System

Abstract

A system is provided.SOLUTION: A system comprising: means for inputting a sign language animation; means for analyzing the inputted sign language animation and converting the analyzed sign language animation into a text in each language; and means for displaying the converted text to a user.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] Because sign languages ​​differ from country to country, it is currently difficult for people who use sign languages ​​to understand the sign languages ​​and written texts of other countries. There is also a need for tools to facilitate communication between people who use sign languages ​​and those who do not understand sign languages. Furthermore, because there are various types of sign languages ​​and ways of expression, precision in converting between sign languages ​​and written texts is required. The purpose of this invention is to eliminate language barriers and facilitate smoother communication. [Means for solving the problem]

[0005] The present invention provides a system that inputs sign language video, analyzes the input sign language video, converts it into text in a variety of languages, and displays the converted text to the user. It also provides a system that inputs text in a variety of languages, analyzes the input text, converts it into a corresponding sign language video, and displays the converted sign language video to the user. It also includes a system that converts speech to text, analyzes the text, converts it into a corresponding sign language video, and displays the generated sign language video to the user. These methods enable two-way translation between sign language and a variety of languages, facilitating multilingual communication.

[0006] "Sign language video" is video data that records specific sign language actions.

[0007] "Input means" refers to the interface that allows the user to provide sign language video, text, and audio to the system.

[0008] An "analysis means" is an algorithm or model that processes input data and converts it into an appropriate language or format.

[0009] The "conversion means" refers to a processing device or software that converts the analyzed data into a different language or format.

[0010] A "means for displaying" is a display device or application that visually presents the converted data in a form that can be understood by a user.

[0011] "National texts" are written texts corresponding to different countries or languages.

[0012] "Means for generating sign language videos" refers to technology for creating videos containing corresponding sign language actions based on text and audio data.

[0013] A "speech-to-text means" is a process that analyzes input speech data and converts it into corresponding sentences.

[0014] "User" is an individual or group who uses the system to translate between sign language and national languages. [Brief explanation of the drawings]

[0015] [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. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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, a 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), and an APU (Accelerated Processing Unit).

[0019] 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.

[0020] 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.

[0021] 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), Bluetooth (registered trademark), etc.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the 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).

[0027] 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.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. 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 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] This invention is a system that enables bidirectional translation between sign language and other languages. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[0037] When converting sign language videos into other languages

[0038] Program processing overview

[0039] 1. Input sign language video

[0040] The user uses the device to record a specific sign language video and upload it to the application.

[0041] 2. Send the sign language video to the server

[0042] The terminal transmits the sign language video data to the server.

[0043] 3. Analyzing sign language videos

[0044] The server receives the sign language video data and analyzes the video based on the generative AI model.

[0045] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[0046] 4. Generate the conversion results

[0047] The server returns to the terminal text in each language generated from the analyzed sign language movements.

[0048] 5. View the conversion results

[0049] The terminal displays the text data received from the server to the user.

[0050] Specific examples

[0051] A user can record a video of themselves speaking Japanese Sign Language on their smartphone and upload it through an application. This video is then sent over the internet to a server, which uses a generative AI model to analyze the sign language movements and convert them into Japanese text, such as "hello." This text is then sent back to the device and displayed to the user.

[0052] Converting text or dialogue from a foreign language into sign language video

[0053] Program processing overview

[0054] 1. Enter text or voice

[0055] The user can enter text using the device's keyboard or by speaking into the microphone.

[0056] 2. Send the input data to the server

[0057] The device sends the captured text or audio data to the server.

[0058] 3. Analyze the input data

[0059] The server receives the text or voice data and converts the voice to text as needed.

[0060] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[0061] 4. Generate sign language video

[0062] The server generates a corresponding sign language video based on the analyzed text data.

[0063] 5. Display sign language video

[0064] The terminal displays the sign language video data received from the server to the user.

[0065] Specific examples

[0066] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to a server, which uses a speech recognition model to convert the speech to text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video and sends it back to the device. Finally, the device displays the video of "Bonjour" in French Sign Language to the user.

[0067] As described above, the present invention realizes two-way translation between sign language and various languages ​​and functions as a system that supports communication between multiple languages.

[0068] The processing flow will be explained below.

[0069] When converting sign language videos into other languages

[0070] Program processing details

[0071] Step 1:

[0072] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[0073] Step 2:

[0074] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[0075] Step 3:

[0076] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[0077] Step 4:

[0078] The generative AI model analyzes the sign language movements for each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[0079] Step 5:

[0080] The server prepares the results of the analyzed sign language movements as text data and returns it to the terminal.

[0081] Step 6:

[0082] The terminal receives the text data in each language received from the server and displays it to the user. For example, the displayed text is "Hello."

[0083] Converting text or dialogue from a foreign language into sign language video

[0084] Program processing details

[0085] Step 1:

[0086] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[0087] Step 2:

[0088] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[0089] Step 3:

[0090] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[0091] Step 4:

[0092] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[0093] Step 5:

[0094] The server converts the generated sign language motion data into a sign language video, for example, generating a video of "Bonjour" in French sign language.

[0095] Step 6:

[0096] The server generates a response to send the generated sign language video data to the terminal, which receives the sign language video data via the Internet.

[0097] Step 7:

[0098] The device receives the sign language video data from the server and displays it to the user, who can then watch the video of "Bonjour" in French sign language on the device screen.

[0099] The above is a specific processing flow of the bidirectional translation system between sign language and other national languages ​​based on the present invention.

[0100] Example 1

[0101] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0102] Conventional translation systems between sign language and other languages ​​lack accuracy and real-time performance, making it difficult to facilitate smooth communication between multiple languages. Furthermore, while there is a need not only to convert sign language video into text in other languages, but also to convert text and audio in other languages ​​into sign language video, there are only a limited number of systems that can achieve this kind of two-way translation function. Therefore, there is a need for a system that can perform two-way translation with higher accuracy and in real time to support multilingual communication.

[0103] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0104] In this invention, the server includes means for a user to input sign language video, means for dividing the input sign language video and analyzing the hand position and shape of each frame, means for generating text in various languages ​​based on the analyzed sign language movements using a generative AI model, means for returning the converted text to the terminal, and means for displaying the converted text to the user, thereby enabling sign language video to be converted into text in various languages ​​with high accuracy and in real time.

[0105] "Sign language video" is video data that records communication using sign language.

[0106] "User" refers to a person who uses the system to input sign language video or input text or audio in various languages.

[0107] A "generative AI model" is a model that generates and converts text data and sign language motion data based on artificial intelligence. Examples include models that use natural language processing and computer vision techniques.

[0108] "Analysis" refers to the process of recognizing the hand position and shape for each frame of the input sign language video and converting it into text in each language based on that.

[0109] "Text" refers to character string data expressed in various languages, and refers to written language information.

[0110] A "terminal" is a device that allows a user to input sign language video, text, or audio, and receive responses from a server. Examples include smartphones and tablets.

[0111] A "server" is a computer system that analyzes input sign language video, text, and audio data, performs any necessary conversions, and then returns the results to the terminal.

[0112] The "Sign Language Animation Library" is a library for expressing sign language movements as animation using 3D motion capture data.

[0113] "Speech-to-text means" refers to any technology or function that analyzes and converts audio data into corresponding text data. An example is voice recognition software.

[0114] "Means for analyzing the position and shape of the hands in each frame" refers to technology or functions that divide sign language video into frames and recognize the position and shape of the hands in each frame.

[0115] "Means for generating sign language videos" refers to technologies and functions that generate new sign language videos based on analyzed sign language movements.

[0116] This invention is a system that realizes two-way translation between sign language and other languages, and is a system that mainly operates through the cooperation of three parties: a server, a terminal, and a user. This system provides functions to convert sign language videos into other languages, and to convert text and audio in other languages ​​into sign language videos.

[0117] A form of converting sign language videos into various languages

[0118] The user records a sign language video using a device such as a smartphone. The recorded sign language video is then sent to a server using a dedicated application on the device. The server divides the received sign language video and analyzes the hand position and shape for each frame using computer vision technology. Libraries such as TENSORFLOW (registered trademark) and OpenCV are used for the analysis.

[0119] Based on the analyzed sign language motion data, a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) is used to generate text in the corresponding language. An example of a generative AI model that can be implemented is OpenAI's GPT-4.

[0120] The generated text in each language is sent back to the device in JSON format and displayed to the user by the device application. As a concrete example, a user records a video of themselves saying "hello" in Japanese sign language and uploads it to the application. This video is sent to a server, which analyzes the sign language movements and converts them into the Japanese text "hello". The converted text is then sent back to the device and displayed to the user.

[0121] A form of converting text or audio in various languages ​​into sign language video

[0122] Users can input text in their local language using an application on their device, or input voice using a microphone. The device then sends the input text or voice data to the server. In the case of voice data, it is first converted into text using voice recognition software (e.g., Google's Speech-to-Text API).

[0123] The server uses a generative AI model to convert the input text into appropriate sign language actions, then uses a sign language animation library (e.g., a library using 3D motion capture data) to generate a sign language video, which is sent back from the server to the device and displayed to the user by the device application.

[0124] As a concrete example, if a user types or speaks "Bonjour" in French, the entered text or voice data is sent from the device to a server. The server uses speech recognition software to convert the speech into text and analyzes it with a generative AI model. Based on this data, the server generates a corresponding French Sign Language video and sends it back to the device. Finally, the sign language video is played in an application on the device.

[0125] Prompt Sentence Examples

[0126] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[0127] "Create a program that converts input text or audio into a sign language video and plays it back."

[0128] As described above, the present invention is a system that realizes bidirectional translation between sign language and various other languages ​​with high accuracy and in real time, thereby supporting communication between multiple languages.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] When converting sign language videos into other languages

[0131] Processing steps and specific explanations

[0132] Step 1:

[0133] The user records a sign language video using the camera on their smartphone and presses the "upload" button in the dedicated application. The input is the recorded sign language video file, and the output is a video file sent from the device to the server.

[0134] Step 2:

[0135] The device sends the uploaded sign language video file to a server via the Internet. Specifically, the HTTPS protocol is used to encrypt data and ensure secure communication. The input is the sign language video file, and the output is the video data sent to the server.

[0136] Step 3:

[0137] The server temporarily stores the received sign language video and begins analyzing it using a generative AI model. The video is first divided into frames, and the hand position and shape of each frame are analyzed using computer vision technology (e.g., OpenCV, TensorFlow). The input is the sign language video stored on the server, and the output is analysis data including the hand position and shape.

[0138] Step 4:

[0139] The server uses a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) to generate text in the corresponding language based on the analyzed sign language movement data. For generation, OpenAI's GPT-4 is used, for example. The input is sign language movement data, and the output is text data in the language.

[0140] Step 5:

[0141] The server converts the generated text in each language into JSON format and returns it to the terminal. The input is the generated text data, and the output is JSON format text data.

[0142] Step 6:

[0143] The terminal parses the JSON-formatted text data received from the server and displays it in the user interface. Specifically, it displays the generated text in each language in a text view within the application. The input is JSON-formatted text data, and the output is the text displayed in the user interface.

[0144] Converting text or audio in any language into sign language video

[0145] Processing steps and specific explanations

[0146] Step 1:

[0147] The user opens the application on their device and enters the text they want to translate into the text field, or presses the voice input button and speaks into the microphone. The input is text or voice data, and the output is data sent from the device to the server.

[0148] Step 2:

[0149] The device sends text data or recorded voice data to the server. In the case of voice data, it is first saved as a voice file on the device and then sent to the server. The input is text data or voice data, and the output is the data sent to the server.

[0150] Step 3:

[0151] The server directly analyzes the received text data, and converts audio data into text using Google's Speech-to-Text API. The input is text data or audio data, and the output is the converted data.

[0152] Step 4:

[0153] The server uses a generative AI model to convert input text in various languages ​​into data corresponding to sign language actions. The input is text data, and the output is sign language action data.

[0154] Step 5:

[0155] The server generates sign language animation using a sign language animation library (e.g., a 3D motion capture library) based on the sign language movement data. The input is the sign language movement data, and the output is the generated sign language video.

[0156] Step 6:

[0157] The terminal plays the sign language video data received from the server on the user interface. The user can view the translated sign language video through the application. The input is sign language video data, and the output is a sign language video that the user can view.

[0158] Prompt Sentence Examples

[0159] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[0160] "Create a program that converts input text or audio into a sign language video and plays it back."

[0161] As described above, by clearly showing the specific processing steps and operations, it has been shown how the system of the present invention provides bidirectional translation between sign language and various other languages ​​with high accuracy and in real time.

[0162] (Application example 1)

[0163] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0164] It aims to address the lack of communication tools that allow sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies.

[0165] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0166] In this invention, the server includes means for analyzing input sign language video and converting it into text in various languages, means for transmitting the input sign language video as an emergency call, and means for converting reply text from the server into sign language video, thereby enabling sign language users to make emergency calls via sign language video and respond quickly.

[0167] "Sign language video" is information including hand movements and facial expressions recorded in video format so that people who use sign language can understand it visually.

[0168] "Text in a national language" refers to character strings or sentences written in a specific language, and is text data in a format that can be read and understood by humans.

[0169] "User" refers to a person who uses this system to input, transmit, and display sign language video.

[0170] A "server" is a computer system that analyzes and converts sign language videos and text and manages communications.

[0171] An "emergency call" is a communication act to warn or request help in an emergency, and refers to the transmission of information that requires a rapid response.

[0172] "Reply text" is a written string or sentence sent from a server or the like as a response to an emergency call.

[0173] "Analysis" refers to the process of understanding the content of sign language video or text and converting it into another format.

[0174] "Conversion" refers to the process of converting data of one format into another format, and in this invention it primarily refers to conversion between sign language video and text in a foreign language.

[0175] This invention is a system that allows sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[0176] Sign language video input and analysis

[0177] A user first uses a device (e.g., a smartphone) to record a video of an emergency situation in sign language. This video is then sent to a server via an application. The device then uploads the video to the server, using a generative AI model to convert it into text in each language.

[0178] Analysis of sign language video and conversion to text

[0179] The server receives the sign language video data and analyzes the video based on the generative AI model. This analysis converts the sign language actions into text in the corresponding language. The converted text is then sent from the server to the relevant authorities (e.g., police or security center) as an emergency call.

[0180] Sending and replying to emergency calls

[0181] When a reply from a relevant organization is sent to the server in text format, the server analyzes the text data and generates a corresponding sign language video. This sign language video is then sent to the device and displayed to the user. This allows the user to quickly report the situation through sign language and receive necessary assistance, even in an emergency.

[0182] Example

[0183] For example, suppose a user discovers a suspicious person at home and uses their smartphone to record a sign language video saying "Help me" and report it. This video is sent to a server and converted into text for the Japanese word "Help me." This text is then sent to the police, who reply, "We will send a patrol immediately." This reply is analyzed by the server, converted into the corresponding sign language video, and displayed on the user's device.

[0184] Hardware and software used

[0185] Hardware: Your smartphone's camera, microphone, and display.

[0186] Software: Server-side generative AI models, speech recognition models, and text analysis tools.

[0187] Specific names: requests, cv2 (OpenCV), and speech_recognition libraries implemented in Python.

[0188] Prompt Sentence Examples

[0189] "Convert the French word 'help' into the corresponding sign language action."

[0190] By inputting this prompt into a generative AI model, the necessary sign language actions are generated. In this way, the present invention enables sign language users to safely make emergency calls and receive prompt responses.

[0191] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0192] Step 1:

[0193] A user uses a smartphone to record a sign language video about an emergency (e.g., discovering a suspicious person at home). A video captured with the smartphone camera is generated as input data for the sign language video. The recorded sign language video is then saved by the user via an application.

[0194] Step 2:

[0195] The device sends the recorded sign language video data to a server. Specifically, an application on the smartphone uses an Internet connection to upload the sign language video to a specified API endpoint on the server. The input is a sign language video file, and the output is the data sent to the server.

[0196] Step 3:

[0197] The server analyzes the received sign language video data using a generative AI model. It identifies each frame of the video and analyzes hand movements and facial expressions. As a result of this analysis, the input sign language video is converted into natural language text (text in each language). The input is sign language video data, and the output is text in the corresponding language.

[0198] Step 4:

[0199] The server sends the parsed text in each language as an emergency call to the appropriate security agency (e.g., police). The text data is communicated through the server's API. The input is the generated text in each language, and the output is the call data to the security agency.

[0200] Step 5:

[0201] A security agency (e.g., the police) receives the call and creates a reply in the local language. This reply text is sent back to the server. Here, as a police operation, the input is the text of the emergency call and the output is the local language text of the reply.

[0202] Step 6:

[0203] The server receives the reply text data from the security agency. This text in each language is analyzed again using the generative AI model and converted into the corresponding sign language video. The input is the reply text data, and the output is the sign language video.

[0204] Step 7:

[0205] The server sends the generated sign language video to the device. The device then displays the video to the user. The input is the sign language video data from the server, and the output is the sign language video displayed on the smartphone.

[0206] Step 8:

[0207] The user checks the sign language video on the device and understands the reply from the security agency in sign language. The input is the sign language video, and the output is what the user understands.

[0208] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0209] This invention is a system that enables two-way translation between sign language and other languages, and also has the ability to adjust the translation results by recognizing the user's emotions. This system mainly works in cooperation with three entities: the server, the terminal, and the user, and by combining an emotion engine, it provides more natural and appropriate translations.

[0210] When converting sign language videos into other languages

[0211] Program processing overview

[0212] 1. Input sign language video

[0213] A user uses a device such as a smartphone or PC to record a specific sign language video and upload it to the application.

[0214] 2. Send the sign language video to the server

[0215] The terminal transmits the sign language video data to the server.

[0216] 3. Analyzing sign language videos

[0217] The server receives the sign language video data and analyzes the video based on the generative AI model.

[0218] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[0219] 4. Emotional Engine Adjustment

[0220] The server uses an emotion engine to recognize the user's emotions.

[0221] Adjust the tone and nuance of the generated text based on the perceived sentiment.

[0222] 5. Generate the conversion results

[0223] The server returns the adjusted text in each language to the terminal.

[0224] 6. View the conversion results

[0225] The terminal displays the text data received from the server to the user.

