System

Abstract

A system is provided.SOLUTION: A system comprising: means for acquiring image data for a hearing-impaired person to communicate using sign language; means for analyzing the acquired image data of the sign language and converting the image data into corresponding text data; means for translating the converted text data into sign language text of another country; and means for transmitting the translated sign language text.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] There are significant barriers to communication between hearing-impaired and non-hearing people, and between hearing-impaired people in different countries. Because hearing-impaired people cannot understand sign language, dialogue between the two is difficult, and differences in the sign languages ​​used in each country make international sign language communication difficult. The purpose of this invention is to solve these problems and realize a world in which more people can communicate smoothly. [Means for solving the problem]

[0005] The present invention is a system that includes a means for acquiring video data for hearing-impaired people to communicate using sign language, a means for analyzing the acquired sign language video data and converting it into corresponding text data, a means for translating the converted text data into sign language text for other countries, and a means for transmitting the translated sign language text. This enables smooth communication via sign language between hearing-impaired people and between hearing-impaired people and hearing-impaired people, and also enables international communication.

[0006] "Hearing impaired" refers to a person who has a hearing impairment and has difficulty understanding speech through hearing.

[0007] "Sign language" is a means of communication that expresses words and meanings using hand, finger, and body movements.

[0008] "Communication" is the act of people exchanging information and ideas.

[0009] A "normal person" is someone who does not have a specific disability.

[0010] "Video data" refers to image or video data captured using a camera or other imaging device.

[0011] "Means of acquisition" refers to methods or devices for collecting specified information using cameras, sensors, etc.

[0012] "Means for analysis" refers to the methods and devices used to analyze collected data and convert it into meaningful information.

[0013] "Text data" refers to data expressed as character information.

[0014] A "translation means" is a method or device for converting information expressed in one language into another language.

[0015] A "transmitting means" is a method or device for sending data to a specific destination.

[0016] A "system" is a configuration in which multiple devices and methods cooperate with each other to perform a specified function. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] The system of the present invention is a technology that uses terminals such as smartphones and PCs and a server. The specific usage and operation are as follows.

[0039] User operations

[0040] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, so the user can simply sign naturally without any special operations.

[0041] Device behavior

[0042] The device captures video of the user's sign language and sends the video to a server. It is important that the device application has a high-performance camera and network connection to maintain video quality. The video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2) to minimize latency.

[0043] Server Operation

[0044] The server analyzes the video data received from the device. This analysis utilizes a neural network model. The server recognizes the sign language movements and generates text data corresponding to those movements. The generated text data is then further refined through natural language processing algorithms and converted into meaningful text.

[0045] Next, the server uses its multilingual translation function to translate the generated text data into other language data. This translation uses natural language processing (NLP) technology, and sign-language-specific translation models are used to accommodate each country's sign language. Through this process, translated sign language text data is generated.

[0046] Sending and displaying translation data

[0047] The server then sends the translated sign language text back to the device. This transmission is also done in real time, and the device that receives it displays the data on its screen. By checking the displayed translated text, the user can communicate smoothly with sign language users in other countries.

[0048] Specific examples

[0049] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs, and the video is captured by the device's camera and sent to a server. The server analyzes the video data and converts it into English text data, which is then translated into Japanese text corresponding to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American Sign Language, enabling smooth communication.

[0050] The system of the present invention will enable smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing.

[0051] The processing flow will be explained below.

[0052] Step 1:

[0053] Device behavior

[0054] The user signs. The device activates the camera and captures the sign language video in real time. This video data is stored in temporary memory.

[0055] Step 2:

[0056] Device behavior

[0057] The acquired video data is sent to the server in real time, and WebSocket or HTTP / 2 is often used as the communication protocol.

[0058] Step 3:

[0059] Server Operation

[0060] The server receives the video data sent from the device and stores it in temporary storage. As soon as the data is received, it is immediately passed to the processing queue.

