI on different types of noise data to enhance its filtering function. For example, it can train the AI ​​on various environmental sounds, such as cafe noise and car engine sounds. The text conversion unit also has the generation AI perform real-time noise filtering to improve the accuracy of noise removal. For example, it can instantly remove sudden noises that occur during live broadcasts. This improves the accuracy of voice recognition by removing background sounds and noise.

[0067] The text conversion unit can use the emotion estimation function to estimate the speaker's emotion and adjust the text expression based on that emotion. For example, the text conversion unit uses the emotion estimation function to analyze the speaker's emotion in real time and adjust the text expression based on that emotion. For example, if the speaker is excited, an exclamation mark is added. The text conversion unit also trains the generation AI with audio data and emotion labels to estimate the speaker's emotion. For example, it trains audio data containing emotions such as joy, sadness, and anger. The text conversion unit also analyzes features of the voice, such as tone, pitch, and speed, to enhance the emotion estimation function. For example, if the speaker's voice is high-pitched and fast, it is estimated that the speaker is excited. This allows the text expression to be adjusted based on the speaker's emotion, thereby providing more natural-sounding subtitles.

[0068] The text conversion unit can simultaneously translate audio into multiple languages ​​when converting audio to text, thereby providing multilingual subtitles. For example, when converting audio to text, the text conversion unit adds the ability to simultaneously translate into multiple languages ​​to the generation AI. For example, English audio can be simultaneously translated into Japanese, Chinese, and Spanish to provide multilingual subtitles. The text conversion unit also trains the generation AI to learn technical terms and slang in each language to improve translation accuracy. For example, it trains the AI ​​on audio data containing medical and technical terms. The text conversion unit also has the generation AI perform translation in real time to provide multilingual subtitles. For example, it can simultaneously translate into multiple languages ​​during a live broadcast and provide them to viewers. This makes it possible to provide multilingual subtitles by translating into multiple languages ​​simultaneously.

[0069] The text conversion unit can add a function to analyze audio data and highlight specific keywords and phrases. For example, the text conversion unit incorporates a keyword extraction algorithm into the generation AI to analyze audio data and highlight specific keywords and phrases. For example, this is done to highlight important information in a news program. The text conversion unit also adds visual effects to the generation AI when highlighting keywords and phrases. For example, important keywords may be displayed in bold or color. The text conversion unit also trains the generation AI on large amounts of text data to improve the accuracy of keyword extraction. For example, it trains the generation AI on news articles and scripts to extract important keywords. This makes it easier to convey important information to viewers by highlighting specific keywords and phrases.

[0070] The text conversion unit can use the emotion estimation function to highlight parts that are likely to interest the user, improving the viewing experience. For example, the text conversion unit incorporates an emotion analysis algorithm into the generation AI to use the emotion estimation function to highlight parts that are likely to interest the user. For example, it highlights parts where the speaker is excited. The text conversion unit also adds visual effects to the generation AI to attract the user's interest. For example, it displays interesting parts in color or with animation. The text conversion unit also has the generation AI learn the user's viewing history and feedback to strengthen the emotion estimation function. For example, it highlights parts based on programs viewed in the past or topics that interested the user. This improves the viewing experience by highlighting parts that are likely to interest the user.

[0071] The text conversion unit allows the generation AI to learn from past broadcast data and create a model specialized for a specific program and genre in order to improve real-time speech recognition accuracy. For example, the text conversion unit has the generation AI learn from past broadcast data in order to improve real-time speech recognition accuracy. For example, the text conversion unit has the generation AI learn from audio data of news programs or dramas to create a model specialized for a specific program. In addition, when creating a model specialized for a specific program or genre, the text conversion unit has the generation AI learn the characteristics of the program and the voice quality of the speaker. For example, the text conversion unit has the generation AI learn from the live audio of a sports program. In addition, in order to improve the accuracy of the model, the text conversion unit periodically adds new broadcast data to the generation AI and has it learn from it. For example, data from a new season of a drama or news program is added. In this way, by learning from past broadcast data, a speech recognition model specialized for a specific program or genre can be created.