[0226] Specific examples

[0227] Suppose a user films a video of themselves speaking Japanese Sign Language on their smartphone and uploads it through an application. The video is sent over the internet to a server, which uses a generative AI model to analyze the sign language and convert it into Japanese text, such as "hello." The emotion engine then recognizes the user's emotion (e.g., joy) and adjusts the nuance of the text to "a very happy 'hello.'" This text is then sent back to the device and displayed to the user.

[0228] Converting text or dialogue from a foreign language into sign language video

[0229] Program processing overview

[0230] 1. Enter text or voice

[0231] The user can input text in various languages ​​using the device's keyboard or by voice input using the microphone.

[0232] 2. Send the input data to the server

[0233] The device sends the captured text or audio data to the server.

[0234] 3. Analyze the input data

[0235] The server receives the text or voice data and converts the voice to text as needed.

[0236] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[0237] 4. Emotional Engine Adjustment

[0238] The server uses an emotion engine to recognize the user's emotions.

[0239] The expression of sign language actions is adjusted based on the recognized emotion.

[0240] 5. Generate sign language video

[0241] The server generates a corresponding sign language video based on the analyzed text data and emotions.

[0242] 6. Display sign language video

[0243] The terminal displays the sign language video data received from the server to the user.

[0244] Specific examples

[0245] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to the server, which uses a speech recognition model to convert the speech into text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video, and an emotion engine recognizes the user's joy and reflects it in the sign language video. The sign language video sent back to the device visualizes "Bonjour" with a clear expression of joyful movements.

[0246] In this way, the combination of the emotion engine makes two-way translation between sign language and national languages ​​more natural and effective, further facilitating multilingual communication.

[0247] The processing flow will be explained below.

[0248] When converting sign language videos into other languages

[0249] Program processing details

[0250] Step 1:

[0251] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[0252] Step 2:

[0253] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[0254] Step 3:

[0255] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[0256] Step 4:

[0257] The generative AI model analyzes the sign language movements in each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[0258] Step 5:

[0259] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[0260] Step 6:

[0261] Based on the recognized emotion, the tone and nuance of the generated text in each language is adjusted. For example, if the user is happy, "Hello" is adjusted to "A very happy 'Hello'."

[0262] Step 7:

[0263] The server generates a request to return the adjusted text to the terminal and sends it over the Internet to the terminal.

[0264] Step 8:

[0265] The terminal receives the text data received from the server and displays it to the user. For example, the displayed text might be "A very happy 'hello'."

[0266] Converting text or dialogue from a foreign language into sign language video

[0267] Program processing details

[0268] Step 1:

[0269] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[0270] Step 2:

[0271] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[0272] Step 3:

[0273] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[0274] Step 4:

[0275] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[0276] Step 5:

[0277] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[0278] Step 6:

[0279] The expression of the sign language actions is adjusted based on the recognized emotion. For example, if the user is happy, the sign language actions are adjusted to be more cheerful and positive.

[0280] Step 7:

[0281] The server generates a sign language video based on the generated sign language movement data and the adjusted expression. For example, it creates a video that cheerfully expresses "Bonjour" in French sign language.

[0282] Step 8:

[0283] The server generates a request to transmit the generated sign language video data to the terminal, and transmits it to the terminal via the Internet.

[0284] Step 9:

[0285] The device receives the sign language video data from the server and displays it to the user. For example, a video of the French sign language "Bonjour" being displayed brightly on the device screen.

[0286] In this way, by combining the emotion engine, two-way translation between sign language and other languages ​​becomes more natural and emotional, providing users with a more effective means of communication.

[0287] Example 2

[0288] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0289] Conventional translation systems between sign language and other languages ​​often produce mechanical results without taking into account the user's emotions. As a result, the translation results cannot be said to support natural and appropriate communication, and there are problems with a lack of accuracy and expressiveness. This is particularly noticeable in communication where emotional nuances are important.

[0290] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0291] In this invention, the server includes a means for inputting a sign language video, a means for analyzing the input sign language video and converting it into text in a national language using a generative AI model, a means for recognizing emotions during the analysis and adjusting the tone and nuance of the generated text, and a means for displaying the adjusted text to the user. This enables natural and appropriate bidirectional translation between sign language and a national language that incorporates emotions.

[0292] "Sign language video" is video data that records the movements and gestures of a person using sign language.

[0293] A "generative AI model" is an artificial intelligence model that automatically generates text, images, etc. based on data.

[0294] A "server" is a computer system for storing, processing, and managing data.

[0295] A "terminal" is a device used by a user to input and display data, such as a smartphone or PC.

[0296] "User" refers to a person who uses this system to translate between sign language and other national languages.

[0297] "Analysis" is the process of breaking down and extracting data and finding meaning and patterns within it.

[0298] An "emotion engine" is a program that recognizes the user's emotions from text or video and adjusts the output results based on those emotions.

[0299] "Text" refers to data that consists of sentences or characters.

[0300] "Tone and nuance" refers to subtle changes and emotional resonance in words and expressions.

[0301] This invention provides a system that enables bidirectional translation between sign language and other languages. This system operates mainly through the cooperation of three entities: the server, the terminal, and the user, and provides more natural and appropriate translation by combining an emotion engine.

[0302] Hardware and Software Configuration

[0303] Device: The device on which users record and upload sign language videos. This can be a smartphone or PC.

[0304] Server: A computer system for data analysis, transformation, and emotion recognition.

[0305] Generative AI models: Use generative AI models (e.g., OpenAI's GPT-4) to convert sign language videos into text in various languages.

[0306] Emotion engine: An emotion engine (e.g., IBM Watson® Tone Analyzer) is used to recognize the user's emotions and adjust the tone and nuance of the translation results based on the results.

[0307] System Operation

[0308] When converting sign language videos into other languages

[0309] 1. Recording and uploading sign language videos

[0310] Users record specific sign language videos on their smartphones or PCs and upload them to the application, which then sends the video to the server.

[0311] 2. Analysis of sign language videos

[0312] The server analyzes the received sign language video, using a generative AI model to analyze the sign language movements for each frame of the video and generate corresponding text in each language.

[0313] 3. Recognizing emotions and adjusting to nuances

[0314] The server uses an emotion engine to recognize the user's emotions from the sign language video and adjust the tone and nuance of the generated text, making it more appropriate and natural-sounding.

[0315] 4. Returning and displaying text

[0316] The server returns the adjusted text data to the terminal, which displays it to the user.

[0317] Specific examples

[0318] A user records a video of themselves saying "hello" in Japanese Sign Language and uploads it through the application. The server analyzes the video and converts it into the Japanese text "hello." The emotion engine recognizes the user's emotion of joy and adjusts the nuance of the text to "a very happy 'hello.'" Finally, this text data is sent back to the user's device and displayed.

[0319] Example prompt sentence:

[0320] "A user has uploaded a video in Japanese Sign Language. Please translate the sign language in this video into Japanese and use the emotion engine to recognize that it contains the emotion 'joy' and adjust appropriately."

[0321] Converting text or dialogue from a foreign language into sign language video

[0322] 1. Text or voice input

[0323] The user inputs text using the device's keyboard or voice using the microphone, and this data is sent to the server.

[0324] 2. Text or speech analysis

[0325] The server analyzes the received text or audio data, converting audio to text as needed, and using generative AI models to convert international text into sign language actions.

[0326] 3. Emotion Recognition and Video Generation

[0327] The server uses an emotion engine to recognize the user's emotion from the input data and adjusts the expression of the sign language video, generating the sign language video based on this.

[0328] 4. Returning and displaying sign language videos

[0329] The server sends the generated sign language video back to the terminal, which displays it to the user.

[0330] Specific examples

[0331] The user types or speaks "Bonjour" in French. This data is sent to the server, which performs speech recognition and then analyzes it using a generative AI model. A sign language video is generated from the analyzed data, and an emotion engine recognizes the user's emotion of joy and reflects it in the video. This sign language video is then sent back to the user's device and displayed.

[0332] Example prompt sentence:

[0333] "A user types 'Bonjour' in French. Convert this text into a sign language video, and then use an emotion engine to recognize that it contains the emotion 'joy' and reflect that in the video."

[0334] In this way, the system enables natural bidirectional translation between sign language and other national languages.

[0335] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0336] When converting sign language videos into other languages

[0337] Step 1:

[0338] Users record specific sign language videos on their smartphones or PCs and upload them to the application.

[0339] Input: Sign language video file

[0340] Output: Video file saved on device

[0341] Specific operation: The user opens the application, presses the record button to record a sign language video, and when the recording is finished, the video file is saved on the device.

[0342] Step 2:

[0343] The terminal transmits the sign language video data to the server.

[0344] Input: Sign language video file saved on the device

[0345] Output: Sign language video file transferred to the server

[0346] Specific operation: When you press the "Send" button in the application, the video file is uploaded to the server via the Internet.

[0347] Step 3:

[0348] The server analyzes the sign language video it receives using a generative AI model.

[0349] Input: Sign language video file uploaded to the server

[0350] Output: Text data generated from sign language actions

[0351] Specific operation: Analyzes sign language video frame by frame on the server and runs a generative AI model that converts sign language movements into text data.

[0352] Step 4:

[0353] The server uses an emotion engine to recognize the user's emotions and adjust the tone and nuance of the generated text.

[0354] Input: Generated text data, facial expressions and movement data from the sign language video

[0355] Output: Text data adjusted based on user sentiment

[0356] What it does: It uses an emotion engine to analyze facial expressions and movements in sign language videos and adjust the nuance and expression of the text based on the recognized emotion (e.g., joy).

[0357] Step 5:

[0358] The server returns the adjusted text data to the terminal.

[0359] Input: Adjusted text data

[0360] Output: Adjusted text data sent to the terminal

[0361] Specific operation: Generates text data in JSON format and sends it to the device.

[0362] Step 6:

[0363] The terminal displays the text data received from the server to the user.

[0364] Input: Text data sent from the server

[0365] Output: Text data displayed on the device screen

[0366] What it does: Displays text data in a text display field within an application so that it can be read by the user.

[0367] Converting text or dialogue from a foreign language into sign language video

[0368] Step 1:

[0369] The user enters text using the device's keyboard or voice using the microphone.

[0370] Input: International text and audio data

[0371] Output: Input data saved on the device

[0372] Specific operation: When inputting text, the user types text on the keyboard, and when inputting voice, the user speaks into the microphone and voice data is recorded.

[0373] Step 2:

[0374] The device sends the captured text or audio data to the server.

[0375] Input: Text or voice data stored on your device

[0376] Output: Text or audio data transmitted to the server

[0377] Specific operation: Pressing the "Submit" button in the application uploads data to the server.

[0378] Step 3:

[0379] The server analyzes the text or audio data, converting the audio to text as needed, and using a generative AI model to convert the international text into sign language actions.

[0380] Input: Text and voice data sent to the server

[0381] Output: Sign language gesture data generated from text in various languages

[0382] Specific actions: Use a speech recognition model to convert speech data into text, then use a generative AI model to convert the text into sign language actions.

[0383] Step 4:

[0384] The server uses an emotion engine to recognize the user's emotions and adjusts the expression of the sign language video.

[0385] Input: Text and voice data, user emotion information

[0386] Output: Emotion-reflecting sign language movement data

[0387] Specific actions: The emotion engine analyzes emotions from text and voice and generates sign language actions appropriate to those emotions.

[0388] Step 5:

[0389] The server generates sign language video based on the analyzed text data and emotional information.

[0390] Input: Emotion-reflecting sign language movement data

[0391] Output: Generated sign language video

[0392] Specific actions: Use sign language animation generation software to create sign language videos based on text and emotional information.

[0393] Step 6:

[0394] The terminal displays the sign language video data received from the server to the user.

[0395] Input: Sign language video data transferred from the server

[0396] Output: Sign language video displayed on the device screen

[0397] What it does: Uses a video player to play a sign language video for the user to review.

[0398] (Application example 2)

[0399] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0400] In modern society, communication between hearing-impaired and hearing-speaking people is extremely important, but adequate means are lacking. In particular, it is difficult for hearing-impaired people to communicate smoothly with store clerks in brick-and-mortar stores, and lack of understanding causes inconvenience to both parties. Existing sign language translation systems do not provide natural-sounding translations that reflect emotions, so more effective communication is needed.

[0401] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting sign language video, a means for analyzing the input sign language video and converting it into text in various languages, a means for recognizing the emotion of the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotion, and a means for displaying the converted text to the user. This enables hearing-impaired people to communicate naturally with store clerks using sign language in physical stores.

[0402] "Sign language video" is a recorded video of sign language used by the hearing impaired to communicate.

[0403] "National language text" is character data in a language used in a particular country or region.

[0404] "Analysis" is the process of breaking down input data and understanding its meaning and intent.

[0405] "Emotion recognition" means determining a person's feelings or emotional state from video and audio.

[0406] "Adjusting the tone" means modifying the way text or images are presented to suit the emotion or situation.

[0407] "Displaying to the user" means providing the translation results visually on the screen of a smartphone, tablet, etc.

[0408] A "server" is a computer system that processes and stores data.

[0409] The system that realizes this application example enables smooth communication between hearing-impaired people and store clerks, especially in brick-and-mortar stores. The system consists of a server, a terminal (a smartphone or tablet), and users (hearing-impaired people and store clerks).

[0410] The system includes the following means:

[0411] 1. How to input sign language video:

[0412] The user uses the device's camera to record a video of the sign language, which is then sent to the server for analysis.

[0413] 2. Means for analyzing input sign language video and converting it into text in various languages:

[0414] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the sign language video frame by frame and generate corresponding text in each language.

[0415] 3. A means for recognizing emotions in the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotions:

[0416] The server uses an emotion engine (e.g., Microsoft® Azure® Emotion API) to recognize emotions in the sign language video and adjust the tone and nuance of the generated text based on those emotions.

[0417] 4. How to display the converted text to the user:

[0418] The terminal displays the text returned by the server to the user.

[0419] 5. Ways to enter text in different languages:

[0420] The store clerk uses the terminal's keyboard to enter text or the microphone to input voice data, which is then sent to the server.

[0421] 6. Means for analyzing text in various languages ​​and converting it into a corresponding sign language video, and means for recognizing emotions in the generated sign language video and adjusting the expression of the sign language video based on the recognized emotions:

[0422] The server analyzes the input text and audio data and generates sign language videos using a generative AI model. An emotion engine then recognizes the emotions in the sign language videos and adjusts the sign language expressions accordingly.

[0423] 7. How to display the converted sign language video to the user:

[0424] The terminal displays the sign language video sent back from the server to the user.

[0425] For illustrative purposes, consider the following prompt:

[0426] Customer Prompt: Hello! I have a question about this product.

[0427] "Store Staff Prompt": Welcome. Is there anything I can help you with?

[0428] By inputting or speaking these prompts, the text or audio is sent to the server, where it is analyzed and an appropriate sign language video is generated. Based on the emotions analyzed by the emotion engine, a natural and emotive sign language video is displayed on the device, enabling smooth communication between the user and the store clerk.

[0429] The hardware used includes the camera and microphone of a smartphone or tablet, a server (equipped with a high-performance GPU) for data processing and storage, and software used includes a generative AI model, a speech recognition model (e.g., Google Speech-to-Text API), and an emotion analysis engine (e.g., Microsoft Azure Emotion API).

[0430] This embodiment is expected to facilitate smooth communication between hearing-impaired and hearing-savvy people in physical stores, improving service quality and customer satisfaction.

[0431] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0432] Step 1:

[0433] The user uses the device's camera to record sign language video. This sign language video is a question the hearing-impaired person wants to ask the store clerk or information they want to convey in sign language. The input sign language video is saved on the device.

[0434] Input: Sign language video

[0435] Output: Sign language video data saved on the device

[0436] Step 2:

[0437] The device sends the sign language video data to a cloud server, where the data is uploaded via the internet. The server receives the data.

[0438] Input: Sign language video data stored on the device

[0439] Output: Sign language video data uploaded to the server

[0440] Step 3:

[0441] The server uses a generative AI model to analyze the received sign language video frame by frame, converting the sign actions into text in each language.

[0442] Input: Sign language video data uploaded to the server

[0443] Output: Frame-by-frame sign language motion analysis data and text data in various languages

[0444] Step 4:

[0445] The server uses an emotion engine to recognize the user's emotions from the analyzed sign language video, which then adjusts the tone and nuance of the generated text in each language.

[0446] Input: Frame-by-frame sign language motion analysis data and text data in various languages

[0447] Output: Emotion-adjusted text data in various languages

[0448] Step 5:

[0449] The server returns the adjusted text in each language to the terminal, which then displays the received text data on its screen.

[0450] Input: Emotion-adjusted text data in various languages

[0451] Output: Adjusted text data displayed on the terminal

[0452] Step 6:

[0453] The store clerk inputs text using the terminal's keyboard or voice input using the microphone, which generates text data or voice data in various languages.

[0454] Input: Text input or voice input

[0455] Output: Text data or audio data in various languages

[0456] Step 7:

[0457] The terminal transmits the generated text data or voice data to the server, which receives the data.

[0458] Input: Text data or audio data in various languages

[0459] Output: Text data or audio data in various languages ​​uploaded to the server

[0460] Step 8:

[0461] The server uses a speech recognition model to convert speech data into text (if speech is input), and a generative AI model to convert text data in each language into the appropriate sign language gestures.

[0462] Input: (if necessary) audio data, text data in various languages

[0463] Output: Text data in various languages, sign language movement data

[0464] Step 9:

[0465] The server again uses the emotion engine to recognize the emotion of the sign language actions based on the input text and audio, and adjusts the expression of the sign language video.

[0466] Input: Sign language movement data

[0467] Output: Emotion-adjusted sign language video data

[0468] Step 10:

[0469] The server then sends the adjusted sign language video data back to the terminal, which then displays the received sign language video on its screen so that the store clerk can check the content.

[0470] Input: Emotion-adjusted sign language video data

[0471] Output: Adjusted sign language video displayed on the device

[0472] 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.

[0473] 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> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. 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 voice, text data indicating text, and image data indicating an image is also input. 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.

[0474] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0475] [Second embodiment]

[0476] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0477] 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.

[0478] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the 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).

[0479] 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.

[0480] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0481] 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 surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0482] 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.

[0483] 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.

[0484] 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 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.

[0485] 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.

[0486] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0487] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0488] This invention is a system that enables bidirectional translation between sign language and other languages. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[0489] When converting sign language videos into other languages

[0490] Program processing overview

[0491] 1. Input sign language video

[0492] The user uses the device to record a specific sign language video and upload it to the application.