[0061] Step 4:

[0062] Server Operation

[0063] The received video data is input into an AI model to analyze the sign language movements, using a convolutional neural network (CNN) or a recurrent neural network (RNN).

[0064] Step 5:

[0065] Server Operation

[0066] The AI ​​model analyzes sign language movements and converts the corresponding sign language words or phrases into text data.

[0067] Step 6:

[0068] Server Operation

[0069] The generated text data is passed to a natural language processing (NLP) component, which composes it into grammatically correct sentences.

[0070] Step 7:

[0071] Server Operation

[0072] Text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[0073] Step 8:

[0074] Server Operation

[0075] The generated translated sign language text is sent to the terminal in real time, again using WebSocket or HTTP / 2 as the communication protocol.

[0076] Step 9:

[0077] Device behavior

[0078] The terminal receives the translated sign language text sent from the server, and displays the received data in an appropriate format so that the user can easily understand it.

[0079] Step 10:

[0080] User Actions

[0081] The translated sign language text data is checked. If necessary, a reply sign is made using the same system. The reply sign is also transmitted to the other party through the same steps.

[0082] Example 1

[0083] 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."

[0084] Conventional systems have had problems with delays and accuracy in sign language understanding and translation when hearing-impaired people communicate with sign language users from other countries who speak different languages. Furthermore, there was a lack of technology to efficiently analyze sign language video data, convert it into text, and translate it into multiple languages. This made it difficult to achieve smooth communication between different languages.

[0085] 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.

[0086] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data in real time and converting it into corresponding text data, means for translating the converted text data into sign language text of another country, and means for transmitting and displaying the translated sign language text, thereby enabling sign language users to communicate between different languages ​​in real time with high accuracy.

[0087] "Hearing impaired" refers to a person who has a reduced or absent ability to hear sound.

[0088] "Sign language" is a visual language that conveys meaning using movements of the hands, fingers, body and face.

[0089] "Video data" is digital data that includes visual information captured by a photographic device such as a camera.

[0090] "Real-time" refers to a state in which processing or reaction occurs immediately without delay.

[0091] "Analysis" is the process of breaking down data and understanding its structure and meaning.

[0092] "Text data" refers to digital data that contains text information.

[0093] "Translation" refers to the conversion of information written in one language into another language.

[0094] A "neural network" is an algorithm that mimics the neural circuits of the human brain, and is an artificial intelligence technology that is particularly used for pattern recognition and learning.

[0095] "Natural Language Processing (NLP)" is a technology that allows computers to understand, interpret, and generate human language.

[0096] A "server" refers to a computer system that provides services to other computers over a network.

[0097] "Terminal" refers to a device that is directly operated by a user and operates as part of a computer network.

[0098] "Transmission" refers to the act of moving data from one place to another.

[0099] "Display" refers to outputting data in a form that is visually recognizable to humans.

[0100] MODE FOR CARRYING OUT THE INVENTION

[0101] The system of the present invention is a technology that enables hearing-impaired people to communicate using sign language, and is a system that uses terminals such as smartphones and PCs and a server. Specific usage and operation are described below.

[0102] User operations

[0103] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, allowing the user to sign naturally without any special operations.

[0104] Device behavior

[0105] The device captures the user's sign language video using a high-performance camera and transmits the video data to a server in real time. Communication protocols such as WebSocket and HTTP / 2 are used to transmit the video data, minimizing latency. Specifically, the device utilizes the smartphone's built-in camera, an external HD camera, Wi-Fi, or 4G / 5G networks to ensure video quality and stable communication.

[0106] Server Operation

[0107] The server uses a deep learning-based neural network model to analyze the video data received from the device. Specifically, a model using TensorFlow and PyTorch asynchronously analyzes sign language movements frame by frame and extracts their features. Based on these features, the server recognizes the sign language movements and generates text data corresponding to those movements.