[0072] The text conversion unit can introduce an algorithm that optimizes the processing speed of the generation AI to minimize delays in speech recognition. For example, the text conversion unit introduces a high-speed processing algorithm into the generation AI to minimize delays in speech recognition. For example, it analyzes voice data in real time and converts it into text instantly. The text conversion unit also incorporates parallel processing technology into the generation AI to optimize processing speed. For example, it analyzes and converts multiple pieces of voice data into text simultaneously. The text conversion unit also uses hardware acceleration in the generation AI to minimize delays. For example, it uses a GPU or dedicated chip to speed up speech recognition processing. This minimizes delays in speech recognition and enables real-time speech recognition.

[0073] The text conversion unit can use an emotion estimation function to analyze a speaker's emotion in real time and adjust the tone of the text based on that emotion. For example, the text conversion unit uses the emotion estimation function to analyze a speaker's emotion in real time and adjust the tone of the text based on that emotion. For example, if the speaker is angry, an emphasized expression is used. To analyze emotions in real time, the text conversion unit also has the generation AI analyze features such as tone, pitch, and speed of the voice. For example, if the speaker's voice is high-pitched and fast, it is estimated that the speaker is excited. To enhance the emotion estimation function, the text conversion unit also has the generation AI train a large amount of emotion-labeled audio data. For example, it trains audio data containing emotions such as joy, sadness, and anger. This allows the text tone to be adjusted based on the speaker's emotion, thereby providing more natural-looking subtitles.

[0074] The text conversion unit can add a function to display the text converted in real time in a visually easy-to-understand format. For example, the text conversion unit adds a chart or graph generation function to the generation AI to display the text converted in real time in a visually easy-to-understand format. For example, it displays important points of news in a graph. The text conversion unit also incorporates a data summary or highlight function into the generation AI when visually displaying the text converted data. For example, it highlights important keywords. The text conversion unit also causes the generation AI to generate interactive graphs and charts to enhance the visual display. For example, it allows users to access detailed information by clicking or zooming. This makes it easier to convey information to viewers by displaying it in a visually easy-to-understand format.

[0075] The text conversion unit can add a function to synchronize the transcribed data with other devices (smartphones and tablets) in real time. For example, the text conversion unit incorporates a data synchronization algorithm into the generation AI to synchronize the transcribed data with other devices in real time. For example, the text data is instantly sent to smartphones and tablets. The text conversion unit also introduces network optimization technology into the generation AI to improve the accuracy of synchronization between devices. For example, it uses a protocol for transmitting data with low latency. The text conversion unit also uses cloud-based data storage in the generation AI to achieve real-time data synchronization. For example, it stores text data on a cloud server and makes it accessible from each device. This allows viewers to view information on multiple devices by synchronizing the transcribed data with other devices in real time.

[0076] The text conversion unit can use the emotion estimation function to provide real-time feedback according to the viewer's emotions and personalize the viewing experience. For example, the text conversion unit incorporates an emotion analysis algorithm into the generation AI to provide real-time feedback according to the viewer's emotions using the emotion estimation function. For example, if the viewer is excited, related content is suggested. The text conversion unit also has the generation AI learn viewer emotional data to personalize the viewing experience. For example, it predicts emotional reactions based on past viewing history and feedback. The text conversion unit also adds interactive functions to the generation AI to provide real-time feedback. For example, it allows the viewer to select content according to their emotions. This makes it possible to personalize the viewing experience by providing feedback according to the viewer's emotions.

[0077] The speech generation unit can add a function that allows the user to customize the speaker's voice quality and accent when converting text to speech. For example, the speech generation unit adds a function that allows the user to select the speaker's voice quality and accent when converting text to speech. For example, it can allow the user to select a male voice, a female voice, or an accent from a specific region. In addition, to enhance the customization function, the speech generation unit trains the generation AI with data on multiple voice qualities and accents. For example, it can train voices of various ages and genders. In addition, the speech generation unit adds a function that allows the user to record their own voice and generate speech based on that voice quality and accent. For example, it can train the user's voice and convert text into speech using that voice. In this way, the speaker's voice quality and accent can be customized to provide speech that suits the user's preferences.