[0493] 2. Send the sign language video to the server

[0494] The terminal transmits the sign language video data to the server.

[0495] 3. Analyzing sign language videos

[0496] The server receives the sign language video data and analyzes the video based on the generative AI model.

[0497] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[0498] 4. Generate the conversion results

[0499] The server returns to the terminal text in each language generated from the analyzed sign language movements.

[0500] 5. View the conversion results

[0501] The terminal displays the text data received from the server to the user.

[0502] Specific examples

[0503] A user can record a video of themselves speaking Japanese Sign Language on their smartphone and upload it through an application. This video is then sent over the internet to a server, which uses a generative AI model to analyze the sign language movements and convert them into Japanese text, such as "hello." This text is then sent back to the device and displayed to the user.

[0504] Converting text or dialogue from a foreign language into sign language video

[0505] Program processing overview

[0506] 1. Enter text or voice

[0507] The user can enter text using the device's keyboard or by speaking into the microphone.

[0508] 2. Send the input data to the server

[0509] The device sends the captured text or audio data to the server.

[0510] 3. Analyze the input data

[0511] The server receives the text or voice data and converts the voice to text as needed.

[0512] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[0513] 4. Generate sign language video

[0514] The server generates a corresponding sign language video based on the analyzed text data.

[0515] 5. Display sign language video

[0516] The terminal displays the sign language video data received from the server to the user.

[0517] Specific examples

[0518] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to a server, which uses a speech recognition model to convert the speech to text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video and sends it back to the device. Finally, the device displays the video of "Bonjour" in French Sign Language to the user.

[0519] As described above, the present invention realizes two-way translation between sign language and various languages ​​and functions as a system that supports communication between multiple languages.

[0520] The processing flow will be explained below.

[0521] When converting sign language videos into other languages

[0522] Program processing details

[0523] Step 1:

[0524] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[0525] Step 2:

[0526] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[0527] Step 3:

[0528] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[0529] Step 4:

[0530] The generative AI model analyzes the sign language movements for each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[0531] Step 5:

[0532] The server prepares the results of the analyzed sign language movements as text data and returns it to the terminal.

[0533] Step 6:

[0534] The terminal receives the text data in each language received from the server and displays it to the user. For example, the displayed text is "Hello."

[0535] Converting text or dialogue from a foreign language into sign language video

[0536] Program processing details

[0537] Step 1:

[0538] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[0539] Step 2:

[0540] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[0541] Step 3:

[0542] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[0543] Step 4:

[0544] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[0545] Step 5:

[0546] The server converts the generated sign language motion data into a sign language video, for example, generating a video of "Bonjour" in French sign language.

[0547] Step 6:

[0548] The server generates a response to send the generated sign language video data to the terminal, which receives the sign language video data via the Internet.

[0549] Step 7:

[0550] The device receives the sign language video data from the server and displays it to the user, who can then watch the video of "Bonjour" in French sign language on the device screen.

[0551] The above is a specific processing flow of the bidirectional translation system between sign language and other national languages ​​based on the present invention.

[0552] Example 1

[0553] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0554] Conventional translation systems between sign language and other languages ​​lack accuracy and real-time performance, making it difficult to facilitate smooth communication between multiple languages. Furthermore, while there is a need not only to convert sign language video into text in other languages, but also to convert text and audio in other languages ​​into sign language video, there are only a limited number of systems that can achieve this kind of two-way translation function. Therefore, there is a need for a system that can perform two-way translation with higher accuracy and in real time to support multilingual communication.

[0555] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0556] In this invention, the server includes means for a user to input sign language video, means for dividing the input sign language video and analyzing the hand position and shape of each frame, means for generating text in various languages ​​based on the analyzed sign language movements using a generative AI model, means for returning the converted text to the terminal, and means for displaying the converted text to the user, thereby enabling sign language video to be converted into text in various languages ​​with high accuracy and in real time.

[0557] "Sign language video" is video data that records communication using sign language.

[0558] "User" refers to a person who uses the system to input sign language video or input text or audio in various languages.

[0559] A "generative AI model" is a model that generates and converts text data and sign language motion data based on artificial intelligence. Examples include models that use natural language processing and computer vision techniques.

[0560] "Analysis" refers to the process of recognizing the hand position and shape for each frame of the input sign language video and converting it into text in each language based on that.

[0561] "Text" refers to character string data expressed in various languages, and refers to written language information.

[0562] A "terminal" is a device that allows a user to input sign language video, text, or audio, and receive responses from a server. Examples include smartphones and tablets.

[0563] A "server" is a computer system that analyzes input sign language video, text, and audio data, performs any necessary conversions, and then returns the results to the terminal.

[0564] The "Sign Language Animation Library" is a library for expressing sign language movements as animation using 3D motion capture data.

[0565] "Speech-to-text means" refers to any technology or function that analyzes and converts audio data into corresponding text data. An example is voice recognition software.

[0566] "Means for analyzing the position and shape of the hands in each frame" refers to technology or functions that divide sign language video into frames and recognize the position and shape of the hands in each frame.

[0567] "Means for generating sign language videos" refers to technologies and functions that generate new sign language videos based on analyzed sign language movements.

[0568] This invention is a system that realizes two-way translation between sign language and other languages, and is a system that mainly operates through the cooperation of three parties: a server, a terminal, and a user. This system provides functions to convert sign language videos into other languages, and to convert text and audio in other languages ​​into sign language videos.

[0569] A form of converting sign language videos into various languages

[0570] Users record sign language videos using devices such as smartphones. The recorded sign language videos are then sent to a server using a dedicated application on the device. The server then divides the received sign language videos and analyzes the hand position and shape for each frame using computer vision technology. Libraries such as TensorFlow and OpenCV are used for the analysis.

[0571] Based on the analyzed sign language movement data, a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) is used to generate text in the corresponding language. One example of a generative AI model that can be implemented is OpenAI's GPT-4.

[0572] The generated text in each language is sent back to the device in JSON format and displayed to the user by the device application. As a concrete example, a user records a video of themselves saying "hello" in Japanese sign language and uploads it to the application. This video is sent to a server, which analyzes the sign language movements and converts them into the Japanese text "hello". The converted text is then sent back to the device and displayed to the user.

[0573] A form of converting text or audio in various languages ​​into sign language video

[0574] Users can enter text in their native language using an application on their device, or use the microphone to input voice. The device then sends the entered text or voice data to the server. In the case of voice data, it is first converted into text using voice recognition software (e.g., Google's Speech-to-Text API).

[0575] The server uses a generative AI model to convert the input text into appropriate sign language actions, then uses a sign language animation library (e.g., a library using 3D motion capture data) to generate a sign language video, which is sent back from the server to the device and displayed to the user by the device application.

[0576] As a concrete example, if a user types or speaks "Bonjour" in French, the entered text or voice data is sent from the device to a server. The server uses speech recognition software to convert the speech into text and analyzes it with a generative AI model. Based on this data, the server generates a corresponding French Sign Language video and sends it back to the device. Finally, the sign language video is played in an application on the device.

[0577] Prompt Sentence Examples

[0578] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[0579] "Create a program that converts input text or audio into a sign language video and plays it back."

[0580] As described above, the present invention is a system that realizes bidirectional translation between sign language and various other languages ​​with high accuracy and in real time, thereby supporting communication between multiple languages.

[0581] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0582] When converting sign language videos into other languages

[0583] Processing steps and specific explanations

[0584] Step 1:

[0585] The user records a sign language video using the camera on their smartphone and presses the "upload" button in the dedicated application. The input is the recorded sign language video file, and the output is a video file sent from the device to the server.

[0586] Step 2:

[0587] The device sends the uploaded sign language video file to a server via the Internet. Specifically, the HTTPS protocol is used to encrypt data and ensure secure communication. The input is the sign language video file, and the output is the video data sent to the server.

[0588] Step 3:

[0589] The server temporarily stores the received sign language video and begins analyzing it using a generative AI model. The video is first divided into frames, and the hand position and shape of each frame are analyzed using computer vision technology (e.g., OpenCV, TensorFlow). The input is the sign language video stored on the server, and the output is analysis data including the hand position and shape.

[0590] Step 4:

[0591] The server uses a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) to generate text in the corresponding language based on the analyzed sign language movement data. For generation, OpenAI's GPT-4 is used, for example. The input is sign language movement data, and the output is text data in the language.

[0592] Step 5:

[0593] The server converts the generated text in each language into JSON format and returns it to the terminal. The input is the generated text data, and the output is JSON format text data.

[0594] Step 6:

[0595] The terminal parses the JSON-formatted text data received from the server and displays it in the user interface. Specifically, it displays the generated text in each language in a text view within the application. The input is JSON-formatted text data, and the output is the text displayed in the user interface.

[0596] Converting text or audio in any language into sign language video

[0597] Processing steps and specific explanations

[0598] Step 1:

[0599] The user opens the application on their device and enters the text they want to translate into the text field, or presses the voice input button and speaks into the microphone. The input is text or voice data, and the output is data sent from the device to the server.

[0600] Step 2:

[0601] The device sends text data or recorded voice data to the server. In the case of voice data, it is first saved as a voice file on the device and then sent to the server. The input is text data or voice data, and the output is the data sent to the server.

[0602] Step 3:

[0603] The server directly analyzes the received text data, and converts audio data into text using Google's Speech-to-Text API. The input is text data or audio data, and the output is the converted data.

[0604] Step 4:

[0605] The server uses a generative AI model to convert input text in various languages ​​into data corresponding to sign language actions. The input is text data, and the output is sign language action data.

[0606] Step 5:

[0607] The server generates sign language animation using a sign language animation library (e.g., a 3D motion capture library) based on the sign language movement data. The input is the sign language movement data, and the output is the generated sign language video.

[0608] Step 6:

[0609] The terminal plays the sign language video data received from the server on the user interface. The user can view the translated sign language video through the application. The input is sign language video data, and the output is a sign language video that the user can view.

[0610] Prompt Sentence Examples

[0611] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[0612] "Create a program that converts input text or audio into a sign language video and plays it back."

[0613] As described above, by clearly showing the specific processing steps and operations, it has been shown how the system of the present invention provides bidirectional translation between sign language and various other languages ​​with high accuracy and in real time.

[0614] (Application example 1)

[0615] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0616] It aims to address the lack of communication tools that allow sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies.

[0617] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0618] In this invention, the server includes means for analyzing input sign language video and converting it into text in various languages, means for transmitting the input sign language video as an emergency call, and means for converting reply text from the server into sign language video, thereby enabling sign language users to make emergency calls via sign language video and respond quickly.

[0619] "Sign language video" is information including hand movements and facial expressions recorded in video format so that people who use sign language can understand it visually.

[0620] "Text in a national language" refers to character strings or sentences written in a specific language, and is text data in a format that can be read and understood by humans.

[0621] "User" refers to a person who uses this system to input, transmit, and display sign language video.

[0622] A "server" is a computer system that analyzes and converts sign language videos and text and manages communications.

[0623] An "emergency call" is a communication act to warn or request help in an emergency, and refers to the transmission of information that requires a rapid response.

[0624] "Reply text" is a written string or sentence sent from a server or the like as a response to an emergency call.

[0625] "Analysis" refers to the process of understanding the content of sign language video or text and converting it into another format.

[0626] "Conversion" refers to the process of converting data of one format into another format, and in this invention it primarily refers to conversion between sign language video and text in a foreign language.

[0627] This invention is a system that allows sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[0628] Sign language video input and analysis

[0629] A user first uses a device (e.g., a smartphone) to record a video of an emergency situation in sign language. This video is then sent to a server via an application. The device then uploads the video to the server, using a generative AI model to convert it into text in each language.

[0630] Analysis of sign language video and conversion to text

[0631] The server receives the sign language video data and analyzes the video based on the generative AI model. This analysis converts the sign language actions into text in the corresponding language. The converted text is then sent from the server to the relevant authorities (e.g., police or security center) as an emergency call.

[0632] Sending and replying to emergency calls

[0633] When a reply from a relevant organization is sent to the server in text format, the server analyzes the text data and generates a corresponding sign language video. This sign language video is then sent to the device and displayed to the user. This allows the user to quickly report the situation through sign language and receive necessary assistance, even in an emergency.

[0634] Example

[0635] For example, suppose a user discovers a suspicious person at home and uses their smartphone to record a sign language video saying "Help me" and report it. This video is sent to a server and converted into text for the Japanese word "Help me." This text is then sent to the police, who reply, "We will send a patrol immediately." This reply is analyzed by the server, converted into the corresponding sign language video, and displayed on the user's device.

[0636] Hardware and software used

[0637] Hardware: Your smartphone's camera, microphone, and display.

[0638] Software: Server-side generative AI models, speech recognition models, and text analysis tools.

[0639] Specific names: requests, cv2 (OpenCV), and speech_recognition libraries implemented in Python.

[0640] Prompt Sentence Examples

[0641] "Convert the French word 'help' into the corresponding sign language action."

[0642] By inputting this prompt into a generative AI model, the necessary sign language actions are generated. In this way, the present invention enables sign language users to safely make emergency calls and receive prompt responses.

[0643] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0644] Step 1:

[0645] A user uses a smartphone to record a sign language video about an emergency (e.g., discovering a suspicious person at home). A video captured with the smartphone camera is generated as input data for the sign language video. The recorded sign language video is then saved by the user via an application.

[0646] Step 2:

[0647] The device sends the recorded sign language video data to a server. Specifically, an application on the smartphone uses an Internet connection to upload the sign language video to a specified API endpoint on the server. The input is a sign language video file, and the output is the data sent to the server.

[0648] Step 3:

[0649] The server analyzes the received sign language video data using a generative AI model. It identifies each frame of the video and analyzes hand movements and facial expressions. As a result of this analysis, the input sign language video is converted into natural language text (text in each language). The input is sign language video data, and the output is text in the corresponding language.

[0650] Step 4:

[0651] The server sends the parsed text in each language as an emergency call to the appropriate security agency (e.g., police). The text data is communicated through the server's API. The input is the generated text in each language, and the output is the call data to the security agency.

[0652] Step 5:

[0653] A security agency (e.g., the police) receives the call and creates a reply in the local language. This reply text is sent back to the server. Here, as a police operation, the input is the text of the emergency call and the output is the local language text of the reply.

[0654] Step 6:

[0655] The server receives the reply text data from the security agency. This text in each language is analyzed again using the generative AI model and converted into the corresponding sign language video. The input is the reply text data, and the output is the sign language video.

[0656] Step 7:

[0657] The server sends the generated sign language video to the device. The device then displays the video to the user. The input is the sign language video data from the server, and the output is the sign language video displayed on the smartphone.

[0658] Step 8:

[0659] The user checks the sign language video on the device and understands the reply from the security agency in sign language. The input is the sign language video, and the output is what the user understands.

[0660] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0661] This invention is a system that enables two-way translation between sign language and other languages, and also has the ability to adjust the translation results by recognizing the user's emotions. This system mainly works in cooperation with three entities: the server, the terminal, and the user, and by combining an emotion engine, it provides more natural and appropriate translations.

[0662] When converting sign language videos into other languages

[0663] Program processing overview

[0664] 1. Input sign language video

[0665] A user uses a device such as a smartphone or PC to record a specific sign language video and upload it to the application.

[0666] 2. Send the sign language video to the server

[0667] The terminal transmits the sign language video data to the server.

[0668] 3. Analyzing sign language videos

[0669] The server receives the sign language video data and analyzes the video based on the generative AI model.

[0670] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[0671] 4. Emotional Engine Adjustment

[0672] The server uses an emotion engine to recognize the user's emotions.

[0673] Adjust the tone and nuance of the generated text based on the perceived sentiment.

[0674] 5. Generate the conversion results

[0675] The server returns the adjusted text in each language to the terminal.

[0676] 6. View the conversion results

[0677] The terminal displays the text data received from the server to the user.

[0678] Specific examples

[0679] Suppose a user films a video of themselves speaking Japanese Sign Language on their smartphone and uploads it through an application. The video is sent over the internet to a server, which uses a generative AI model to analyze the sign language and convert it into Japanese text, such as "hello." The emotion engine then recognizes the user's emotion (e.g., joy) and adjusts the nuance of the text to "a very happy 'hello.'" This text is then sent back to the device and displayed to the user.

[0680] Converting text or dialogue from a foreign language into sign language video

[0681] Program processing overview

[0682] 1. Enter text or voice

[0683] The user can input text in various languages ​​using the device's keyboard or by voice input using the microphone.

[0684] 2. Send the input data to the server

[0685] The device sends the captured text or audio data to the server.

[0686] 3. Analyze the input data

[0687] The server receives the text or voice data and converts the voice to text as needed.

[0688] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[0689] 4. Emotional Engine Adjustment

[0690] The server uses an emotion engine to recognize the user's emotions.

[0691] The expression of sign language actions is adjusted based on the recognized emotion.

[0692] 5. Generate sign language video

[0693] The server generates a corresponding sign language video based on the analyzed text data and emotions.

[0694] 6. Display sign language video

[0695] The terminal displays the sign language video data received from the server to the user.

[0696] Specific examples

[0697] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to the server, which uses a speech recognition model to convert the speech into text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video, and an emotion engine recognizes the user's joy and reflects it in the sign language video. The sign language video sent back to the device visualizes "Bonjour" with a clear expression of joyful movements.

[0698] In this way, the combination of the emotion engine makes two-way translation between sign language and national languages ​​more natural and effective, further facilitating multilingual communication.

[0699] The processing flow will be explained below.

[0700] When converting sign language videos into other languages

[0701] Program processing details

[0702] Step 1:

[0703] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[0704] Step 2:

[0705] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[0706] Step 3:

[0707] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[0708] Step 4:

[0709] The generative AI model analyzes the sign language movements in each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[0710] Step 5:

[0711] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[0712] Step 6:

[0713] Based on the recognized emotion, the tone and nuance of the generated text in each language is adjusted. For example, if the user is happy, "Hello" is adjusted to "A very happy 'Hello'."

[0714] Step 7:

[0715] The server generates a request to return the adjusted text to the terminal and sends it over the Internet to the terminal.

[0716] Step 8:

[0717] The terminal receives the text data received from the server and displays it to the user. For example, the displayed text might be "A very happy 'hello'."

[0718] Converting text or dialogue from a foreign language into sign language video

[0719] Program processing details

[0720] Step 1:

[0721] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[0722] Step 2:

[0723] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[0724] Step 3:

[0725] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[0726] Step 4:

[0727] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[0728] Step 5:

[0729] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[0730] Step 6:

[0731] The expression of the sign language actions is adjusted based on the recognized emotion. For example, if the user is happy, the sign language actions are adjusted to be more cheerful and positive.