[0108] The generated text data is then translated into other languages ​​using natural language processing (NLP) technology. For example, the generated text is automatically translated using the Google Translate API or DeepL API. Furthermore, a translation model specialized for sign language is used to accurately convert the unique expressions and meanings of sign language into other languages.

[0109] Sending and displaying translation data

[0110] The server sends the generated translated sign language text to the device in real time. The device then displays the text data on the screen so that the user can confirm it. This display method can be an application UI or a pop-up message. The text displayed on the screen is displayed in a large font for easy viewing.

[0111] Specific examples

[0112] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user makes a sign to say "hello" in sign language, and the video of this is captured by the device's camera and sent to the server. The server analyzes the video data and converts it into text data for the English word "hello," which is then translated into the Japanese text "hello" that corresponds to the Japanese sign language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American sign language and achieve smooth communication.

[0113] Prompt Sentence Examples

[0114] "Please translate American Sign Language into Japanese text and display it."

[0115] This enables the system of the present invention to translate sign language in real time with high accuracy, supporting international communication.

[0116] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0117] Step 1:

[0118] The user begins signing.

[0119] Input: The user's sign language action.

[0120] Action: The user signs naturally towards the camera, specifically to express a greeting such as "hello."

[0121] Output: Video of a user signing.

[0122] Step 2:

[0123] The device captures the sign language video.

[0124] Input: Signing actions of the user seen on camera.

[0125] Actions: The device's camera records the user's sign language movements with high precision. The video is captured frame by frame and stored as digital video data.

[0126] Output: Captured video data of the user's sign language.

[0127] Step 3:

[0128] The terminal transmits the video data to the server.

[0129] Input: Captured video data of the user's sign language.

[0130] How it works: The device uses WebSocket or HTTP / 2 to send video data to the server with minimal latency, using Wi-Fi or 4G / 5G networks.

[0131] Output: Video data sent to the server.

[0132] Step 4:

[0133] The server analyzes the video data.

[0134] Input: Sign language video data sent from the device.

[0135] How it works: The server uses a deep learning-based neural network to analyze sign language movements frame by frame. Specifically, a model using TensorFlow and PyTorch extracts features.

[0136] Output: Analysis results of sign language movements (feature data).

[0137] Step 5:

[0138] The server converts the sign language into text.

[0139] Input: Analysis results of sign language movements (feature data).

[0140] Operation: The server recognizes sign language based on the feature data and generates the corresponding text. For example, the sign language for "hello" is converted into the text data "Hello."

[0141] Output: Generated English text data.

[0142] Step 6:

[0143] The server translates the text into other languages.

[0144] Input: Generated English text data.

[0145] How it works: The server uses natural language processing (NLP) techniques and sign language-specific translation models to translate text into other languages. Specifically, it uses the Google Translate API and DeepL API to convert "Hello" into "Hello."

[0146] Output: Translated Japanese text data.

[0147] Step 7:

[0148] The server sends the translated text to the device.

[0149] Input: Translated Japanese text data.

[0150] How it works: The server uses WebSocket or HTTP / 2 to send translation data to the device in real time.

[0151] Output: Japanese text data sent to the terminal.

[0152] Step 8:

[0153] The device displays the translated text.

[0154] Input: Japanese text data sent to the terminal.

[0155] How it works: Your device will display the translated text on-screen, either in the application UI or as a pop-up message, in a large, easy-to-read font.

[0156] Output: Japanese text "Hello" displayed on the screen.

[0157] Step 9:

[0158] The user checks the translated text.

[0159] Input: Japanese text "Hello" displayed on the screen.

[0160] How it works: The user reads the translated text displayed on the device and understands the sign language of the other country, enabling smooth communication.

[0161] Output: A user who understands the sign language content.

[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] The present invention aims to provide a means for hearing-impaired people to work efficiently in factories. With conventional systems, it is difficult for hearing-impaired people to use sign language to give instructions to work machines and robots, resulting in problems with smooth communication and work efficiency. By solving this problem, the present invention aims to achieve smooth communication between hearing-impaired people working in factories and work machines, thereby improving work efficiency.