[0078] The speech generation unit can incorporate an algorithm that automatically adjusts the tone and speed of the speech according to the content of the text. For example, the speech generation unit incorporates an algorithm into the generation AI that automatically adjusts the tone and speed of the speech according to the content of the text. For example, the tone may be emphasized and the speed may be slowed down when conveying important information. The speech generation unit also has the generation AI analyze the context and emotion of the text in order to adjust the tone and speed of the speech. For example, the tone may be raised and the speed increased in emotionally charged parts. The speech generation unit also adds a function that allows the user to customize the tone and speed of the speech. For example, the user can set their preferred tone and speed and generate speech based on that. This makes it possible to provide a more natural speech by adjusting the tone and speed of the speech according to the content of the text.

[0079] The speech generation unit can use the emotion estimation function to adjust the speech expression based on the emotion of the text. For example, the speech generation unit incorporates an emotion analysis algorithm into the generation AI to adjust the speech expression based on the emotion of the text using the emotion estimation function. For example, a brighter tone is used for text containing the emotion of joy. The speech generation unit also trains the generation AI with a large amount of emotion-labeled text data to analyze the emotion of the text. For example, text data containing emotions such as joy, sadness, and anger is trained. The speech generation unit also analyzes features of the speech such as tone, pitch, and speed to enhance the emotion estimation function. For example, the tone is raised and the speed is increased in parts where emotions are heightened. In this way, by adjusting the speech expression based on the emotion of the text, a more emotional speech can be provided.

[0080] The speech generation unit can incorporate a function to automatically add background music and sound effects when converting text to speech. For example, the speech generation unit incorporates a function to automatically add background music and sound effects into the generation AI when converting text to speech. For example, music suitable for the background of news is added. The speech generation unit also has the generation AI analyze the content and context of the text to select background music and sound effects. For example, moving music is added for moving scenes. The speech generation unit also incorporates a function that allows the user to customize background music and sound effects. For example, the user can select their favorite music and sound effects and generate speech based on them. This allows for a richer audio experience by adding background music and sound effects.

[0081] The speech generation unit can simulate the voices of multiple speakers when converting text to speech and generate conversational speech. For example, the speech generation unit adds a function to the generation AI to simulate the voices of multiple speakers when converting text to speech. For example, it converts conversational text into speech using multiple voices. In addition, the speech generation unit trains the generation AI with data on various voice qualities and accents to simulate the voices of multiple speakers. For example, it trains voices of different ages and genders. In addition, the speech generation unit adds a function that allows the user to customize the speaker voice. For example, the user can select their favorite voice and generate conversational speech based on that. In this way, conversational speech can be provided by simulating the voices of multiple speakers.

[0082] The voice generation unit can use the emotion estimation function to provide voice feedback according to the user's emotions, thereby improving the viewing experience. For example, the voice generation unit incorporates an emotion analysis algorithm into the generation AI to provide voice feedback according to the user's emotions using the emotion estimation function. For example, if the user is excited, it provides encouraging voice feedback. The voice generation unit also has the generation AI learn the user's facial expressions and voice data to analyze the user's emotions. For example, it analyzes the user's facial expressions and tone of voice to estimate emotions. The voice generation unit also adds an interactive voice feedback function to the generation AI to improve the viewing experience. For example, it allows the user to select voice feedback according to their emotions. This improves the viewing experience by providing voice feedback according to the user's emotions.

[0083] The speech generation unit can add a function to automatically adjust the intonation and emphasis of the speech according to the content of the text data. For example, the speech generation unit adds a function to the generation AI that automatically adjusts the intonation and emphasis of the speech according to the content of the text data. For example, the intonation can be emphasized when conveying important information. The speech generation unit also has the generation AI analyze the context and emotions of the text to adjust the intonation and emphasis of the speech. For example, the intonation can be raised in emotionally charged parts. The speech generation unit also adds a function that allows the user to customize the intonation and emphasis of the speech. For example, the user can set their preferred intonation and emphasis and generate speech based on that. This makes it possible to provide a more natural speech by adjusting the intonation and emphasis of the speech according to the content of the text data.