[0732] Step 7:

[0733] The server generates a sign language video based on the generated sign language movement data and the adjusted expression. For example, it creates a video that cheerfully expresses "Bonjour" in French sign language.

[0734] Step 8:

[0735] The server generates a request to transmit the generated sign language video data to the terminal, and transmits it to the terminal via the Internet.

[0736] Step 9:

[0737] The device receives the sign language video data from the server and displays it to the user. For example, a video of the French sign language "Bonjour" being displayed brightly on the device screen.

[0738] In this way, by combining the emotion engine, two-way translation between sign language and other languages ​​becomes more natural and emotional, providing users with a more effective means of communication.

[0739] Example 2

[0740] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0741] Conventional translation systems between sign language and other languages ​​often produce mechanical results without taking into account the user's emotions. As a result, the translation results cannot be said to support natural and appropriate communication, and there are problems with a lack of accuracy and expressiveness. This is particularly noticeable in communication where emotional nuances are important.

[0742] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0743] In this invention, the server includes a means for inputting a sign language video, a means for analyzing the input sign language video and converting it into text in a national language using a generative AI model, a means for recognizing emotions during the analysis and adjusting the tone and nuance of the generated text, and a means for displaying the adjusted text to the user. This enables natural and appropriate bidirectional translation between sign language and a national language that incorporates emotions.

[0744] "Sign language video" is video data that records the movements and gestures of a person using sign language.

[0745] A "generative AI model" is an artificial intelligence model that automatically generates text, images, etc. based on data.

[0746] A "server" is a computer system for storing, processing, and managing data.

[0747] A "terminal" is a device used by a user to input and display data, such as a smartphone or PC.

[0748] "User" refers to a person who uses this system to translate between sign language and other national languages.

[0749] "Analysis" is the process of breaking down and extracting data and finding meaning and patterns within it.

[0750] An "emotion engine" is a program that recognizes the user's emotions from text or video and adjusts the output results based on those emotions.

[0751] "Text" refers to data that consists of sentences or characters.

[0752] "Tone and nuance" refers to subtle changes and emotional resonance in words and expressions.

[0753] This invention provides a system that enables bidirectional translation between sign language and other languages. This system operates mainly through the cooperation of three entities: the server, the terminal, and the user, and provides more natural and appropriate translation by combining an emotion engine.

[0754] Hardware and Software Configuration

[0755] Device: The device on which users record and upload sign language videos. This can be a smartphone or PC.

[0756] Server: A computer system for data analysis, transformation, and emotion recognition.

[0757] Generative AI models: Use generative AI models (e.g., OpenAI's GPT-4) to convert sign language videos into text in various languages.

[0758] Emotion engine: An emotion engine (e.g., IBM Watson's Tone Analyzer) is used to recognize the user's emotions and adjust the tone and nuance of the translation results based on the results.

[0759] System Operation

[0760] When converting sign language videos into other languages

[0761] 1. Recording and uploading sign language videos

[0762] Users record specific sign language videos on their smartphones or PCs and upload them to the application, which then sends the video to the server.

[0763] 2. Analysis of sign language videos

[0764] The server analyzes the received sign language video, using a generative AI model to analyze the sign language movements for each frame of the video and generate corresponding text in each language.

[0765] 3. Recognizing emotions and adjusting to nuances

[0766] The server uses an emotion engine to recognize the user's emotions from the sign language video and adjust the tone and nuance of the generated text, making it more appropriate and natural-sounding.

[0767] 4. Returning and displaying text

[0768] The server returns the adjusted text data to the terminal, which displays it to the user.

[0769] Specific examples

[0770] A user records a video of themselves saying "hello" in Japanese Sign Language and uploads it through the application. The server analyzes the video and converts it into the Japanese text "hello." The emotion engine recognizes the user's emotion of joy and adjusts the nuance of the text to "a very happy 'hello.'" Finally, this text data is sent back to the user's device and displayed.

[0771] Example prompt sentence:

[0772] "A user has uploaded a video in Japanese Sign Language. Please translate the sign language in this video into Japanese and use the emotion engine to recognize that it contains the emotion 'joy' and adjust appropriately."

[0773] Converting text or dialogue from a foreign language into sign language video

[0774] 1. Text or voice input

[0775] The user inputs text using the device's keyboard or voice using the microphone, and this data is sent to the server.

[0776] 2. Text or speech analysis

[0777] The server analyzes the received text or audio data, converting audio to text as needed, and using generative AI models to convert international text into sign language actions.

[0778] 3. Emotion Recognition and Video Generation

[0779] The server uses an emotion engine to recognize the user's emotion from the input data and adjusts the expression of the sign language video, generating the sign language video based on this.

[0780] 4. Returning and displaying sign language videos

[0781] The server sends the generated sign language video back to the terminal, which displays it to the user.

[0782] Specific examples

[0783] The user types or speaks "Bonjour" in French. This data is sent to the server, which performs speech recognition and then analyzes it using a generative AI model. A sign language video is generated from the analyzed data, and an emotion engine recognizes the user's emotion of joy and reflects it in the video. This sign language video is then sent back to the user's device and displayed.

[0784] Example prompt sentence:

[0785] "A user types 'Bonjour' in French. Convert this text into a sign language video, and then use an emotion engine to recognize that it contains the emotion 'joy' and reflect that in the video."

[0786] In this way, the system enables natural bidirectional translation between sign language and other national languages.

[0787] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0788] When converting sign language videos into other languages

[0789] Step 1:

[0790] Users record specific sign language videos on their smartphones or PCs and upload them to the application.

[0791] Input: Sign language video file

[0792] Output: Video file saved on device

[0793] Specific operation: The user opens the application, presses the record button to record a sign language video, and when the recording is finished, the video file is saved on the device.

[0794] Step 2:

[0795] The terminal transmits the sign language video data to the server.

[0796] Input: Sign language video file saved on the device

[0797] Output: Sign language video file transferred to the server

[0798] Specific operation: When you press the "Send" button in the application, the video file is uploaded to the server via the Internet.

[0799] Step 3:

[0800] The server analyzes the sign language video it receives using a generative AI model.

[0801] Input: Sign language video file uploaded to the server

[0802] Output: Text data generated from sign language actions

[0803] Specific operation: Analyzes sign language video frame by frame on the server and runs a generative AI model that converts sign language movements into text data.

[0804] Step 4:

[0805] The server uses an emotion engine to recognize the user's emotions and adjust the tone and nuance of the generated text.

[0806] Input: Generated text data, facial expressions and movement data from the sign language video

[0807] Output: Text data adjusted based on user sentiment

[0808] What it does: It uses an emotion engine to analyze facial expressions and movements in sign language videos and adjust the nuance and expression of the text based on the recognized emotion (e.g., joy).

[0809] Step 5:

[0810] The server returns the adjusted text data to the terminal.

[0811] Input: Adjusted text data

[0812] Output: Adjusted text data sent to the terminal

[0813] Specific operation: Generates text data in JSON format and sends it to the device.

[0814] Step 6:

[0815] The terminal displays the text data received from the server to the user.

[0816] Input: Text data sent from the server

[0817] Output: Text data displayed on the device screen

[0818] What it does: Displays text data in a text display field within an application so that it can be read by the user.

[0819] Converting text or dialogue from a foreign language into sign language video

[0820] Step 1:

[0821] The user enters text using the device's keyboard or voice using the microphone.

[0822] Input: International text and audio data

[0823] Output: Input data saved on the device

[0824] Specific operation: When inputting text, the user types text on the keyboard, and when inputting voice, the user speaks into the microphone and voice data is recorded.

[0825] Step 2:

[0826] The device sends the captured text or audio data to the server.

[0827] Input: Text or voice data stored on your device

[0828] Output: Text or audio data transmitted to the server

[0829] Specific operation: Pressing the "Submit" button in the application uploads data to the server.

[0830] Step 3:

[0831] The server analyzes the text or audio data, converting the audio to text as needed, and using a generative AI model to convert the international text into sign language actions.

[0832] Input: Text and voice data sent to the server

[0833] Output: Sign language gesture data generated from text in various languages

[0834] Specific actions: Use a speech recognition model to convert speech data into text, then use a generative AI model to convert the text into sign language actions.

[0835] Step 4:

[0836] The server uses an emotion engine to recognize the user's emotions and adjusts the expression of the sign language video.

[0837] Input: Text and voice data, user emotion information

[0838] Output: Emotion-reflecting sign language movement data

[0839] Specific actions: The emotion engine analyzes emotions from text and voice and generates sign language actions appropriate to those emotions.

[0840] Step 5:

[0841] The server generates sign language video based on the analyzed text data and emotional information.

[0842] Input: Emotion-reflecting sign language movement data

[0843] Output: Generated sign language video

[0844] Specific actions: Use sign language animation generation software to create sign language videos based on text and emotional information.

[0845] Step 6:

[0846] The terminal displays the sign language video data received from the server to the user.

[0847] Input: Sign language video data transferred from the server

[0848] Output: Sign language video displayed on the device screen

[0849] What it does: Uses a video player to play a sign language video for the user to review.

[0850] (Application example 2)

[0851] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0852] In modern society, communication between hearing-impaired and hearing-speaking people is extremely important, but adequate means are lacking. In particular, it is difficult for hearing-impaired people to communicate smoothly with store clerks in brick-and-mortar stores, and lack of understanding causes inconvenience to both parties. Existing sign language translation systems do not provide natural-sounding translations that reflect emotions, so more effective communication is needed.

[0853] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting sign language video, a means for analyzing the input sign language video and converting it into text in various languages, a means for recognizing the emotion of the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotion, and a means for displaying the converted text to the user. This enables hearing-impaired people to communicate naturally with store clerks using sign language in physical stores.

[0854] "Sign language video" is a recorded video of sign language used by the hearing impaired to communicate.

[0855] "National language text" is character data in a language used in a particular country or region.

[0856] "Analysis" is the process of breaking down input data and understanding its meaning and intent.

[0857] "Emotion recognition" means determining a person's feelings or emotional state from video and audio.

[0858] "Adjusting the tone" means modifying the way text or images are presented to suit the emotion or situation.

[0859] "Displaying to the user" means providing the translation results visually on the screen of a smartphone, tablet, etc.

[0860] A "server" is a computer system that processes and stores data.

[0861] The system that realizes this application example enables smooth communication between hearing-impaired people and store clerks, especially in brick-and-mortar stores. The system consists of a server, a terminal (a smartphone or tablet), and users (hearing-impaired people and store clerks).

[0862] The system includes the following means:

[0863] 1. How to input sign language video:

[0864] The user uses the device's camera to record a video of the sign language, which is then sent to the server for analysis.

[0865] 2. Means for analyzing input sign language video and converting it into text in various languages:

[0866] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the sign language video frame by frame and generate corresponding text in each language.

[0867] 3. A means for recognizing emotions in the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotions:

[0868] The server uses an emotion engine (e.g., Microsoft Azure Emotion API) to recognize emotions in the sign language video and adjust the tone and nuance of the generated text based on those emotions.

[0869] 4. How to display the converted text to the user:

[0870] The terminal displays the text returned by the server to the user.

[0871] 5. Ways to enter text in different languages:

[0872] The store clerk uses the terminal's keyboard to enter text or the microphone to input voice data, which is then sent to the server.

[0873] 6. Means for analyzing text in various languages ​​and converting it into a corresponding sign language video, and means for recognizing emotions in the generated sign language video and adjusting the expression of the sign language video based on the recognized emotions:

[0874] The server analyzes the input text and audio data and generates sign language videos using a generative AI model. An emotion engine then recognizes the emotions in the sign language videos and adjusts the sign language expressions accordingly.

[0875] 7. How to display the converted sign language video to the user:

[0876] The terminal displays the sign language video sent back from the server to the user.

[0877] For illustrative purposes, consider the following prompt:

[0878] Customer Prompt: Hello! I have a question about this product.

[0879] "Store Staff Prompt": Welcome. Is there anything I can help you with?

[0880] By inputting or speaking these prompts, the text or audio is sent to the server, where it is analyzed and an appropriate sign language video is generated. Based on the emotions analyzed by the emotion engine, a natural and emotive sign language video is displayed on the device, enabling smooth communication between the user and the store clerk.

[0881] The hardware used includes the camera and microphone of a smartphone or tablet, a server (equipped with a high-performance GPU) for data processing and storage, and software used includes a generative AI model, a speech recognition model (e.g., Google Speech-to-Text API), and an emotion analysis engine (e.g., Microsoft Azure Emotion API).

[0882] This embodiment is expected to facilitate smooth communication between hearing-impaired and hearing-savvy people in physical stores, improving service quality and customer satisfaction.

[0883] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0884] Step 1:

[0885] The user uses the device's camera to record sign language video. This sign language video is a question the hearing-impaired person wants to ask the store clerk or information they want to convey in sign language. The input sign language video is saved on the device.

[0886] Input: Sign language video

[0887] Output: Sign language video data saved on the device

[0888] Step 2:

[0889] The device sends the sign language video data to a cloud server, where the data is uploaded via the internet. The server receives the data.

[0890] Input: Sign language video data stored on the device

[0891] Output: Sign language video data uploaded to the server

[0892] Step 3:

[0893] The server uses a generative AI model to analyze the received sign language video frame by frame, converting the sign actions into text in each language.

[0894] Input: Sign language video data uploaded to the server

[0895] Output: Frame-by-frame sign language motion analysis data and text data in various languages

[0896] Step 4:

[0897] The server uses an emotion engine to recognize the user's emotions from the analyzed sign language video, which then adjusts the tone and nuance of the generated text in each language.

[0898] Input: Frame-by-frame sign language motion analysis data and text data in various languages

[0899] Output: Emotion-adjusted text data in various languages

[0900] Step 5:

[0901] The server returns the adjusted text in each language to the terminal, which then displays the received text data on its screen.

[0902] Input: Emotion-adjusted text data in various languages

[0903] Output: Adjusted text data displayed on the terminal

[0904] Step 6:

[0905] The store clerk inputs text using the terminal's keyboard or voice input using the microphone, which generates text data or voice data in various languages.

[0906] Input: Text input or voice input

[0907] Output: Text data or audio data in various languages

[0908] Step 7:

[0909] The terminal transmits the generated text data or voice data to the server, which receives the data.

[0910] Input: Text data or audio data in various languages

[0911] Output: Text data or audio data in various languages ​​uploaded to the server

[0912] Step 8:

[0913] The server uses a speech recognition model to convert speech data into text (if speech is input), and a generative AI model to convert text data in each language into the appropriate sign language gestures.

[0914] Input: (if necessary) audio data, text data in various languages

[0915] Output: Text data in various languages, sign language movement data

[0916] Step 9:

[0917] The server again uses the emotion engine to recognize the emotion of the sign language actions based on the input text and audio, and adjusts the expression of the sign language video.

[0918] Input: Sign language movement data

[0919] Output: Emotion-adjusted sign language video data

[0920] Step 10:

[0921] The server then sends the adjusted sign language video data back to the terminal, which then displays the received sign language video on its screen so that the store clerk can check the content.

[0922] Input: Emotion-adjusted sign language video data

[0923] Output: Adjusted sign language video displayed on the device

[0924] 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.

[0925] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. 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 voice, text data indicating text, and image data indicating an image is also input. 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.

[0926] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0927] [Third embodiment]

[0928] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0929] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0930] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the 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).

[0931] 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.

[0932] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0933] 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 surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0934] 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.

[0935] 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.

[0936] 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 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.

[0937] 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.

[0938] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0939] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0940] This invention is a system that enables bidirectional translation between sign language and other languages. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[0941] When converting sign language videos into other languages

[0942] Program processing overview

[0943] 1. Input sign language video

[0944] The user uses the device to record a specific sign language video and upload it to the application.

[0945] 2. Send the sign language video to the server

[0946] The terminal transmits the sign language video data to the server.

[0947] 3. Analyzing sign language videos

[0948] The server receives the sign language video data and analyzes the video based on the generative AI model.

[0949] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[0950] 4. Generate the conversion results

[0951] The server returns to the terminal text in each language generated from the analyzed sign language movements.

[0952] 5. View the conversion results

[0953] The terminal displays the text data received from the server to the user.

[0954] Specific examples

[0955] A user can record a video of themselves speaking Japanese Sign Language on their smartphone and upload it through an application. This video is then sent over the internet to a server, which uses a generative AI model to analyze the sign language movements and convert them into Japanese text, such as "hello." This text is then sent back to the device and displayed to the user.

[0956] Converting text or dialogue from a foreign language into sign language video

[0957] Program processing overview

[0958] 1. Enter text or voice

[0959] The user can enter text using the device's keyboard or by speaking into the microphone.

[0960] 2. Send the input data to the server

[0961] The device sends the captured text or audio data to the server.

[0962] 3. Analyze the input data

[0963] The server receives the text or voice data and converts the voice to text as needed.

[0964] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[0965] 4. Generate sign language video

[0966] The server generates a corresponding sign language video based on the analyzed text data.

[0967] 5. Display sign language video

[0968] The terminal displays the sign language video data received from the server to the user.

[0969] Specific examples

[0970] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to a server, which uses a speech recognition model to convert the speech to text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video and sends it back to the device. Finally, the device displays the video of "Bonjour" in French Sign Language to the user.

[0971] As described above, the present invention realizes two-way translation between sign language and various languages ​​and functions as a system that supports communication between multiple languages.

[0972] The processing flow will be explained below.

[0973] When converting sign language videos into other languages

[0974] Program processing details

[0975] Step 1:

[0976] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[0977] Step 2:

[0978] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[0979] Step 3:

[0980] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[0981] Step 4:

[0982] The generative AI model analyzes the sign language movements for each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[0983] Step 5:

[0984] The server prepares the results of the analyzed sign language movements as text data and returns it to the terminal.

[0985] Step 6:

[0986] The terminal receives the text data in each language received from the server and displays it to the user. For example, the displayed text is "Hello."

[0987] Converting text or dialogue from a foreign language into sign language video

[0988] Program processing details

[0989] Step 1:

[0990] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[0991] Step 2:

[0992] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[0993] Step 3:

[0994] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[0995] Step 4:

[0996] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[0997] Step 5:

[0998] The server converts the generated sign language motion data into a sign language video, for example, generating a video of "Bonjour" in French sign language.