[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 the acquired sign language video data and converting it into corresponding text data, means for converting the text data into sign language text in another language using a multilingual translation algorithm, and means for receiving the analyzed sign language data and operating factory work machines. This enables hearing-impaired people to issue instructions to work machines in real time using sign language, facilitating smooth communication within the factory and improving work efficiency.

[0167] "Video data" is digital information captured by a camera of hand movements and gestures used in sign language.

[0168] "Means" refers to a method or device for achieving a specific purpose.

[0169] "Means for acquiring sign language video data" refers to a function that uses cameras, sensors, etc. to record the movements of a person signing and saves them as digital video data.

[0170] The "means for analyzing sign language video data" refers to a technology that processes captured video data, recognizes specific sign language movements, and converts them into corresponding text data.

[0171] "Text data" is character information corresponding to sign language actions, and is data used to communicate intentions.

[0172] "Translation means" is a function for converting text data expressed in one language into text data in another language.

[0173] "Means for sending" refers to the function of sending converted or translated data to other devices or servers.

[0174] "Means of operating factory machinery" means the ability to receive instructions given in sign language to control machinery in a factory and perform specific tasks.

[0175] A "neural network" is a computational algorithm modeled on the neural circuits of the human brain, and is a technology that learns and recognizes specific patterns.

[0176] "Natural language processing algorithms" are technologies that allow computers to understand, generate, and translate human language.

[0177] A system based on an application example of the present invention enables hearing-impaired people to operate work machines in factories using sign language. This system involves a series of processes: acquiring video data of a person signing, analyzing it, converting it into text data necessary for machine control, and operating the work machine based on that text data.

[0178] System Components

[0179] Hardware

[0180] 1. Device camera: Using a device equipped with a high-performance camera, the actions of the person signing are captured in real time.

[0181] 2. Factory machinery: Factory machinery connected to the system, such as robot arms and transport devices.

[0182] software

[0183] 1. Video Analysis Module: A neural network is used to analyze sign language video data and convert it into corresponding text data. This model is built using libraries such as TensorFlow and Keras.

[0184] 2. Natural Language Processing (NLP) Module: This module uses natural language processing algorithms to refine the analyzed text data and convert it into a form suitable for operational instructions. This module includes a natural language processing library with multilingual translation capabilities.

[0185] 3. Robot control module: Includes an API that receives the translated operating instructions and controls the work machines in the factory.

[0186] Operational flow

[0187] User operations

[0188] When using sign language, users simply move their hands toward the device's camera, allowing them to sign naturally without any special operations.

[0189] Device behavior

[0190] The device captures the user's sign language movements in real time through a camera and transmits the video data to a server. This process uses communication protocols (e.g., WebSocket and HTTP / 2) to maintain video quality and minimize latency.

[0191] Server Operation

[0192] The server analyzes the video data received from the device using a neural network to recognize sign language movements. The recognized sign language movements are then converted into text data. This text data is further refined through a natural language processing module and converted into appropriate instructions for controlling factory machinery. These instructions are sent to the factory machinery in real time, and the machinery performs the operations according to the user's instructions.

[0193] Specific examples

[0194] For example, if a hearing-impaired person in a factory signs "go forward," this action is captured by the device's camera and sent as video data to a server. The server analyzes the video data, recognizes it as "go forward," and converts it into corresponding text data. This text data is then converted into an operational command for "go forward" and sent to a work machine (e.g., a transport robot). The transport robot receives this command and moves forward.

[0195] Example of input prompt for generative AI model

[0196] "Certain sign language actions are captured by the camera. Analyze them in real time and convert them into work instructions. For example, if the sign recognizes 'move forward,' give the robot the command to move forward."

[0197] This will enable smooth communication between hearing-impaired people and machines within the factory, improving work efficiency.