[0084] The speech generation unit can analyze text data and generate speech that emphasizes specific keywords and phrases. For example, the speech generation unit analyzes text data and incorporates a keyword extraction algorithm into the generation AI to emphasize specific keywords and phrases. For example, it generates speech that emphasizes important points of news. The speech generation unit also adds visual effects to the generation AI when emphasizing keywords and phrases. For example, it displays important keywords in bold or color. The speech generation unit also trains the generation AI on large amounts of text data to improve the accuracy of keyword extraction. For example, it trains the AI ​​on news articles and scripts to extract important keywords. This makes it easier to convey important information to viewers by emphasizing specific keywords and phrases.

[0085] The speech generation unit can use the emotion estimation function to adjust the tone of the speech based on the emotion of the text. For example, the speech generation unit incorporates an emotion analysis algorithm into the generation AI to adjust the tone of the speech based on the emotion of the text using the emotion estimation function. For example, a brighter tone is used for text containing the emotion of joy. The speech generation unit also trains the generation AI with a large amount of emotion-labeled text data to analyze the emotion of the text. For example, it trains text data containing emotions such as joy, sadness, and anger. The speech generation unit also analyzes features of the speech, such as tone, pitch, and speed, to enhance the emotion estimation function. For example, it raises the tone and speed in parts where emotions are heightened. In this way, by adjusting the tone of the speech based on the emotion of the text, it is possible to provide a more emotional speech.

[0086] The speech generation unit can add a function to simultaneously generate speech in different languages ​​when converting text data into speech. For example, the speech generation unit adds a function to the generation AI to simultaneously generate speech in different languages ​​when converting text data into speech. For example, English text can be simultaneously converted into speech in Japanese, Chinese, and Spanish. The speech generation unit also trains the generation AI to learn technical terms and slang in each language to improve translation accuracy. For example, it trains the AI ​​on text data containing medical and technical terms. The speech generation unit also has the generation AI perform translation in real time to provide multilingual speech. For example, it can translate into multiple languages ​​simultaneously during a live broadcast and provide the results to viewers. This allows multilingual speech to be provided by simultaneously generating speech in different languages.

[0087] The speech generation unit can incorporate a function that allows customization of the speech speed and pitch when converting text data into speech. For example, the speech generation unit incorporates a function that allows the generation AI to customize the speech speed and pitch when converting text data into speech. For example, a user can set their preferred speed and pitch, and speech is generated based on that. The speech generation unit also has the generation AI analyze the context and emotion of the text to adjust the speech speed and pitch. For example, the speed can be increased and the pitch can be raised in emotionally charged parts. The speech generation unit also adds a function that allows the user to adjust the speech speed and pitch in real time. For example, the user can change the speed and pitch during playback. This allows the speech speed and pitch to be customized, thereby providing speech that meets the user's preferences.

[0088] The voice generation unit can use the emotion estimation function to provide voice feedback according to the user's emotions, thereby improving the viewing experience. For example, the voice generation unit incorporates an emotion analysis algorithm into the generation AI to provide voice feedback according to the user's emotions using the emotion estimation function. For example, if the user is excited, it provides encouraging voice feedback. The voice generation unit also has the generation AI learn the user's facial expressions and voice data to analyze the user's emotions. For example, it analyzes the user's facial expressions and tone of voice to estimate emotions. The voice generation unit also adds an interactive voice feedback function to the generation AI to improve the viewing experience. For example, it allows the user to select voice feedback according to their emotions. This improves the viewing experience by providing voice feedback according to the user's emotions.

[0089] The subtitle providing unit allows the generation AI to learn past subtitle data and create a model specialized for a specific program and genre in order to improve the accuracy of providing audio subtitles in real time. For example, the subtitle providing unit has the generation AI learn past subtitle data in order to improve the accuracy of providing audio subtitles in real time. For example, the subtitle providing unit has the generation AI learn subtitle data from news programs or dramas to create a model specialized for a specific program. In addition, when creating a model specialized for a specific program or genre, the subtitle providing unit has the generation AI learn the characteristics of the program and the voice quality of the speaker. For example, the subtitle providing unit has the generation AI learn the live audio of a sports program. In addition, in order to improve the accuracy of the model, the subtitle providing unit periodically adds new subtitle data to the generation AI and has it learn from it. For example, data from a new season of a drama or news program is added. In this way, by learning from past subtitle data, an audio subtitle providing model specialized for a specific program or genre can be created.