[0999] Step 6:

[1000] The server generates a response to send the generated sign language video data to the terminal, which receives the sign language video data via the Internet.

[1001] Step 7:

[1002] The device receives the sign language video data from the server and displays it to the user, who can then watch the video of "Bonjour" in French sign language on the device screen.

[1003] The above is a specific processing flow of the bidirectional translation system between sign language and other national languages ​​based on the present invention.

[1004] Example 1

[1005] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1006] Conventional translation systems between sign language and other languages ​​lack accuracy and real-time performance, making it difficult to facilitate smooth communication between multiple languages. Furthermore, while there is a need not only to convert sign language video into text in other languages, but also to convert text and audio in other languages ​​into sign language video, there are only a limited number of systems that can achieve this kind of two-way translation function. Therefore, there is a need for a system that can perform two-way translation with higher accuracy and in real time to support multilingual communication.

[1007] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1008] In this invention, the server includes means for a user to input sign language video, means for dividing the input sign language video and analyzing the hand position and shape of each frame, means for generating text in various languages ​​based on the analyzed sign language movements using a generative AI model, means for returning the converted text to the terminal, and means for displaying the converted text to the user, thereby enabling sign language video to be converted into text in various languages ​​with high accuracy and in real time.

[1009] "Sign language video" is video data that records communication using sign language.

[1010] "User" refers to a person who uses the system to input sign language video or input text or audio in various languages.

[1011] A "generative AI model" is a model that generates and converts text data and sign language motion data based on artificial intelligence. Examples include models that use natural language processing and computer vision techniques.

[1012] "Analysis" refers to the process of recognizing the hand position and shape for each frame of the input sign language video and converting it into text in each language based on that.

[1013] "Text" refers to character string data expressed in various languages, and refers to written language information.

[1014] A "terminal" is a device that allows a user to input sign language video, text, or audio, and receive responses from a server. Examples include smartphones and tablets.

[1015] A "server" is a computer system that analyzes input sign language video, text, and audio data, performs any necessary conversions, and then returns the results to the terminal.

[1016] The "Sign Language Animation Library" is a library for expressing sign language movements as animation using 3D motion capture data.

[1017] "Speech-to-text means" refers to any technology or function that analyzes and converts audio data into corresponding text data. An example is voice recognition software.

[1018] "Means for analyzing the position and shape of the hands in each frame" refers to technology or functions that divide sign language video into frames and recognize the position and shape of the hands in each frame.

[1019] "Means for generating sign language videos" refers to technologies and functions that generate new sign language videos based on analyzed sign language movements.

[1020] This invention is a system that realizes two-way translation between sign language and other languages, and is a system that mainly operates through the cooperation of three parties: a server, a terminal, and a user. This system provides functions to convert sign language videos into other languages, and to convert text and audio in other languages ​​into sign language videos.

[1021] A form of converting sign language videos into various languages

[1022] Users record sign language videos using devices such as smartphones. The recorded sign language videos are then sent to a server using a dedicated application on the device. The server then divides the received sign language videos and analyzes the hand position and shape for each frame using computer vision technology. Libraries such as TensorFlow and OpenCV are used for the analysis.

[1023] Based on the analyzed sign language movement data, a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) is used to generate text in the corresponding language. One example of a generative AI model that can be implemented is OpenAI's GPT-4.

[1024] The generated text in each language is sent back to the device in JSON format and displayed to the user by the device application. As a concrete example, a user records a video of themselves saying "hello" in Japanese sign language and uploads it to the application. This video is sent to a server, which analyzes the sign language movements and converts them into the Japanese text "hello". The converted text is then sent back to the device and displayed to the user.

[1025] A form of converting text or audio in various languages ​​into sign language video

[1026] Users can enter text in their native language using an application on their device, or use the microphone to input voice. The device then sends the entered text or voice data to the server. In the case of voice data, it is first converted into text using voice recognition software (e.g., Google's Speech-to-Text API).

[1027] The server uses a generative AI model to convert the input text into appropriate sign language actions, then uses a sign language animation library (e.g., a library using 3D motion capture data) to generate a sign language video, which is sent back from the server to the device and displayed to the user by the device application.

[1028] As a concrete example, if a user types or speaks "Bonjour" in French, the entered text or voice data is sent from the device to a server. The server uses speech recognition software to convert the speech into text and analyzes it with a generative AI model. Based on this data, the server generates a corresponding French Sign Language video and sends it back to the device. Finally, the sign language video is played in an application on the device.

[1029] Prompt Sentence Examples

[1030] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[1031] "Create a program that converts input text or audio into a sign language video and plays it back."

[1032] As described above, the present invention is a system that realizes bidirectional translation between sign language and various other languages ​​with high accuracy and in real time, thereby supporting communication between multiple languages.

[1033] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1034] When converting sign language videos into other languages

[1035] Processing steps and specific explanations

[1036] Step 1:

[1037] The user records a sign language video using the camera on their smartphone and presses the "upload" button in the dedicated application. The input is the recorded sign language video file, and the output is a video file sent from the device to the server.

[1038] Step 2:

[1039] The device sends the uploaded sign language video file to a server via the Internet. Specifically, the HTTPS protocol is used to encrypt data and ensure secure communication. The input is the sign language video file, and the output is the video data sent to the server.

[1040] Step 3:

[1041] The server temporarily stores the received sign language video and begins analyzing it using a generative AI model. The video is first divided into frames, and the hand position and shape of each frame are analyzed using computer vision technology (e.g., OpenCV, TensorFlow). The input is the sign language video stored on the server, and the output is analysis data including the hand position and shape.

[1042] Step 4:

[1043] The server uses a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) to generate text in the corresponding language based on the analyzed sign language movement data. For generation, OpenAI's GPT-4 is used, for example. The input is sign language movement data, and the output is text data in the language.

[1044] Step 5:

[1045] The server converts the generated text in each language into JSON format and returns it to the terminal. The input is the generated text data, and the output is JSON format text data.

[1046] Step 6:

[1047] The terminal parses the JSON-formatted text data received from the server and displays it in the user interface. Specifically, it displays the generated text in each language in a text view within the application. The input is JSON-formatted text data, and the output is the text displayed in the user interface.

[1048] Converting text or audio in any language into sign language video

[1049] Processing steps and specific explanations

[1050] Step 1:

[1051] The user opens the application on their device and enters the text they want to translate into the text field, or presses the voice input button and speaks into the microphone. The input is text or voice data, and the output is data sent from the device to the server.

[1052] Step 2:

[1053] The device sends text data or recorded voice data to the server. In the case of voice data, it is first saved as a voice file on the device and then sent to the server. The input is text data or voice data, and the output is the data sent to the server.

[1054] Step 3:

[1055] The server directly analyzes the received text data, and converts audio data into text using Google's Speech-to-Text API. The input is text data or audio data, and the output is the converted data.

[1056] Step 4:

[1057] The server uses a generative AI model to convert input text in various languages ​​into data corresponding to sign language actions. The input is text data, and the output is sign language action data.

[1058] Step 5:

[1059] The server generates sign language animation using a sign language animation library (e.g., a 3D motion capture library) based on the sign language movement data. The input is the sign language movement data, and the output is the generated sign language video.

[1060] Step 6:

[1061] The terminal plays the sign language video data received from the server on the user interface. The user can view the translated sign language video through the application. The input is sign language video data, and the output is a sign language video that the user can view.

[1062] Prompt Sentence Examples

[1063] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[1064] "Create a program that converts input text or audio into a sign language video and plays it back."

[1065] As described above, by clearly showing the specific processing steps and operations, it has been shown how the system of the present invention provides bidirectional translation between sign language and various other languages ​​with high accuracy and in real time.

[1066] (Application example 1)

[1067] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1068] It aims to address the lack of communication tools that allow sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies.

[1069] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1070] In this invention, the server includes means for analyzing input sign language video and converting it into text in various languages, means for transmitting the input sign language video as an emergency call, and means for converting reply text from the server into sign language video, thereby enabling sign language users to make emergency calls via sign language video and respond quickly.

[1071] "Sign language video" is information including hand movements and facial expressions recorded in video format so that people who use sign language can understand it visually.

[1072] "Text in a national language" refers to character strings or sentences written in a specific language, and is text data in a format that can be read and understood by humans.

[1073] "User" refers to a person who uses this system to input, transmit, and display sign language video.

[1074] A "server" is a computer system that analyzes and converts sign language videos and text and manages communications.

[1075] An "emergency call" is a communication act to warn or request help in an emergency, and refers to the transmission of information that requires a rapid response.

[1076] "Reply text" is a written string or sentence sent from a server or the like as a response to an emergency call.

[1077] "Analysis" refers to the process of understanding the content of sign language video or text and converting it into another format.

[1078] "Conversion" refers to the process of converting data of one format into another format, and in this invention it primarily refers to conversion between sign language video and text in a foreign language.

[1079] This invention is a system that allows sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[1080] Sign language video input and analysis

[1081] A user first uses a device (e.g., a smartphone) to record a video of an emergency situation in sign language. This video is then sent to a server via an application. The device then uploads the video to the server, using a generative AI model to convert it into text in each language.

[1082] Analysis of sign language video and conversion to text

[1083] The server receives the sign language video data and analyzes the video based on the generative AI model. This analysis converts the sign language actions into text in the corresponding language. The converted text is then sent from the server to the relevant authorities (e.g., police or security center) as an emergency call.

[1084] Sending and replying to emergency calls

[1085] When a reply from a relevant organization is sent to the server in text format, the server analyzes the text data and generates a corresponding sign language video. This sign language video is then sent to the device and displayed to the user. This allows the user to quickly report the situation through sign language and receive necessary assistance, even in an emergency.

[1086] Example

[1087] For example, suppose a user discovers a suspicious person at home and uses their smartphone to record a sign language video saying "Help me" and report it. This video is sent to a server and converted into text for the Japanese word "Help me." This text is then sent to the police, who reply, "We will send a patrol immediately." This reply is analyzed by the server, converted into the corresponding sign language video, and displayed on the user's device.

[1088] Hardware and software used

[1089] Hardware: Your smartphone's camera, microphone, and display.

[1090] Software: Server-side generative AI models, speech recognition models, and text analysis tools.

[1091] Specific names: requests, cv2 (OpenCV), and speech_recognition libraries implemented in Python.

[1092] Prompt Sentence Examples

[1093] "Convert the French word 'help' into the corresponding sign language action."

[1094] By inputting this prompt into a generative AI model, the necessary sign language actions are generated. In this way, the present invention enables sign language users to safely make emergency calls and receive prompt responses.

[1095] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1096] Step 1:

[1097] A user uses a smartphone to record a sign language video about an emergency (e.g., discovering a suspicious person at home). A video captured with the smartphone camera is generated as input data for the sign language video. The recorded sign language video is then saved by the user via an application.

[1098] Step 2:

[1099] The device sends the recorded sign language video data to a server. Specifically, an application on the smartphone uses an Internet connection to upload the sign language video to a specified API endpoint on the server. The input is a sign language video file, and the output is the data sent to the server.

[1100] Step 3:

[1101] The server analyzes the received sign language video data using a generative AI model. It identifies each frame of the video and analyzes hand movements and facial expressions. As a result of this analysis, the input sign language video is converted into natural language text (text in each language). The input is sign language video data, and the output is text in the corresponding language.

[1102] Step 4:

[1103] The server sends the parsed text in each language as an emergency call to the appropriate security agency (e.g., police). The text data is communicated through the server's API. The input is the generated text in each language, and the output is the call data to the security agency.

[1104] Step 5:

[1105] A security agency (e.g., the police) receives the call and creates a reply in the local language. This reply text is sent back to the server. Here, as a police operation, the input is the text of the emergency call and the output is the local language text of the reply.

[1106] Step 6:

[1107] The server receives the reply text data from the security agency. This text in each language is analyzed again using the generative AI model and converted into the corresponding sign language video. The input is the reply text data, and the output is the sign language video.

[1108] Step 7:

[1109] The server sends the generated sign language video to the device. The device then displays the video to the user. The input is the sign language video data from the server, and the output is the sign language video displayed on the smartphone.

[1110] Step 8:

[1111] The user checks the sign language video on the device and understands the reply from the security agency in sign language. The input is the sign language video, and the output is what the user understands.

[1112] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1113] This invention is a system that enables two-way translation between sign language and other languages, and also has the ability to adjust the translation results by recognizing the user's emotions. This system mainly works in cooperation with three entities: the server, the terminal, and the user, and by combining an emotion engine, it provides more natural and appropriate translations.

[1114] When converting sign language videos into other languages

[1115] Program processing overview

[1116] 1. Input sign language video

[1117] A user uses a device such as a smartphone or PC to record a specific sign language video and upload it to the application.

[1118] 2. Send the sign language video to the server

[1119] The terminal transmits the sign language video data to the server.

[1120] 3. Analyzing sign language videos

[1121] The server receives the sign language video data and analyzes the video based on the generative AI model.

[1122] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[1123] 4. Emotional Engine Adjustment

[1124] The server uses an emotion engine to recognize the user's emotions.

[1125] Adjust the tone and nuance of the generated text based on the perceived sentiment.

[1126] 5. Generate the conversion results

[1127] The server returns the adjusted text in each language to the terminal.

[1128] 6. View the conversion results

[1129] The terminal displays the text data received from the server to the user.

[1130] Specific examples

[1131] Suppose a user films a video of themselves speaking Japanese Sign Language on their smartphone and uploads it through an application. The video is sent over the internet to a server, which uses a generative AI model to analyze the sign language and convert it into Japanese text, such as "hello." The emotion engine then recognizes the user's emotion (e.g., joy) and adjusts the nuance of the text to "a very happy 'hello.'" This text is then sent back to the device and displayed to the user.

[1132] Converting text or dialogue from a foreign language into sign language video

[1133] Program processing overview

[1134] 1. Enter text or voice

[1135] The user can input text in various languages ​​using the device's keyboard or by voice input using the microphone.

[1136] 2. Send the input data to the server

[1137] The device sends the captured text or audio data to the server.

[1138] 3. Analyze the input data

[1139] The server receives the text or voice data and converts the voice to text as needed.

[1140] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[1141] 4. Emotional Engine Adjustment

[1142] The server uses an emotion engine to recognize the user's emotions.

[1143] The expression of sign language actions is adjusted based on the recognized emotion.

[1144] 5. Generate sign language video

[1145] The server generates a corresponding sign language video based on the analyzed text data and emotions.

[1146] 6. Display sign language video

[1147] The terminal displays the sign language video data received from the server to the user.

[1148] Specific examples

[1149] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to the server, which uses a speech recognition model to convert the speech into text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video, and an emotion engine recognizes the user's joy and reflects it in the sign language video. The sign language video sent back to the device visualizes "Bonjour" with a clear expression of joyful movements.

[1150] In this way, the combination of the emotion engine makes two-way translation between sign language and national languages ​​more natural and effective, further facilitating multilingual communication.

[1151] The processing flow will be explained below.

[1152] When converting sign language videos into other languages

[1153] Program processing details

[1154] Step 1:

[1155] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[1156] Step 2:

[1157] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[1158] Step 3:

[1159] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[1160] Step 4:

[1161] The generative AI model analyzes the sign language movements in each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[1162] Step 5:

[1163] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[1164] Step 6:

[1165] Based on the recognized emotion, the tone and nuance of the generated text in each language is adjusted. For example, if the user is happy, "Hello" is adjusted to "A very happy 'Hello'."

[1166] Step 7:

[1167] The server generates a request to return the adjusted text to the terminal and sends it over the Internet to the terminal.

[1168] Step 8:

[1169] The terminal receives the text data received from the server and displays it to the user. For example, the displayed text might be "A very happy 'hello'."

[1170] Converting text or dialogue from a foreign language into sign language video

[1171] Program processing details

[1172] Step 1:

[1173] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[1174] Step 2:

[1175] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[1176] Step 3:

[1177] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[1178] Step 4:

[1179] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[1180] Step 5:

[1181] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[1182] Step 6:

[1183] The expression of the sign language actions is adjusted based on the recognized emotion. For example, if the user is happy, the sign language actions are adjusted to be more cheerful and positive.

[1184] Step 7:

[1185] The server generates a sign language video based on the generated sign language movement data and the adjusted expression. For example, it creates a video that cheerfully expresses "Bonjour" in French sign language.

[1186] Step 8:

[1187] The server generates a request to transmit the generated sign language video data to the terminal, and transmits it to the terminal via the Internet.

[1188] Step 9:

[1189] The device receives the sign language video data from the server and displays it to the user. For example, a video of the French sign language "Bonjour" being displayed brightly on the device screen.

[1190] In this way, by combining the emotion engine, two-way translation between sign language and other languages ​​becomes more natural and emotional, providing users with a more effective means of communication.

[1191] Example 2

[1192] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1193] Conventional translation systems between sign language and other languages ​​often produce mechanical results without taking into account the user's emotions. As a result, the translation results cannot be said to support natural and appropriate communication, and there are problems with a lack of accuracy and expressiveness. This is particularly noticeable in communication where emotional nuances are important.

[1194] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1195] In this invention, the server includes a means for inputting a sign language video, a means for analyzing the input sign language video and converting it into text in a national language using a generative AI model, a means for recognizing emotions during the analysis and adjusting the tone and nuance of the generated text, and a means for displaying the adjusted text to the user. This enables natural and appropriate bidirectional translation between sign language and a national language that incorporates emotions.

[1196] "Sign language video" is video data that records the movements and gestures of a person using sign language.

[1197] A "generative AI model" is an artificial intelligence model that automatically generates text, images, etc. based on data.

[1198] A "server" is a computer system for storing, processing, and managing data.

[1199] A "terminal" is a device used by a user to input and display data, such as a smartphone or PC.

[1200] "User" refers to a person who uses this system to translate between sign language and other national languages.

[1201] "Analysis" is the process of breaking down and extracting data and finding meaning and patterns within it.

[1202] An "emotion engine" is a program that recognizes the user's emotions from text or video and adjusts the output results based on those emotions.

[1203] "Text" refers to data that consists of sentences or characters.

[1204] "Tone and nuance" refers to subtle changes and emotional resonance in words and expressions.

[1205] This invention provides a system that enables bidirectional translation between sign language and other languages. This system operates mainly through the cooperation of three entities: the server, the terminal, and the user, and provides more natural and appropriate translation by combining an emotion engine.