[0198] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0199] Step 1:

[0200] The user signs.

[0201] First, the user signs in front of the device's camera. At this time, the user does not need to perform any special operations; they simply sign naturally. The image data input is the user's hand movements.

[0202] Step 2:

[0203] The terminal acquires the video data.

[0204] The device uses a camera to capture the user's sign language movements. The camera acquires video data in real time and processes it to maintain high quality. The input on the device is the video data captured by the camera, and the output is the same video data.

[0205] Step 3:

[0206] The terminal transmits the video data to the server.

[0207] The captured video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2). The device manages the transmission process to minimize the delay of the video data. The input is the captured video data, and the output is the data sent to the server.

[0208] Step 4:

[0209] The server analyzes the video data.

[0210] The server analyzes the received video data using a neural network model to recognize sign language movements. Specifically, the video frames are input into the neural network model, and the sign language movements are converted into text data. The input is the video data sent from the device, and the output is the analyzed text data.

[0211] Step 5:

[0212] The server translates the text data.

[0213] The server uses natural language processing algorithms to translate the acquired text data into sign language text in other languages, and then converts it into appropriate operating instructions for the factory's work machines. The input is the analyzed text data, and the output is the translated operating instructions.

[0214] Step 6:

[0215] The server transmits the operation instruction data.

[0216] The server then transmits the translated operation instruction data to the factory machine using a communication protocol. This process also takes place in real time. The input is the translated operation instruction data, and the output is the data sent to the factory machine.

[0217] Step 7:

[0218] The factory's work machines carry out the operation instructions.

[0219] The factory's work machines receive the operation instruction data sent from the server and perform specific operations based on it. For example, if a transport robot receives the instruction to "move forward," it will move forward. The input is the operation instruction data sent from the server, and the output is the corresponding machine operation.

[0220] This enables users to control factory machinery in real time using sign language, enabling smooth communication between hearing-impaired people and factory machinery.

[0221] 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.

[0222] The system of the present invention is a technology that combines a terminal such as a smartphone or PC, a server, and an emotion engine that recognizes the user's emotions. The specific implementation method and operation flow are as follows.

[0223] User operations

[0224] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures this sign language movement. The device is equipped with an emotion engine that recognizes emotions from the user's facial expressions and movements. This video data is analyzed in real time and sent to a server.

[0225] Device behavior

[0226] The device captures video of the user's sign language and sends the video data to the server. It is important that the device application has a high-performance camera and network connection to maintain video quality. In addition, an emotion engine analyzes the user's facial expressions and movements to extract emotional data. This emotional data is also sent to the server along with the video data.

[0227] Server Operation

[0228] The server processes the video and emotion data received from the device. The video data is input into an AI model (e.g., a neural network) that analyzes the sign language movements and converts the corresponding sign language words or phrases into text data. Regarding the emotion data, the emotion engine reflects the analyzed information in the text data and generates text that expresses the emotion.

[0229] The generated text data is converted into text using a natural language processing (NLP) algorithm. The server then uses a multilingual translation function to translate the generated text data into other languages. This translation is performed using the Google Translate API or a proprietary translation AI model. The translated text data is then input back into the sign language generation model to generate sign language text corresponding to the sign language of the other country.

[0230] Sending and displaying translation data

[0231] The server sends the generated translated sign language text to the terminal. The terminal receives the translated sign language text from the server and displays it to the user. The displayed translated text is provided in an appropriate format that is easy for the user to understand.

[0232] Specific examples

[0233] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs with emotion, and the video and emotional data are captured by the device's camera and sent to the server. The server analyzes the video data and converts it into English text data, and the emotion engine reflects the analyzed emotional information in the text. The English text data is then translated into Japanese text that corresponds to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the content of the American sign language and its emotions, achieving smooth communication.

[0234] The system of the present invention enables smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing, while reflecting the user's emotions.

[0235] The processing flow will be explained below.