[0090] The subtitle providing unit can introduce an algorithm that improves the processing speed of the generation AI to optimize the timing of displaying audio subtitles. For example, the subtitle providing unit introduces a high-speed processing algorithm into the generation AI to optimize the timing of displaying audio subtitles. For example, it analyzes audio data in real time and displays subtitles instantly. The subtitle providing unit also incorporates parallel processing technology into the generation AI to optimize the processing speed. For example, it analyzes multiple pieces of audio data simultaneously and generates subtitles. The subtitle providing unit also uses hardware acceleration in the generation AI to optimize the display timing. For example, it uses a GPU or dedicated chip to speed up speech recognition processing. This makes it possible to provide subtitles in real time by optimizing the timing of displaying audio subtitles.

[0091] The subtitle providing unit can use the emotion estimation function to add emotional expressions according to the content of the subtitles. For example, the subtitle providing unit incorporates an emotion analysis algorithm into the generation AI to add emotional expressions according to the content of the subtitles using the emotion estimation function. For example, if the speaker is excited, an exclamation mark is added. The subtitle providing unit also has the generation AI analyze features such as tone, pitch, and speed of the voice to enhance the emotional expression of the subtitles. For example, if the speaker's voice is high-pitched and fast, it is estimated that the speaker is excited. The subtitle providing unit also has the generation AI train a large amount of emotion-labeled audio data to enhance the emotion estimation function. For example, it trains audio data containing emotions such as joy, sadness, and anger. This allows the addition of emotional expressions according to the content of the subtitles, thereby providing viewers with more emotionally rich subtitles.

[0092] The subtitle providing unit can add a function to display real-time audio subtitles in a visually easy-to-understand format. For example, the subtitle providing unit adds color coding and font change functions to the generation AI to display real-time audio subtitles in a visually easy-to-understand format. For example, important information may be displayed in red. The subtitle providing unit also adds interactive effects to the generation AI to enhance the visual display of the subtitles. For example, the subtitle providing unit allows the user to access detailed information by clicking or zooming. The subtitle providing unit also adds a function to customize the visual display. For example, the user may select their preferred color or font and display subtitles based on that. This makes it easier to convey information to viewers by displaying it in a visually easy-to-understand format.

[0093] The subtitle provider can add a function to synchronize audio subtitles with other devices (smartphones and tablets) in real time. For example, the subtitle provider incorporates a data synchronization algorithm into the generation AI to synchronize audio subtitles with other devices in real time. For example, the subtitle provider instantly sends subtitle data to smartphones and tablets. The subtitle provider also introduces network optimization technology into the generation AI to improve synchronization accuracy between devices. For example, the subtitle provider uses a protocol for transmitting data with low latency. The subtitle provider also uses cloud-based data storage in the generation AI to achieve real-time data synchronization. For example, the subtitle data is stored on a cloud server and made accessible from each device. This allows viewers to view information on multiple devices by synchronizing audio subtitles with other devices in real time.

[0094] The subtitle providing unit can use the emotion estimation function to provide real-time feedback according to the viewer's emotions and personalize the viewing experience. For example, the subtitle providing unit incorporates an emotion analysis algorithm into the generation AI to provide real-time feedback according to the viewer's emotions using the emotion estimation function. For example, if the viewer is excited, related content is suggested. The subtitle providing unit also has the generation AI learn viewer emotion data to personalize the viewing experience. For example, it predicts emotional responses based on past viewing history and feedback. The subtitle providing unit also adds interactive functions to the generation AI to provide real-time feedback. For example, it allows the viewer to select content according to their emotions. This makes it possible to personalize the viewing experience by providing feedback according to the viewer's emotions.