[1206] Hardware and Software Configuration

[1207] Device: The device on which users record and upload sign language videos. This can be a smartphone or PC.

[1208] Server: A computer system for data analysis, transformation, and emotion recognition.

[1209] Generative AI models: Use generative AI models (e.g., OpenAI's GPT-4) to convert sign language videos into text in various languages.

[1210] Emotion engine: An emotion engine (e.g., IBM Watson's Tone Analyzer) is used to recognize the user's emotions and adjust the tone and nuance of the translation results based on the results.

[1211] System Operation

[1212] When converting sign language videos into other languages

[1213] 1. Recording and uploading sign language videos

[1214] Users record specific sign language videos on their smartphones or PCs and upload them to the application, which then sends the video to the server.

[1215] 2. Analysis of sign language videos

[1216] The server analyzes the received sign language video, using a generative AI model to analyze the sign language movements for each frame of the video and generate corresponding text in each language.

[1217] 3. Recognizing emotions and adjusting to nuances

[1218] The server uses an emotion engine to recognize the user's emotions from the sign language video and adjust the tone and nuance of the generated text, making it more appropriate and natural-sounding.

[1219] 4. Returning and displaying text

[1220] The server returns the adjusted text data to the terminal, which displays it to the user.

[1221] Specific examples

[1222] A user records a video of themselves saying "hello" in Japanese Sign Language and uploads it through the application. The server analyzes the video and converts it into the Japanese text "hello." The emotion engine recognizes the user's emotion of joy and adjusts the nuance of the text to "a very happy 'hello.'" Finally, this text data is sent back to the user's device and displayed.

[1223] Example prompt sentence:

[1224] "A user has uploaded a video in Japanese Sign Language. Please translate the sign language in this video into Japanese and use the emotion engine to recognize that it contains the emotion 'joy' and adjust appropriately."

[1225] Converting text or dialogue from a foreign language into sign language video

[1226] 1. Text or voice input

[1227] The user inputs text using the device's keyboard or voice using the microphone, and this data is sent to the server.

[1228] 2. Text or speech analysis

[1229] The server analyzes the received text or audio data, converting audio to text as needed, and using generative AI models to convert international text into sign language actions.

[1230] 3. Emotion Recognition and Video Generation

[1231] The server uses an emotion engine to recognize the user's emotion from the input data and adjusts the expression of the sign language video, generating the sign language video based on this.

[1232] 4. Returning and displaying sign language videos

[1233] The server sends the generated sign language video back to the terminal, which displays it to the user.

[1234] Specific examples

[1235] The user types or speaks "Bonjour" in French. This data is sent to the server, which performs speech recognition and then analyzes it using a generative AI model. A sign language video is generated from the analyzed data, and an emotion engine recognizes the user's emotion of joy and reflects it in the video. This sign language video is then sent back to the user's device and displayed.

[1236] Example prompt sentence:

[1237] "A user types 'Bonjour' in French. Convert this text into a sign language video, and then use an emotion engine to recognize that it contains the emotion 'joy' and reflect that in the video."

[1238] In this way, the system enables natural bidirectional translation between sign language and other national languages.

[1239] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1240] When converting sign language videos into other languages

[1241] Step 1:

[1242] Users record specific sign language videos on their smartphones or PCs and upload them to the application.

[1243] Input: Sign language video file

[1244] Output: Video file saved on device

[1245] Specific operation: The user opens the application, presses the record button to record a sign language video, and when the recording is finished, the video file is saved on the device.

[1246] Step 2:

[1247] The terminal transmits the sign language video data to the server.

[1248] Input: Sign language video file saved on the device

[1249] Output: Sign language video file transferred to the server

[1250] Specific operation: When you press the "Send" button in the application, the video file is uploaded to the server via the Internet.

[1251] Step 3:

[1252] The server analyzes the sign language video it receives using a generative AI model.

[1253] Input: Sign language video file uploaded to the server

[1254] Output: Text data generated from sign language actions

[1255] Specific operation: Analyzes sign language video frame by frame on the server and runs a generative AI model that converts sign language movements into text data.

[1256] Step 4:

[1257] The server uses an emotion engine to recognize the user's emotions and adjust the tone and nuance of the generated text.

[1258] Input: Generated text data, facial expressions and movement data from the sign language video

[1259] Output: Text data adjusted based on user sentiment

[1260] What it does: It uses an emotion engine to analyze facial expressions and movements in sign language videos and adjust the nuance and expression of the text based on the recognized emotion (e.g., joy).

[1261] Step 5:

[1262] The server returns the adjusted text data to the terminal.

[1263] Input: Adjusted text data

[1264] Output: Adjusted text data sent to the terminal

[1265] Specific operation: Generates text data in JSON format and sends it to the device.

[1266] Step 6:

[1267] The terminal displays the text data received from the server to the user.

[1268] Input: Text data sent from the server

[1269] Output: Text data displayed on the device screen

[1270] What it does: Displays text data in a text display field within an application so that it can be read by the user.

[1271] Converting text or dialogue from a foreign language into sign language video

[1272] Step 1:

[1273] The user enters text using the device's keyboard or voice using the microphone.

[1274] Input: International text and audio data

[1275] Output: Input data saved on the device

[1276] Specific operation: When inputting text, the user types text on the keyboard, and when inputting voice, the user speaks into the microphone and voice data is recorded.

[1277] Step 2:

[1278] The device sends the captured text or audio data to the server.

[1279] Input: Text or voice data stored on your device

[1280] Output: Text or audio data transmitted to the server

[1281] Specific operation: Pressing the "Submit" button in the application uploads data to the server.

[1282] Step 3:

[1283] The server analyzes the text or audio data, converting the audio to text as needed, and using a generative AI model to convert the international text into sign language actions.

[1284] Input: Text and voice data sent to the server

[1285] Output: Sign language gesture data generated from text in various languages

[1286] Specific actions: Use a speech recognition model to convert speech data into text, then use a generative AI model to convert the text into sign language actions.

[1287] Step 4:

[1288] The server uses an emotion engine to recognize the user's emotions and adjusts the expression of the sign language video.

[1289] Input: Text and voice data, user emotion information

[1290] Output: Emotion-reflecting sign language movement data

[1291] Specific actions: The emotion engine analyzes emotions from text and voice and generates sign language actions appropriate to those emotions.

[1292] Step 5:

[1293] The server generates sign language video based on the analyzed text data and emotional information.

[1294] Input: Emotion-reflecting sign language movement data

[1295] Output: Generated sign language video

[1296] Specific actions: Use sign language animation generation software to create sign language videos based on text and emotional information.

[1297] Step 6:

[1298] The terminal displays the sign language video data received from the server to the user.

[1299] Input: Sign language video data transferred from the server

[1300] Output: Sign language video displayed on the device screen

[1301] What it does: Uses a video player to play a sign language video for the user to review.

[1302] (Application example 2)

[1303] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1304] In modern society, communication between hearing-impaired and hearing-speaking people is extremely important, but adequate means are lacking. In particular, it is difficult for hearing-impaired people to communicate smoothly with store clerks in brick-and-mortar stores, and lack of understanding causes inconvenience to both parties. Existing sign language translation systems do not provide natural-sounding translations that reflect emotions, so more effective communication is needed.

[1305] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting sign language video, a means for analyzing the input sign language video and converting it into text in various languages, a means for recognizing the emotion of the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotion, and a means for displaying the converted text to the user. This enables hearing-impaired people to communicate naturally with store clerks using sign language in physical stores.

[1306] "Sign language video" is a recorded video of sign language used by the hearing impaired to communicate.

[1307] "National language text" is character data in a language used in a particular country or region.

[1308] "Analysis" is the process of breaking down input data and understanding its meaning and intent.

[1309] "Emotion recognition" means determining a person's feelings or emotional state from video and audio.

[1310] "Adjusting the tone" means modifying the way text or images are presented to suit the emotion or situation.

[1311] "Displaying to the user" means providing the translation results visually on the screen of a smartphone, tablet, etc.

[1312] A "server" is a computer system that processes and stores data.

[1313] The system that realizes this application example enables smooth communication between hearing-impaired people and store clerks, especially in brick-and-mortar stores. The system consists of a server, a terminal (a smartphone or tablet), and users (hearing-impaired people and store clerks).

[1314] The system includes the following means:

[1315] 1. How to input sign language video:

[1316] The user uses the device's camera to record a video of the sign language, which is then sent to the server for analysis.

[1317] 2. Means for analyzing input sign language video and converting it into text in various languages:

[1318] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the sign language video frame by frame and generate corresponding text in each language.

[1319] 3. A means for recognizing emotions in the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotions:

[1320] The server uses an emotion engine (e.g., Microsoft Azure Emotion API) to recognize emotions in the sign language video and adjust the tone and nuance of the generated text based on those emotions.

[1321] 4. How to display the converted text to the user:

[1322] The terminal displays the text returned by the server to the user.

[1323] 5. Ways to enter text in different languages:

[1324] The store clerk uses the terminal's keyboard to enter text or the microphone to input voice data, which is then sent to the server.

[1325] 6. Means for analyzing text in various languages ​​and converting it into a corresponding sign language video, and means for recognizing emotions in the generated sign language video and adjusting the expression of the sign language video based on the recognized emotions:

[1326] The server analyzes the input text and audio data and generates sign language videos using a generative AI model. An emotion engine then recognizes the emotions in the sign language videos and adjusts the sign language expressions accordingly.

[1327] 7. How to display the converted sign language video to the user:

[1328] The terminal displays the sign language video sent back from the server to the user.

[1329] For illustrative purposes, consider the following prompt:

[1330] Customer Prompt: Hello! I have a question about this product.

[1331] "Store Staff Prompt": Welcome. Is there anything I can help you with?

[1332] By inputting or speaking these prompts, the text or audio is sent to the server, where it is analyzed and an appropriate sign language video is generated. Based on the emotions analyzed by the emotion engine, a natural and emotive sign language video is displayed on the device, enabling smooth communication between the user and the store clerk.

[1333] The hardware used includes the camera and microphone of a smartphone or tablet, a server (equipped with a high-performance GPU) for data processing and storage, and software used includes a generative AI model, a speech recognition model (e.g., Google Speech-to-Text API), and an emotion analysis engine (e.g., Microsoft Azure Emotion API).

[1334] This embodiment is expected to facilitate smooth communication between hearing-impaired and hearing-savvy people in physical stores, improving service quality and customer satisfaction.

[1335] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1336] Step 1:

[1337] The user uses the device's camera to record sign language video. This sign language video is a question the hearing-impaired person wants to ask the store clerk or information they want to convey in sign language. The input sign language video is saved on the device.

[1338] Input: Sign language video

[1339] Output: Sign language video data saved on the device

[1340] Step 2:

[1341] The device sends the sign language video data to a cloud server, where the data is uploaded via the internet. The server receives the data.

[1342] Input: Sign language video data stored on the device

[1343] Output: Sign language video data uploaded to the server

[1344] Step 3:

[1345] The server uses a generative AI model to analyze the received sign language video frame by frame, converting the sign actions into text in each language.

[1346] Input: Sign language video data uploaded to the server

[1347] Output: Frame-by-frame sign language motion analysis data and text data in various languages

[1348] Step 4:

[1349] The server uses an emotion engine to recognize the user's emotions from the analyzed sign language video, which then adjusts the tone and nuance of the generated text in each language.

[1350] Input: Frame-by-frame sign language motion analysis data and text data in various languages

[1351] Output: Emotion-adjusted text data in various languages

[1352] Step 5:

[1353] The server returns the adjusted text in each language to the terminal, which then displays the received text data on its screen.

[1354] Input: Emotion-adjusted text data in various languages

[1355] Output: Adjusted text data displayed on the terminal

[1356] Step 6:

[1357] The store clerk inputs text using the terminal's keyboard or voice input using the microphone, which generates text data or voice data in various languages.

[1358] Input: Text input or voice input

[1359] Output: Text data or audio data in various languages

[1360] Step 7:

[1361] The terminal transmits the generated text data or voice data to the server, which receives the data.

[1362] Input: Text data or audio data in various languages

[1363] Output: Text data or audio data in various languages ​​uploaded to the server

[1364] Step 8:

[1365] The server uses a speech recognition model to convert speech data into text (if speech is input), and a generative AI model to convert text data in each language into the appropriate sign language gestures.

[1366] Input: (if necessary) audio data, text data in various languages

[1367] Output: Text data in various languages, sign language movement data

[1368] Step 9:

[1369] The server again uses the emotion engine to recognize the emotion of the sign language actions based on the input text and audio, and adjusts the expression of the sign language video.

[1370] Input: Sign language movement data

[1371] Output: Emotion-adjusted sign language video data

[1372] Step 10:

[1373] The server then sends the adjusted sign language video data back to the terminal, which then displays the received sign language video on its screen so that the store clerk can check the content.

[1374] Input: Emotion-adjusted sign language video data

[1375] Output: Adjusted sign language video displayed on the device

[1376] 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.

[1377] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. 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 voice, text data indicating text, and image data indicating an image is also input. 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.

[1378] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1379] [Fourth embodiment]

[1380] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1381] 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.

[1382] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the 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).

[1383] 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.

[1384] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1385] 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 surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1386] 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.

[1387] The control object 443 includes a display device, LEDs in the eyes, and motors for driving 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.

[1388] 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.

[1389] 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 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.

[1390] 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.

[1391] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1392] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1393] This invention is a system that enables bidirectional translation between sign language and other languages. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[1394] When converting sign language videos into other languages

[1395] Program processing overview

[1396] 1. Input sign language video

[1397] The user uses the device to record a specific sign language video and upload it to the application.

[1398] 2. Send the sign language video to the server

[1399] The terminal transmits the sign language video data to the server.

[1400] 3. Analyzing sign language videos

[1401] The server receives the sign language video data and analyzes the video based on the generative AI model.

[1402] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[1403] 4. Generate the conversion results

[1404] The server returns to the terminal text in each language generated from the analyzed sign language movements.

[1405] 5. View the conversion results

[1406] The terminal displays the text data received from the server to the user.

[1407] Specific examples

[1408] A user can record a video of themselves speaking Japanese Sign Language on their smartphone and upload it through an application. This video is then sent over the internet to a server, which uses a generative AI model to analyze the sign language movements and convert them into Japanese text, such as "hello." This text is then sent back to the device and displayed to the user.

[1409] Converting text or dialogue from a foreign language into sign language video

[1410] Program processing overview

[1411] 1. Enter text or voice

[1412] The user can enter text using the device's keyboard or by speaking into the microphone.

[1413] 2. Send the input data to the server

[1414] The device sends the captured text or audio data to the server.

[1415] 3. Analyze the input data

[1416] The server receives the text or voice data and converts the voice to text as needed.

[1417] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[1418] 4. Generate sign language video

[1419] The server generates a corresponding sign language video based on the analyzed text data.

[1420] 5. Display sign language video

[1421] The terminal displays the sign language video data received from the server to the user.

[1422] Specific examples

[1423] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to a server, which uses a speech recognition model to convert the speech to text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video and sends it back to the device. Finally, the device displays the video of "Bonjour" in French Sign Language to the user.

[1424] As described above, the present invention realizes two-way translation between sign language and various languages ​​and functions as a system that supports communication between multiple languages.

[1425] The processing flow will be explained below.

[1426] When converting sign language videos into other languages

[1427] Program processing details

[1428] Step 1:

[1429] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[1430] Step 2:

[1431] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[1432] Step 3:

[1433] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[1434] Step 4:

[1435] The generative AI model analyzes the sign language movements for each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[1436] Step 5:

[1437] The server prepares the results of the analyzed sign language movements as text data and returns it to the terminal.

[1438] Step 6:

[1439] The terminal receives the text data in each language received from the server and displays it to the user. For example, the displayed text is "Hello."

[1440] Converting text or dialogue from a foreign language into sign language video

[1441] Program processing details

[1442] Step 1:

[1443] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[1444] Step 2:

[1445] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[1446] Step 3:

[1447] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[1448] Step 4:

[1449] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[1450] Step 5:

[1451] The server converts the generated sign language motion data into a sign language video, for example, generating a video of "Bonjour" in French sign language.

[1452] Step 6:

[1453] The server generates a response to send the generated sign language video data to the terminal, which receives the sign language video data via the Internet.

[1454] Step 7:

[1455] The device receives the sign language video data from the server and displays it to the user, who can then watch the video of "Bonjour" in French sign language on the device screen.

[1456] The above is a specific processing flow of the bidirectional translation system between sign language and other national languages ​​based on the present invention.

[1457] Example 1

[1458] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1459] Conventional translation systems between sign language and other languages ​​lack accuracy and real-time performance, making it difficult to facilitate smooth communication between multiple languages. Furthermore, while there is a need not only to convert sign language video into text in other languages, but also to convert text and audio in other languages ​​into sign language video, there are only a limited number of systems that can achieve this kind of two-way translation function. Therefore, there is a need for a system that can perform two-way translation with higher accuracy and in real time to support multilingual communication.

[1460] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1461] In this invention, the server includes means for a user to input sign language video, means for dividing the input sign language video and analyzing the hand position and shape of each frame, means for generating text in various languages ​​based on the analyzed sign language movements using a generative AI model, means for returning the converted text to the terminal, and means for displaying the converted text to the user, thereby enabling sign language video to be converted into text in various languages ​​with high accuracy and in real time.

[1462] "Sign language video" is video data that records communication using sign language.

[1463] "User" refers to a person who uses the system to input sign language video or input text or audio in various languages.

[1464] A "generative AI model" is a model that generates and converts text data and sign language motion data based on artificial intelligence. Examples include models that use natural language processing and computer vision techniques.

[1465] "Analysis" refers to the process of recognizing the hand position and shape for each frame of the input sign language video and converting it into text in each language based on that.

[1466] "Text" refers to character string data expressed in various languages, and refers to written language information.

[1467] A "terminal" is a device that allows a user to input sign language video, text, or audio, and receive responses from a server. Examples include smartphones and tablets.

[1468] A "server" is a computer system that analyzes input sign language video, text, and audio data, performs any necessary conversions, and then returns the results to the terminal.

[1469] The "Sign Language Animation Library" is a library for expressing sign language movements as animation using 3D motion capture data.

[1470] "Speech-to-text means" refers to any technology or function that analyzes and converts audio data into corresponding text data. An example is voice recognition software.

[1471] "Means for analyzing the position and shape of the hands in each frame" refers to technology or functions that divide sign language video into frames and recognize the position and shape of the hands in each frame.