[0236] Step 1:

[0237] Device behavior

[0238] The user signs into the device's camera. The device's built-in camera captures the sign language video data, and the emotion engine simultaneously analyzes the user's facial expressions and movements to obtain emotional data. The video data and emotional data are temporarily stored in memory.

[0239] Step 2:

[0240] Device behavior

[0241] The device transmits the captured video data and emotion data to the server in real time using communication protocols such as WebSocket and HTTP / 2, and appropriate compression techniques are used to minimize delays during data transmission.

[0242] Step 3:

[0243] Server Operation

[0244] The server receives the video and emotion data sent from the device and stores it in temporary storage. The received data is immediately passed to the processing queue, where analysis begins.

[0245] Step 4:

[0246] Server Operation

[0247] The video data is fed into an AI model that analyzes the sign language movements. The AI ​​model used here is typically a neural network (e.g., CNN or RNN) that recognizes each sign movement and identifies the corresponding sign word or phrase.

[0248] Step 5:

[0249] Server Operation

[0250] After identifying the sign language actions, they are converted into corresponding text data, which is then fed into an NLP algorithm to be reconstructed into natural-sounding sentences.

[0251] Step 6:

[0252] Server Operation

[0253] Emotional data is analyzed to identify the user's emotions. NLP algorithms are used to reflect the emotional data in the text, so that the generated text contains the user's emotional information.

[0254] Step 7:

[0255] Server Operation

[0256] The text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[0257] Step 8:

[0258] Server Operation

[0259] The translated sign language text data is sent to the device, again using a communication protocol (WebSocket or HTTP / 2) to transfer the data in real time.

[0260] Step 9:

[0261] Device behavior

[0262] The terminal receives the translated sign language text data sent from the server and displays it to the user in an appropriate format, which also reflects the original user's emotions.

[0263] Step 10:

[0264] User Actions

[0265] The received translated sign language text data is checked, and a reply is made in sign language if necessary. The reply sign language is also transmitted to the other party through the same steps.

[0266] Specific examples

[0267] For example, consider the case where a hearing-impaired person in the United States (User A) communicates with a hearing-impaired person in Japan (User B). User A uses sign language, and the emotion engine analyzes his or her emotions. This video and emotion data is sent from the device to the server. The server analyzes the data, converts the sign language movements into English text data, and also reflects the emotion data. Next, the English text is translated into Japanese text and Japanese Sign Language. The translation result is sent to User B's device and displayed. This allows User B to understand the content of User A's sign language, including his or her emotions, achieving smooth communication.

[0268] This system enables communication through sign language that reflects the user's emotions, realizing smooth dialogue between hearing-impaired people in different countries and between hearing-impaired people and hearing-speaking people.

[0269] Example 2

[0270] 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."

[0271] There is a lack of efficient means to accurately convey sign language and emotions in communication between hearing-impaired and hearing-disabled people. Conventional sign language translation systems have difficulty conveying rich communication, including emotions, and are inadequate for translating sign language between multiple languages. This makes it difficult for hearing-impaired people to communicate smoothly with users of different cultures and languages.

[0272] 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.

[0273] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data and user emotion data and converting them into corresponding text data, means for analyzing the converted text data including the emotion data and translating it into sign language text for another country, and means for transmitting the translated sign language text. This enables multilingual translation of sign language including emotion, enabling smooth communication between hearing-impaired people and those without hearing.

[0274] "Hearing impaired" refers to a person who has difficulty communicating verbally and uses sign language or other visual methods to communicate.

[0275] "Sign language" is a method of expressing language using visual gestures, hand shapes, and facial expressions, and refers to a means of communication primarily used by people with hearing impairments.

[0276] "Video data" refers to digital data that includes visual information such as a user's sign language actions and facial expressions.

[0277] "Emotion data" refers to information about emotions analyzed from the user's facial expressions and movements.

[0278] "Analysis" refers to the process of analyzing collected data to extract useful information.