[0095] When the system converts speech to text in real time and converts the text to speech to provide subtitles, the generation AI can learn from past data to create a model specialized for a specific program and genre in order to improve the accuracy of real-time speech recognition and speech generation. For example, the system has the generation AI learn from past data to improve the accuracy of real-time speech recognition and speech generation. For example, the system has the generation AI learn from audio data of news programs or dramas to create a model specialized for a specific program. In addition, when creating a model specialized for a specific program or genre, the system has the generation AI learn the characteristics of the program and the voice quality of the speaker. For example, the system has the generation AI learn from the commentary audio of a sports program. In addition, the system periodically adds new data to the generation AI to improve the accuracy of the model. For example, data from a new season of a drama or news program is added. In this way, by learning from past data, the accuracy of speech recognition and speech generation specialized for a specific program or genre is improved.

[0096] The system can introduce an algorithm that optimizes the processing speed of the generative AI to minimize delays in speech recognition and speech generation. For example, the system introduces a high-speed processing algorithm into the generative AI to minimize delays in speech recognition and speech generation. For example, the system analyzes speech data in real time and instantly converts it into text and speech. The system also incorporates parallel processing technology into the generative AI to optimize processing speed. For example, the system simultaneously analyzes multiple pieces of speech data, converts them into text, and converts them into speech. The system also uses hardware acceleration in the generative AI to minimize delays. For example, the system uses a GPU or dedicated chip to speed up speech recognition and speech generation processing. This minimizes delays in speech recognition and speech generation, enabling real-time speech recognition and speech generation.

[0097] The system can use the emotion estimation function to analyze a speaker's emotion in real time and adjust the text and voice tone based on that emotion. For example, the system uses the emotion estimation function to analyze a speaker's emotion in real time and adjust the text and voice tone based on that emotion. For example, if the speaker is angry, an emphasized expression is used. To analyze emotions in real time, the system also has the generation AI analyze features such as tone, pitch, and speed of the voice. For example, if the speaker's voice is high-pitched and fast, it is estimated that the speaker is excited. To enhance the emotion estimation function, the system also trains the generation AI with a large amount of emotion-labeled voice data. For example, it trains voice data containing emotions such as joy, sadness, and anger. This allows the system to adjust the text and voice tone based on the speaker's emotion, thereby providing more natural-looking subtitles and voice.

[0098] The system can add a function to display real-time text data and audio in a visually easy-to-understand format. For example, the system adds a chart or graph generation function to the generation AI to display real-time text data and audio in a visually easy-to-understand format. For example, the system displays important points of news in a graph. The system also incorporates a data summary or highlight function into the generation AI when visually displaying text data and audio. For example, the system highlights important keywords. The system also causes the generation AI to generate interactive graphs and charts to enhance the visual display. For example, the system allows users to access more detailed information by clicking or zooming. This makes it easier to convey information to viewers by displaying it in a visually easy-to-understand format.

[0099] The system can add a function for synchronizing the transcribed data and audio with other devices (smartphones and tablets) in real time. For example, the system incorporates a data synchronization algorithm into the generation AI to synchronize the transcribed data and audio with other devices in real time. For example, the system instantly sends text and audio data to smartphones and tablets. The system also introduces network optimization technology into the generation AI to improve the accuracy of synchronization between devices. For example, the system uses a protocol for transmitting data with low latency. The system also uses cloud-based data storage in the generation AI to achieve real-time data synchronization. For example, the system stores text and audio data on a cloud server and makes it accessible from each device. This allows viewers to view information on multiple devices by synchronizing the transcribed data and audio with other devices in real time.

[0100] The system can use the emotion estimation function to provide real-time feedback according to the viewer's emotions and personalize the viewing experience. For example, the system incorporates an emotion analysis algorithm into the generation AI to use the emotion estimation function to provide real-time feedback according to the viewer's emotions. For example, if the viewer is excited, related content is suggested. The system also has the generation AI learn viewer emotional data to personalize the viewing experience. For example, it predicts emotional reactions based on past viewing history and feedback. The system also adds interactive functions to the generation AI to provide real-time feedback. For example, it allows the viewer to select content according to their emotions. This makes it possible to personalize the viewing experience by providing feedback according to the viewer's emotions.