[1472] "Means for generating sign language videos" refers to technologies and functions that generate new sign language videos based on analyzed sign language movements.

[1473] This invention is a system that realizes two-way translation between sign language and other languages, and is a system that mainly operates through the cooperation of three parties: a server, a terminal, and a user. This system provides functions to convert sign language videos into other languages, and to convert text and audio in other languages ​​into sign language videos.

[1474] A form of converting sign language videos into various languages

[1475] Users record sign language videos using devices such as smartphones. The recorded sign language videos are then sent to a server using a dedicated application on the device. The server then divides the received sign language videos and analyzes the hand position and shape for each frame using computer vision technology. Libraries such as TensorFlow and OpenCV are used for the analysis.

[1476] Based on the analyzed sign language movement data, a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) is used to generate text in the corresponding language. One example of a generative AI model that can be implemented is OpenAI's GPT-4.

[1477] The generated text in each language is sent back to the device in JSON format and displayed to the user by the device application. As a concrete example, a user records a video of themselves saying "hello" in Japanese sign language and uploads it to the application. This video is sent to a server, which analyzes the sign language movements and converts them into the Japanese text "hello". The converted text is then sent back to the device and displayed to the user.

[1478] A form of converting text or audio in various languages ​​into sign language video

[1479] Users can enter text in their native language using an application on their device, or use the microphone to input voice. The device then sends the entered text or voice data to the server. In the case of voice data, it is first converted into text using voice recognition software (e.g., Google's Speech-to-Text API).

[1480] The server uses a generative AI model to convert the input text into appropriate sign language actions, then uses a sign language animation library (e.g., a library using 3D motion capture data) to generate a sign language video, which is sent back from the server to the device and displayed to the user by the device application.

[1481] As a concrete example, if a user types or speaks "Bonjour" in French, the entered text or voice data is sent from the device to a server. The server uses speech recognition software to convert the speech into text and analyzes it with a generative AI model. Based on this data, the server generates a corresponding French Sign Language video and sends it back to the device. Finally, the sign language video is played in an application on the device.

[1482] Prompt Sentence Examples

[1483] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[1484] "Create a program that converts input text or audio into a sign language video and plays it back."

[1485] As described above, the present invention is a system that realizes bidirectional translation between sign language and various other languages ​​with high accuracy and in real time, thereby supporting communication between multiple languages.

[1486] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1487] When converting sign language videos into other languages

[1488] Processing steps and specific explanations

[1489] Step 1:

[1490] The user records a sign language video using the camera on their smartphone and presses the "upload" button in the dedicated application. The input is the recorded sign language video file, and the output is a video file sent from the device to the server.

[1491] Step 2:

[1492] The device sends the uploaded sign language video file to a server via the Internet. Specifically, the HTTPS protocol is used to encrypt data and ensure secure communication. The input is the sign language video file, and the output is the video data sent to the server.

[1493] Step 3:

[1494] The server temporarily stores the received sign language video and begins analyzing it using a generative AI model. The video is first divided into frames, and the hand position and shape of each frame are analyzed using computer vision technology (e.g., OpenCV, TensorFlow). The input is the sign language video stored on the server, and the output is analysis data including the hand position and shape.

[1495] Step 4:

[1496] The server uses a generative AI model (e.g., an AI model specialized in natural language processing or video analysis) to generate text in the corresponding language based on the analyzed sign language movement data. For generation, OpenAI's GPT-4 is used, for example. The input is sign language movement data, and the output is text data in the language.

[1497] Step 5:

[1498] The server converts the generated text in each language into JSON format and returns it to the terminal. The input is the generated text data, and the output is JSON format text data.

[1499] Step 6:

[1500] The terminal parses the JSON-formatted text data received from the server and displays it in the user interface. Specifically, it displays the generated text in each language in a text view within the application. The input is JSON-formatted text data, and the output is the text displayed in the user interface.

[1501] Converting text or audio in any language into sign language video

[1502] Processing steps and specific explanations

[1503] Step 1:

[1504] The user opens the application on their device and enters the text they want to translate into the text field, or presses the voice input button and speaks into the microphone. The input is text or voice data, and the output is data sent from the device to the server.

[1505] Step 2:

[1506] The device sends text data or recorded voice data to the server. In the case of voice data, it is first saved as a voice file on the device and then sent to the server. The input is text data or voice data, and the output is the data sent to the server.

[1507] Step 3:

[1508] The server directly analyzes the received text data, and converts audio data into text using Google's Speech-to-Text API. The input is text data or audio data, and the output is the converted data.

[1509] Step 4:

[1510] The server uses a generative AI model to convert input text in various languages ​​into data corresponding to sign language actions. The input is text data, and the output is sign language action data.

[1511] Step 5:

[1512] The server generates sign language animation using a sign language animation library (e.g., a 3D motion capture library) based on the sign language movement data. The input is the sign language movement data, and the output is the generated sign language video.

[1513] Step 6:

[1514] The terminal plays the sign language video data received from the server on the user interface. The user can view the translated sign language video through the application. The input is sign language video data, and the output is a sign language video that the user can view.

[1515] Prompt Sentence Examples

[1516] "Please create a program that converts Japanese Sign Language videos into various languages ​​and displays them."

[1517] "Create a program that converts input text or audio into a sign language video and plays it back."

[1518] As described above, by clearly showing the specific processing steps and operations, it has been shown how the system of the present invention provides bidirectional translation between sign language and various other languages ​​with high accuracy and in real time.

[1519] (Application example 1)

[1520] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1521] It aims to address the lack of communication tools that allow sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies.

[1522] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1523] In this invention, the server includes means for analyzing input sign language video and converting it into text in various languages, means for transmitting the input sign language video as an emergency call, and means for converting reply text from the server into sign language video, thereby enabling sign language users to make emergency calls via sign language video and respond quickly.

[1524] "Sign language video" is information including hand movements and facial expressions recorded in video format so that people who use sign language can understand it visually.

[1525] "Text in a national language" refers to character strings or sentences written in a specific language, and is text data in a format that can be read and understood by humans.

[1526] "User" refers to a person who uses this system to input, transmit, and display sign language video.

[1527] A "server" is a computer system that analyzes and converts sign language videos and text and manages communications.

[1528] An "emergency call" is a communication act to warn or request help in an emergency, and refers to the transmission of information that requires a rapid response.

[1529] "Reply text" is a written string or sentence sent from a server or the like as a response to an emergency call.

[1530] "Analysis" refers to the process of understanding the content of sign language video or text and converting it into another format.

[1531] "Conversion" refers to the process of converting data of one format into another format, and in this invention it primarily refers to conversion between sign language video and text in a foreign language.

[1532] This invention is a system that allows sign language users to efficiently and quickly report emergencies and security issues and receive corresponding sign language video replies. This system mainly operates in cooperation with three entities: a server, a terminal, and a user.

[1533] Sign language video input and analysis

[1534] A user first uses a device (e.g., a smartphone) to record a video of an emergency situation in sign language. This video is then sent to a server via an application. The device then uploads the video to the server, using a generative AI model to convert it into text in each language.

[1535] Analysis of sign language video and conversion to text

[1536] The server receives the sign language video data and analyzes the video based on the generative AI model. This analysis converts the sign language actions into text in the corresponding language. The converted text is then sent from the server to the relevant authorities (e.g., police or security center) as an emergency call.

[1537] Sending and replying to emergency calls

[1538] When a reply from a relevant organization is sent to the server in text format, the server analyzes the text data and generates a corresponding sign language video. This sign language video is then sent to the device and displayed to the user. This allows the user to quickly report the situation through sign language and receive necessary assistance, even in an emergency.

[1539] Example

[1540] For example, suppose a user discovers a suspicious person at home and uses their smartphone to record a sign language video saying "Help me" and report it. This video is sent to a server and converted into text for the Japanese word "Help me." This text is then sent to the police, who reply, "We will send a patrol immediately." This reply is analyzed by the server, converted into the corresponding sign language video, and displayed on the user's device.

[1541] Hardware and software used

[1542] Hardware: Your smartphone's camera, microphone, and display.

[1543] Software: Server-side generative AI models, speech recognition models, and text analysis tools.

[1544] Specific names: requests, cv2 (OpenCV), and speech_recognition libraries implemented in Python.

[1545] Prompt Sentence Examples

[1546] "Convert the French word 'help' into the corresponding sign language action."

[1547] By inputting this prompt into a generative AI model, the necessary sign language actions are generated. In this way, the present invention enables sign language users to safely make emergency calls and receive prompt responses.

[1548] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1549] Step 1:

[1550] A user uses a smartphone to record a sign language video about an emergency (e.g., discovering a suspicious person at home). A video captured with the smartphone camera is generated as input data for the sign language video. The recorded sign language video is then saved by the user via an application.

[1551] Step 2:

[1552] The device sends the recorded sign language video data to a server. Specifically, an application on the smartphone uses an Internet connection to upload the sign language video to a specified API endpoint on the server. The input is a sign language video file, and the output is the data sent to the server.

[1553] Step 3:

[1554] The server analyzes the received sign language video data using a generative AI model. It identifies each frame of the video and analyzes hand movements and facial expressions. As a result of this analysis, the input sign language video is converted into natural language text (text in each language). The input is sign language video data, and the output is text in the corresponding language.

[1555] Step 4:

[1556] The server sends the parsed text in each language as an emergency call to the appropriate security agency (e.g., police). The text data is communicated through the server's API. The input is the generated text in each language, and the output is the call data to the security agency.

[1557] Step 5:

[1558] A security agency (e.g., the police) receives the call and creates a reply in the local language. This reply text is sent back to the server. Here, as a police operation, the input is the text of the emergency call and the output is the local language text of the reply.

[1559] Step 6:

[1560] The server receives the reply text data from the security agency. This text in each language is analyzed again using the generative AI model and converted into the corresponding sign language video. The input is the reply text data, and the output is the sign language video.

[1561] Step 7:

[1562] The server sends the generated sign language video to the device. The device then displays the video to the user. The input is the sign language video data from the server, and the output is the sign language video displayed on the smartphone.

[1563] Step 8:

[1564] The user checks the sign language video on the device and understands the reply from the security agency in sign language. The input is the sign language video, and the output is what the user understands.

[1565] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1566] This invention is a system that enables two-way translation between sign language and other languages, and also has the ability to adjust the translation results by recognizing the user's emotions. This system mainly works in cooperation with three entities: the server, the terminal, and the user, and by combining an emotion engine, it provides more natural and appropriate translations.

[1567] When converting sign language videos into other languages

[1568] Program processing overview

[1569] 1. Input sign language video

[1570] A user uses a device such as a smartphone or PC to record a specific sign language video and upload it to the application.

[1571] 2. Send the sign language video to the server

[1572] The terminal transmits the sign language video data to the server.

[1573] 3. Analyzing sign language videos

[1574] The server receives the sign language video data and analyzes the video based on the generative AI model.

[1575] It analyzes sign language movements for each frame of video and generates corresponding text in each language.

[1576] 4. Emotional Engine Adjustment

[1577] The server uses an emotion engine to recognize the user's emotions.

[1578] Adjust the tone and nuance of the generated text based on the perceived sentiment.

[1579] 5. Generate the conversion results

[1580] The server returns the adjusted text in each language to the terminal.

[1581] 6. View the conversion results

[1582] The terminal displays the text data received from the server to the user.

[1583] Specific examples

[1584] Suppose a user films a video of themselves speaking Japanese Sign Language on their smartphone and uploads it through an application. The video is sent over the internet to a server, which uses a generative AI model to analyze the sign language and convert it into Japanese text, such as "hello." The emotion engine then recognizes the user's emotion (e.g., joy) and adjusts the nuance of the text to "a very happy 'hello.'" This text is then sent back to the device and displayed to the user.

[1585] Converting text or dialogue from a foreign language into sign language video

[1586] Program processing overview

[1587] 1. Enter text or voice

[1588] The user can input text in various languages ​​using the device's keyboard or by voice input using the microphone.

[1589] 2. Send the input data to the server

[1590] The device sends the captured text or audio data to the server.

[1591] 3. Analyze the input data

[1592] The server receives the text or voice data and converts the voice to text as needed.

[1593] The server converts text in each language into appropriate sign language actions based on a generative AI model.

[1594] 4. Emotional Engine Adjustment

[1595] The server uses an emotion engine to recognize the user's emotions.

[1596] The expression of sign language actions is adjusted based on the recognized emotion.

[1597] 5. Generate sign language video

[1598] The server generates a corresponding sign language video based on the analyzed text data and emotions.

[1599] 6. Display sign language video

[1600] The terminal displays the sign language video data received from the server to the user.

[1601] Specific examples

[1602] Suppose a user types or speaks "Bonjour" in French. The input data is sent from the device to the server, which uses a speech recognition model to convert the speech into text and analyzes it with a generative AI model. From this data, the server generates a corresponding sign language video, and an emotion engine recognizes the user's joy and reflects it in the sign language video. The sign language video sent back to the device visualizes "Bonjour" with a clear expression of joyful movements.

[1603] In this way, the combination of the emotion engine makes two-way translation between sign language and national languages ​​more natural and effective, further facilitating multilingual communication.

[1604] The processing flow will be explained below.

[1605] When converting sign language videos into other languages

[1606] Program processing details

[1607] Step 1:

[1608] The user records a specific sign language video using a device such as a smartphone or PC, and then uploads the recorded video to an automatic translation application.

[1609] Step 2:

[1610] The terminal generates a request to send the sign language video data uploaded by the user to the server, and then sends the sign language video data to the server via the Internet.

[1611] Step 3:

[1612] The server receives the sign language video data, inputs the video data into the generative AI model, and begins analyzing the sign language movements.

[1613] Step 4:

[1614] The generative AI model analyzes the sign language movements in each frame of the video and converts them into text in each language. For example, if the sign language movement indicates "hello," it will generate the Japanese text "hello."

[1615] Step 5:

[1616] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[1617] Step 6:

[1618] Based on the recognized emotion, the tone and nuance of the generated text in each language is adjusted. For example, if the user is happy, "Hello" is adjusted to "A very happy 'Hello'."

[1619] Step 7:

[1620] The server generates a request to return the adjusted text to the terminal and sends it over the Internet to the terminal.

[1621] Step 8:

[1622] The terminal receives the text data received from the server and displays it to the user. For example, the displayed text might be "A very happy 'hello'."

[1623] Converting text or dialogue from a foreign language into sign language video

[1624] Program processing details

[1625] Step 1:

[1626] The user can enter text in their local language using the device's keyboard or by speaking into the microphone, for example, by saying "Bonjour" in French.

[1627] Step 2:

[1628] The device generates a request to send the captured text or voice data to the server, and the device sends the input data to the server over the Internet.

[1629] Step 3:

[1630] The server receives text or voice data. In the case of voice data, the server uses a speech recognition model to convert the voice to text. For example, the voice "Bonjour" is converted to the French text "Bonjour."

[1631] Step 4:

[1632] The server inputs the French text "Bonjour" into the generative AI model, analyzes the corresponding sign language actions, and generates sign language action data corresponding to "Bonjour."

[1633] Step 5:

[1634] The server recognizes the user's emotions using an emotion engine, which analyzes the user's facial expressions, tone of voice, etc.

[1635] Step 6:

[1636] The expression of the sign language actions is adjusted based on the recognized emotion. For example, if the user is happy, the sign language actions are adjusted to be more cheerful and positive.

[1637] Step 7:

[1638] The server generates a sign language video based on the generated sign language movement data and the adjusted expression. For example, it creates a video that cheerfully expresses "Bonjour" in French sign language.

[1639] Step 8:

[1640] The server generates a request to transmit the generated sign language video data to the terminal, and transmits it to the terminal via the Internet.

[1641] Step 9:

[1642] The device receives the sign language video data from the server and displays it to the user. For example, a video of the French sign language "Bonjour" being displayed brightly on the device screen.

[1643] In this way, by combining the emotion engine, two-way translation between sign language and other languages ​​becomes more natural and emotional, providing users with a more effective means of communication.

[1644] Example 2

[1645] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1646] Conventional translation systems between sign language and other languages ​​often produce mechanical results without taking into account the user's emotions. As a result, the translation results cannot be said to support natural and appropriate communication, and there are problems with a lack of accuracy and expressiveness. This is particularly noticeable in communication where emotional nuances are important.

[1647] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1648] In this invention, the server includes a means for inputting a sign language video, a means for analyzing the input sign language video and converting it into text in a national language using a generative AI model, a means for recognizing emotions during the analysis and adjusting the tone and nuance of the generated text, and a means for displaying the adjusted text to the user. This enables natural and appropriate bidirectional translation between sign language and a national language that incorporates emotions.

[1649] "Sign language video" is video data that records the movements and gestures of a person using sign language.

[1650] A "generative AI model" is an artificial intelligence model that automatically generates text, images, etc. based on data.

[1651] A "server" is a computer system for storing, processing, and managing data.

[1652] A "terminal" is a device used by a user to input and display data, such as a smartphone or PC.

[1653] "User" refers to a person who uses this system to translate between sign language and other national languages.

[1654] "Analysis" is the process of breaking down and extracting data and finding meaning and patterns within it.

[1655] An "emotion engine" is a program that recognizes the user's emotions from text or video and adjusts the output results based on those emotions.

[1656] "Text" refers to data that consists of sentences or characters.

[1657] "Tone and nuance" refers to subtle changes and emotional resonance in words and expressions.

[1658] This invention provides a system that enables bidirectional translation between sign language and other languages. This system operates mainly through the cooperation of three entities: the server, the terminal, and the user, and provides more natural and appropriate translation by combining an emotion engine.

[1659] Hardware and Software Configuration

[1660] Device: The device on which users record and upload sign language videos. This can be a smartphone or PC.

[1661] Server: A computer system for data analysis, transformation, and emotion recognition.

[1662] Generative AI models: Use generative AI models (e.g., OpenAI's GPT-4) to convert sign language videos into text in various languages.

[1663] Emotion engine: An emotion engine (e.g., IBM Watson's Tone Analyzer) is used to recognize the user's emotions and adjust the tone and nuance of the translation results based on the results.

[1664] System Operation

[1665] When converting sign language videos into other languages

[1666] 1. Recording and uploading sign language videos

[1667] Users record specific sign language videos on their smartphones or PCs and upload them to the application, which then sends the video to the server.