[0101] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0102] The TV assistant system can also include a recommendation unit that analyzes the user's viewing history and recommends programs based on the viewer's preferences. For example, it can analyze the genre and content of previously viewed programs and recommend similar programs. The recommendation unit can also recommend programs at optimal times, taking into account the viewer's viewing time and frequency. Furthermore, the recommendation unit can collect viewer feedback and use it to improve the recommendation algorithm. This enables program recommendations based on the viewer's preferences, improving the viewing experience.

[0103] The TV auxiliary system can also be equipped with a health management unit that monitors the viewer's health and encourages appropriate breaks. For example, sensors can detect the viewer's heart rate and posture and suggest breaks to reduce the health risks associated with long viewing periods. The health management unit can also automatically adjust the screen brightness and contrast to reduce eye strain. Furthermore, the health management unit can collect the viewer's health data and provide personalized health advice. This allows for a comfortable viewing experience while maintaining the viewer's health.

[0104] The TV auxiliary system may further include an advertisement serving unit that estimates the viewer's emotions and customizes advertisements based on the estimated emotions. For example, if the viewer is excited, an energetic advertisement may be displayed. Alternatively, if the viewer is relaxed, an advertisement with a relaxing effect may be displayed. Furthermore, the advertisement serving unit may collect viewer's emotional data and use it to analyze the effectiveness of advertisements. This allows the effectiveness of advertisements to be maximized by providing advertisements that correspond to the viewer's emotions.

[0105] The TV auxiliary system may further include a scene highlighting unit that estimates the viewer's emotions and highlights program scenes based on the estimated emotions. For example, scenes that excite the viewer may be highlighted. Scenes that move the viewer may also be highlighted. Furthermore, the scene highlighting unit may collect viewer emotion data and use it for editing the program. This improves the viewing experience by providing scene highlights that correspond to the viewer's emotions.

[0106] The TV auxiliary system may further include an audio adjuster that estimates the viewer's emotion and adjusts the tone of the audio based on the estimated emotion. For example, if the viewer is excited, the audio tone may be increased. Alternatively, if the viewer is relaxed, the audio tone may be decreased. Furthermore, the audio adjuster may collect viewer emotion data and use it to improve the audio tone adjustment algorithm. This allows for a more natural viewing experience by providing an audio tone that corresponds to the viewer's emotion.

[0107] The TV auxiliary system may further include a subtitle color change unit that estimates the viewer's emotion and changes the subtitle color based on the estimated emotion. For example, if the viewer is excited, the subtitle color may be changed to red. Alternatively, if the viewer is relaxed, the subtitle color may be changed to blue. Furthermore, the subtitle color change unit may collect viewer emotion data and use it to improve the subtitle color change algorithm. This may improve the visual viewing experience by providing subtitle colors that correspond to the viewer's emotion.

[0108] The TV auxiliary system can also include a recording unit that analyzes the viewer's viewing history and automatically records programs the viewer missed. For example, it can analyze the genre and content of programs the viewer has previously watched and automatically record programs that the viewer missed. The recording unit can also record programs at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the recording unit can collect viewer feedback and use it to improve the recording algorithm. This improves the viewing experience by automatically recording programs the viewer missed.

[0109] The TV auxiliary system can also include a trailer player that analyzes the viewer's viewing history and automatically plays trailers for programs that the viewer might be interested in. For example, it can analyze the genre and content of programs previously viewed and play trailers for similar programs. The trailer player can also play trailers at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the trailer player can collect viewer feedback and use it to improve the trailer playback algorithm. This improves the viewing experience by automatically playing trailers for programs that the viewer might be interested in.

[0110] The TV auxiliary system can further include an information provider that analyzes the viewer's viewing history and provides related information about programs that may be of interest to the viewer. For example, it can analyze the genre and content of previously viewed programs and provide related news articles and interviews. The information provider can also provide related information at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the information provider can collect viewer feedback and use it to improve the information provision algorithm. This improves the viewing experience by providing related information about programs that may be of interest to the viewer.

[0111] The TV assistance system can further include a feature creation unit that analyzes the viewer's viewing history and automatically creates special features of programs that the viewer may be interested in. For example, it can analyze the genre and content of programs viewed in the past and create special features by collecting similar programs. The feature creation unit can also provide special features at the optimal time, taking into account the viewer's viewing time and frequency. Furthermore, the feature creation unit can collect viewer feedback and use it to improve the feature creation algorithm. This improves the viewing experience by automatically creating special features of programs that the viewer may be interested in.

[0112] The processing flow of the second embodiment will be briefly explained below.

[0113] Step 1: The audio capture unit captures audio from a television program. For example, the audio capture unit captures audio directly from the audio output of a television. The audio capture unit can also capture streaming audio via the Internet. Furthermore, the audio capture unit can also capture ambient audio using a microphone. Step 2: The text conversion unit converts the speech captured by the speech capture unit into text in real time. For example, the text conversion unit converts speech into text using a generation AI (e.g., a generation AI using speech recognition technology). The text conversion unit can also use the generation AI to analyze speech data and convert it into text data. Furthermore, the text conversion unit can also use the generation AI to summarize the content of the speech data and convert it into text. Step 3: The speech generation unit converts the text converted by the text conversion unit into speech. For example, the speech generation unit converts the text into speech using a generation AI (e.g., a generation AI using text-to-speech technology). The speech generation unit can also analyze the text data using the generation AI and generate natural speech. Furthermore, the speech generation unit can also generate speech according to the content of the text data using the generation AI. Step 4: The subtitle providing unit provides the audio generated by the audio generating unit as subtitles. For example, the subtitle providing unit displays the audio as subtitles using a generation AI (e.g., a generation AI using subtitle generation technology). The subtitle providing unit can also analyze audio data using the generation AI and generate subtitle data. Furthermore, the subtitle providing unit can also generate subtitles according to the content of the audio data using the generation AI.

[0114] 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.

[0115] 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.

[0116] 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.

[0117] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0118] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0119] 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.

[0120] 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.

[0121] 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.

[0122] 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).

[0123] 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.

[0124] 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.

[0125] 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.

[0126] 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.

[0127] 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.

[0128] 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.

[0129] 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.

[0130] 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.

[0131] 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.

[0132] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0133] 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.

[0134] 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.

[0135] 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.

[0136] 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.

[0137] 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).

[0138] 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.

[0139] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type 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.

[0140] 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.

[0141] 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.

[0142] In the headset type terminal 314, 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. 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 specific processing unit 290 using these models.

[0143] 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.

[0144] 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.

[0145] 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.

[0146] 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.

[0147] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0148] 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.

[0149] 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.

[0150] 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.

[0151] 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.

[0152] 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).

[0153] 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.

[0154] 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.

[0155] 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.

[0156] 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.

[0157] 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.

[0158] In the robot 414, 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 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 processing similar to that of the specific processing unit 290 using these models.

[0159] 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.

[0160] 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.

[0161] 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.

[0162] 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.

[0163] 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.

[0164] 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.

[0165] 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.

[0166] 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).

[0167] 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 "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[0168] 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."

[0169] 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.

[0170] 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.

[0171] 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.

[0172] 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.

[0173] 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.

[0174] 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.

[0175] 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.

[0176] 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.

[0177] 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.

[0178] 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.

[0179] 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, to avoid confusion and 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.

[0180] 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. [Explanation of symbols]

[0181] 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. an audio acquisition unit that acquires audio from a television program; a text conversion unit that converts the voice acquired by the voice acquisition unit into text in real time; a voice generating unit that converts the text converted by the text conversion unit into voice; a subtitle providing unit that provides the audio generated by the audio generating unit as subtitles. A system characterized by:

2. The text conversion unit The generative AI creates a custom model that learns the voice quality and accent of a specific speaker.

2. The system of claim 1.

3. The text conversion unit Add a filtering function that automatically removes background sounds and noise from audio data and converts only clear audio into text.

2. The system of claim 1.

4. The text conversion unit Estimating the speaker's emotion and adjusting the presentation of the text based on the emotion 2. The system of claim 1.

5. The text conversion unit When converting the audio into text, it translates it into multiple languages ​​simultaneously to provide multilingual subtitles.

2. The system of claim 1.

6. The text conversion unit Add the ability to analyze voice data and highlight specific keywords and phrases 2. The system of claim 1.

Images