[1668] 2. Analysis of sign language videos

[1669] The server analyzes the received sign language video, using a generative AI model to analyze the sign language movements for each frame of the video and generate corresponding text in each language.

[1670] 3. Recognizing emotions and adjusting to nuances

[1671] The server uses an emotion engine to recognize the user's emotions from the sign language video and adjust the tone and nuance of the generated text, making it more appropriate and natural-sounding.

[1672] 4. Returning and displaying text

[1673] The server returns the adjusted text data to the terminal, which displays it to the user.

[1674] Specific examples

[1675] A user records a video of themselves saying "hello" in Japanese Sign Language and uploads it through the application. The server analyzes the video and converts it into the Japanese text "hello." The emotion engine recognizes the user's emotion of joy and adjusts the nuance of the text to "a very happy 'hello.'" Finally, this text data is sent back to the user's device and displayed.

[1676] Example prompt sentence:

[1677] "A user has uploaded a video in Japanese Sign Language. Please translate the sign language in this video into Japanese and use the emotion engine to recognize that it contains the emotion 'joy' and adjust appropriately."

[1678] Converting text or dialogue from a foreign language into sign language video

[1679] 1. Text or voice input

[1680] The user inputs text using the device's keyboard or voice using the microphone, and this data is sent to the server.

[1681] 2. Text or speech analysis

[1682] The server analyzes the received text or audio data, converting audio to text as needed, and using generative AI models to convert international text into sign language actions.

[1683] 3. Emotion Recognition and Video Generation

[1684] The server uses an emotion engine to recognize the user's emotion from the input data and adjusts the expression of the sign language video, generating the sign language video based on this.

[1685] 4. Returning and displaying sign language videos

[1686] The server sends the generated sign language video back to the terminal, which displays it to the user.

[1687] Specific examples

[1688] The user types or speaks "Bonjour" in French. This data is sent to the server, which performs speech recognition and then analyzes it using a generative AI model. A sign language video is generated from the analyzed data, and an emotion engine recognizes the user's emotion of joy and reflects it in the video. This sign language video is then sent back to the user's device and displayed.

[1689] Example prompt sentence:

[1690] "A user types 'Bonjour' in French. Convert this text into a sign language video, and then use an emotion engine to recognize that it contains the emotion 'joy' and reflect that in the video."

[1691] In this way, the system enables natural bidirectional translation between sign language and other national languages.

[1692] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1693] When converting sign language videos into other languages

[1694] Step 1:

[1695] Users record specific sign language videos on their smartphones or PCs and upload them to the application.

[1696] Input: Sign language video file

[1697] Output: Video file saved on device

[1698] Specific operation: The user opens the application, presses the record button to record a sign language video, and when the recording is finished, the video file is saved on the device.

[1699] Step 2:

[1700] The terminal transmits the sign language video data to the server.

[1701] Input: Sign language video file saved on the device

[1702] Output: Sign language video file transferred to the server

[1703] Specific operation: When you press the "Send" button in the application, the video file is uploaded to the server via the Internet.

[1704] Step 3:

[1705] The server analyzes the sign language video it receives using a generative AI model.

[1706] Input: Sign language video file uploaded to the server

[1707] Output: Text data generated from sign language actions

[1708] Specific operation: Analyzes sign language video frame by frame on the server and runs a generative AI model that converts sign language movements into text data.

[1709] Step 4:

[1710] The server uses an emotion engine to recognize the user's emotions and adjust the tone and nuance of the generated text.

[1711] Input: Generated text data, facial expressions and movement data from the sign language video

[1712] Output: Text data adjusted based on user sentiment

[1713] What it does: It uses an emotion engine to analyze facial expressions and movements in sign language videos and adjust the nuance and expression of the text based on the recognized emotion (e.g., joy).

[1714] Step 5:

[1715] The server returns the adjusted text data to the terminal.

[1716] Input: Adjusted text data

[1717] Output: Adjusted text data sent to the terminal

[1718] Specific operation: Generates text data in JSON format and sends it to the device.

[1719] Step 6:

[1720] The terminal displays the text data received from the server to the user.

[1721] Input: Text data sent from the server

[1722] Output: Text data displayed on the device screen

[1723] What it does: Displays text data in a text display field within an application so that it can be read by the user.

[1724] Converting text or dialogue from a foreign language into sign language video

[1725] Step 1:

[1726] The user enters text using the device's keyboard or voice using the microphone.

[1727] Input: International text and audio data

[1728] Output: Input data saved on the device

[1729] Specific operation: When inputting text, the user types text on the keyboard, and when inputting voice, the user speaks into the microphone and voice data is recorded.

[1730] Step 2:

[1731] The device sends the captured text or audio data to the server.

[1732] Input: Text or voice data stored on your device

[1733] Output: Text or audio data transmitted to the server

[1734] Specific operation: Pressing the "Submit" button in the application uploads data to the server.

[1735] Step 3:

[1736] The server analyzes the text or audio data, converting the audio to text as needed, and using a generative AI model to convert the international text into sign language actions.

[1737] Input: Text and voice data sent to the server

[1738] Output: Sign language gesture data generated from text in various languages

[1739] Specific actions: Use a speech recognition model to convert speech data into text, then use a generative AI model to convert the text into sign language actions.

[1740] Step 4:

[1741] The server uses an emotion engine to recognize the user's emotions and adjusts the expression of the sign language video.

[1742] Input: Text and voice data, user emotion information

[1743] Output: Emotion-reflecting sign language movement data

[1744] Specific actions: The emotion engine analyzes emotions from text and voice and generates sign language actions appropriate to those emotions.

[1745] Step 5:

[1746] The server generates sign language video based on the analyzed text data and emotional information.

[1747] Input: Emotion-reflecting sign language movement data

[1748] Output: Generated sign language video

[1749] Specific actions: Use sign language animation generation software to create sign language videos based on text and emotional information.

[1750] Step 6:

[1751] The terminal displays the sign language video data received from the server to the user.

[1752] Input: Sign language video data transferred from the server

[1753] Output: Sign language video displayed on the device screen

[1754] What it does: Uses a video player to play a sign language video for the user to review.

[1755] (Application example 2)

[1756] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1757] In modern society, communication between hearing-impaired and hearing-speaking people is extremely important, but adequate means are lacking. In particular, it is difficult for hearing-impaired people to communicate smoothly with store clerks in brick-and-mortar stores, and lack of understanding causes inconvenience to both parties. Existing sign language translation systems do not provide natural-sounding translations that reflect emotions, so more effective communication is needed.

[1758] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for inputting sign language video, a means for analyzing the input sign language video and converting it into text in various languages, a means for recognizing the emotion of the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotion, and a means for displaying the converted text to the user. This enables hearing-impaired people to communicate naturally with store clerks using sign language in physical stores.

[1759] "Sign language video" is a recorded video of sign language used by the hearing impaired to communicate.

[1760] "National language text" is character data in a language used in a particular country or region.

[1761] "Analysis" is the process of breaking down input data and understanding its meaning and intent.

[1762] "Emotion recognition" means determining a person's feelings or emotional state from video and audio.

[1763] "Adjusting the tone" means modifying the way text or images are presented to suit the emotion or situation.

[1764] "Displaying to the user" means providing the translation results visually on the screen of a smartphone, tablet, etc.

[1765] A "server" is a computer system that processes and stores data.

[1766] The system that realizes this application example enables smooth communication between hearing-impaired people and store clerks, especially in brick-and-mortar stores. The system consists of a server, a terminal (a smartphone or tablet), and users (hearing-impaired people and store clerks).

[1767] The system includes the following means:

[1768] 1. How to input sign language video:

[1769] The user uses the device's camera to record a video of the sign language, which is then sent to the server for analysis.

[1770] 2. Means for analyzing input sign language video and converting it into text in various languages:

[1771] The server uses a generative AI model (e.g., OpenAI GPT) to analyze the sign language video frame by frame and generate corresponding text in each language.

[1772] 3. A means for recognizing emotions in the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotions:

[1773] The server uses an emotion engine (e.g., Microsoft Azure Emotion API) to recognize emotions in the sign language video and adjust the tone and nuance of the generated text based on those emotions.

[1774] 4. How to display the converted text to the user:

[1775] The terminal displays the text returned by the server to the user.

[1776] 5. Ways to enter text in different languages:

[1777] The store clerk uses the terminal's keyboard to enter text or the microphone to input voice data, which is then sent to the server.

[1778] 6. Means for analyzing text in various languages ​​and converting it into a corresponding sign language video, and means for recognizing emotions in the generated sign language video and adjusting the expression of the sign language video based on the recognized emotions:

[1779] The server analyzes the input text and audio data and generates sign language videos using a generative AI model. An emotion engine then recognizes the emotions in the sign language videos and adjusts the sign language expressions accordingly.

[1780] 7. How to display the converted sign language video to the user:

[1781] The terminal displays the sign language video sent back from the server to the user.

[1782] For illustrative purposes, consider the following prompt:

[1783] Customer Prompt: Hello! I have a question about this product.

[1784] "Store Staff Prompt": Welcome. Is there anything I can help you with?

[1785] By inputting or speaking these prompts, the text or audio is sent to the server, where it is analyzed and an appropriate sign language video is generated. Based on the emotions analyzed by the emotion engine, a natural and emotive sign language video is displayed on the device, enabling smooth communication between the user and the store clerk.

[1786] The hardware used includes the camera and microphone of a smartphone or tablet, a server (equipped with a high-performance GPU) for data processing and storage, and software used includes a generative AI model, a speech recognition model (e.g., Google Speech-to-Text API), and an emotion analysis engine (e.g., Microsoft Azure Emotion API).

[1787] This embodiment is expected to facilitate smooth communication between hearing-impaired and hearing-savvy people in physical stores, improving service quality and customer satisfaction.

[1788] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1789] Step 1:

[1790] The user uses the device's camera to record sign language video. This sign language video is a question the hearing-impaired person wants to ask the store clerk or information they want to convey in sign language. The input sign language video is saved on the device.

[1791] Input: Sign language video

[1792] Output: Sign language video data saved on the device

[1793] Step 2:

[1794] The device sends the sign language video data to a cloud server, where the data is uploaded via the internet. The server receives the data.

[1795] Input: Sign language video data stored on the device

[1796] Output: Sign language video data uploaded to the server

[1797] Step 3:

[1798] The server uses a generative AI model to analyze the received sign language video frame by frame, converting the sign actions into text in each language.

[1799] Input: Sign language video data uploaded to the server

[1800] Output: Frame-by-frame sign language motion analysis data and text data in various languages

[1801] Step 4:

[1802] The server uses an emotion engine to recognize the user's emotions from the analyzed sign language video, which then adjusts the tone and nuance of the generated text in each language.

[1803] Input: Frame-by-frame sign language motion analysis data and text data in various languages

[1804] Output: Emotion-adjusted text data in various languages

[1805] Step 5:

[1806] The server returns the adjusted text in each language to the terminal, which then displays the received text data on its screen.

[1807] Input: Emotion-adjusted text data in various languages

[1808] Output: Adjusted text data displayed on the terminal

[1809] Step 6:

[1810] The store clerk inputs text using the terminal's keyboard or voice input using the microphone, which generates text data or voice data in various languages.

[1811] Input: Text input or voice input

[1812] Output: Text data or audio data in various languages

[1813] Step 7:

[1814] The terminal transmits the generated text data or voice data to the server, which receives the data.

[1815] Input: Text data or audio data in various languages

[1816] Output: Text data or audio data in various languages ​​uploaded to the server

[1817] Step 8:

[1818] The server uses a speech recognition model to convert speech data into text (if speech is input), and a generative AI model to convert text data in each language into the appropriate sign language gestures.

[1819] Input: (if necessary) audio data, text data in various languages

[1820] Output: Text data in various languages, sign language movement data

[1821] Step 9:

[1822] The server again uses the emotion engine to recognize the emotion of the sign language actions based on the input text and audio, and adjusts the expression of the sign language video.

[1823] Input: Sign language movement data

[1824] Output: Emotion-adjusted sign language video data

[1825] Step 10:

[1826] The server then sends the adjusted sign language video data back to the terminal, which then displays the received sign language video on its screen so that the store clerk can check the content.

[1827] Input: Emotion-adjusted sign language video data

[1828] Output: Adjusted sign language video displayed on the device

[1829] 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.

[1830] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. 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 voice, text data indicating text, and image data indicating an image is also input. 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.

[1831] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1832] 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.

[1833] 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 includes both affect 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.

[1834] 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.

[1835] 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).

[1836] 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 indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, 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 indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, 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.

[1837] 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."

[1838] 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.

[1839] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1840] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1841] 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.

[1842] 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.

[1843] 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.

[1844] The hardware resource for executing a specific process can be any of the following processors: An example of a processor 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. Another example of a processor is 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.

[1845] The hardware resource that executes the specific processing 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 processing may be a single processor.

[1846] 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.

[1847] 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.

[1848] 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.

[1849] 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.

[1850] The following is further disclosed regarding the above embodiment.

[1851] (Claim 1)

[1852] A means for inputting sign language video;

[1853] A means to analyze input sign language video and convert it into text in various languages,

[1854] means for displaying the converted text to a user;

[1855] A system including:

[1856] (Claim 2)

[1857] a means for inputting text in various languages;

[1858] A means for analyzing input text and converting it into a corresponding sign language video;

[1859] means for displaying the converted sign language video to a user;

[1860] 10. The system of claim 1.

[1861] (Claim 3)

[1862] a means for converting speech to text;

[1863] means for analyzing and converting the converted text into a corresponding sign language video;

[1864] means for displaying the generated sign language video to a user;

[1865] 10. The system of claim 1.

[1866] "Example 1"

[1867] (Claim 1)

[1868] A means for a user to input sign language video;

[1869] A means for dividing the input sign language video and analyzing the position and shape of the hands in each frame;

[1870] A means for generating text in various languages ​​based on the analyzed sign language movements using an AI model;

[1871] means for transmitting the converted text back to the terminal;

[1872] means for displaying the converted text to a user;

[1873] A system including:

[1874] (Claim 2)

[1875] a means for inputting text or voice in each language;

[1876] means for transmitting input text or voice data to a server;

[1877] a means for converting the audio data into text;

[1878] A means for converting input text in various languages ​​into sign language actions using a generative AI model;

[1879] a means for generating a sign language video based on the converted sign language actions;

[1880] means for displaying the generated sign language video to a user;

[1881] 10. The system of claim 1.

[1882] (Claim 3)

[1883] a means for converting speech to text;

[1884] A means for converting the text converted by the generative AI model into sign language actions;

[1885] a means for generating sign language videos using a sign language animation library;

[1886] means for displaying the generated sign language video to a user;

[1887] 10. The system of claim 1.

[1888] "Application Example 1"

[1889] (Claim 1)

[1890] A means for inputting sign language video;

[1891] A means to analyze input sign language video and convert it into text in various languages,

[1892] means for displaying the converted text to a user;

[1893] means for transmitting the input sign language video as an emergency call to a specific server;

[1894] means for converting the reply text from the server into a corresponding sign language video and displaying it to the user;

[1895] A system including:

[1896] (Claim 2)

[1897] a means for inputting text in various languages;

[1898] A means for analyzing input text and converting it into a corresponding sign language video;

[1899] means for displaying the converted sign language video to a user;

[1900] a means for transmitting text in various languages ​​as an emergency call to a specific server;

[1901] means for converting the reply text from the server into a corresponding sign language video and displaying it to the user;

[1902] 10. The system of claim 1, comprising:

[1903] (Claim 3)

[1904] a means for converting speech to text;

[1905] means for analyzing and converting the converted text into a corresponding sign language video;

[1906] a means for displaying the generated sign language video to a user;

[1907] means for transmitting the voice as an emergency call to a specific server;

[1908] means for converting the reply text from the server into a corresponding sign language video and displaying it to the user;

[1909] 10. The system of claim 1, comprising:

[1910] "Example 2: Combining Emotion Engines"

[1911] (Claim 1)

[1912] A means for inputting sign language video;

[1913] A means to analyze input sign language video and convert it into text in various languages ​​using a generative AI model;

[1914] A means of recognizing emotions during analysis and adjusting the tone and nuance of the generated text;

[1915] means for displaying the adjusted text to a user;

[1916] A system including:

[1917] (Claim 2)

[1918] a means for inputting text in various languages;

[1919] A means for analyzing input text and converting it into a corresponding sign language video using a generative AI model; and

[1920] A means for recognizing emotions and adjusting the expressions of the sign language video during conversion;

[1921] means for displaying the adjusted sign language video to a user;

[1922] 10. The system of claim 1.

[1923] (Claim 3)

[1924] a means for converting speech to text;

[1925] means for analyzing the converted text and converting it into a corresponding sign language video using a generative AI model; and

[1926] A means for recognizing emotions and adjusting the expressions of the sign language video during conversion;

[1927] means for displaying the generated sign language video to a user;

[1928] 10. The system of claim 1.

[1929] "Application example 2 when combining emotion engines"

[1930] (Claim 1)

[1931] A means for inputting sign language video;

[1932] A means to analyze input sign language video and convert it into text in various languages,

[1933] means for recognizing emotions in the analyzed sign language video and adjusting the tone of the converted text based on the recognized emotions;

[1934] means for displaying the converted text to a user;

[1935] A system including:

[1936] (Claim 2)

[1937] a means for inputting text in various languages;

[1938] A means for analyzing input text and converting it into a corresponding sign language video;

[1939] means for recognizing emotions in the generated sign language video and adjusting the expression of the sign language video based on the recognized emotions;

[1940] means for displaying the converted sign language video to a user;

[1941] 10. The system of claim 1, comprising:

[1942] (Claim 3)

[1943] a means for converting speech to text;

[1944] means for analyzing and converting the converted text into a corresponding sign language video;

[1945] means for recognizing emotions in the generated sign language video and adjusting the expression of the sign language video based on the recognized emotions;

[1946] a means for displaying the generated sign language video to a user;

[1947] 10. The system of claim 1, comprising: [Explanation of symbols]

[1948] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for inputting sign language video; A means to analyze input sign language video and convert it into text in various languages, means for displaying the converted text to a user; A system including:

2. a means for inputting text in various languages; A means for analyzing input text and converting it into a corresponding sign language video; means for displaying the converted sign language video to a user; The system of claim 1 .

3. a means for converting speech to text; means for analyzing and converting the converted text into a corresponding sign language video; means for displaying the generated sign language video to a user; The system of claim 1 .

Images