[0279] "Text data" refers to digital data that expresses linguistic information as characters.

[0280] "Translation" refers to the process of converting content expressed in one language into another language.

[0281] A "neural network" is an algorithm that mimics the neural circuits of living organisms, and refers to a machine learning model that is particularly used for recognizing images and sounds.

[0282] An "emotion engine" refers to a software or hardware system that automatically analyzes emotions from a user's facial expressions and voice.

[0283] "Natural language processing algorithms" refer to computational methods for analyzing and generating language data, and are particularly used for text translation and sentence generation.

[0284] A "sign language generation model" refers to a machine learning model or algorithm for converting text data into sign language.

[0285] A "server" refers to a computer system that provides services or functions to other computers over a network.

[0286] A "terminal" refers to a device that a user directly operates to input and display data, such as a smartphone or PC.

[0287] The system for implementing this invention operates in cooperation with users, terminals, and servers. Its purpose is to acquire video data for hearing-impaired people to communicate using sign language, analyze that data, convert it into corresponding text data, translate it into multiple languages, and then transmit the translated sign language text.

[0288] Hardware and software configuration

[0289] Device: A device that is directly operated by the user, such as a smartphone or PC. The device is equipped with a high-performance camera that captures the user's sign language movements and facial expressions.

[0290] Camera: A camera built into the device, designed to capture high-resolution, real-time video.

[0291] Emotion engine: Installed on the device, it uses technologies such as OpenCV and Emotion AI SDK to recognize emotions from the user's facial expressions and movements.

[0292] Server: A computer system that performs key processes such as data analysis, sign language to text conversion, multilingual translation, etc. The server analyzes the data using neural networks (e.g., models using TensorFlow or PyTorch) or natural language processing algorithms (e.g., spaCy or BERT models).

[0293] System Operation

[0294] This system operates in the following manner.

[0295] 1. User operations

[0296] The user expresses their feelings by signing, for example, "Hello, how are you?"

[0297] 2. Video and Emotion Data Capture

[0298] The device's camera captures the user's sign language and facial expressions, and the emotion engine analyzes the user's emotions from their facial expressions, generating video data of the sign language movements and emotional data.

[0299] 3. Sending data to the server

[0300] The generated video data and emotion data are transmitted to a server in real time using a secure communication protocol (e.g., HTTPS).

[0301] 4. Analysis of video data

[0302] The server inputs the received video data into a neural network, analyzes the sign language movements, and converts them into sign language words and phrases.

[0303] 5. Emotion Data Analysis

[0304] The server reflects emotional information in the text data generated from the sign language based on the emotional data from the emotion engine. For example, it adds the annotation "with a smile" to "Hello, how are you?"

[0305] 6. Multilingual Translation of Text Data

[0306] The server translates the generated text data using the Google Translate API or a proprietary translation AI model. For example, English text is translated into Japanese.

[0307] 7. Sign Language Text Generation

[0308] The translated text data is input into a sign language generation model to generate corresponding sign language text, for example, translated Japanese sign language text.

[0309] 8. Sending and displaying translation data to your device

[0310] The server sends the translated sign language text to the terminal. The terminal displays the received sign language text to the user in an easy-to-understand format. For example, the Japanese Sign Language text is displayed on the terminal of a Japanese user.

[0311] Specific examples

[0312] For example, if a deaf person in the United States wants to sign to a deaf person in Japan, "Hello, how are you?", they can use the following prompt:

[0313] Example prompt: "A deaf person in the United States signs 'Hello, how are you?' with emotion. Please translate this sign into Japanese sign language. Please also reflect the emotion."

[0314] As described above, this system enables multilingual translation of sign language, including emotional expressions, enabling smooth communication between hearing-impaired and hearing-disabled people.

[0315] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0316] Step 1:

[0317] The user signs.

[0318] As a specific action, the user expresses his / her feelings by using sign language. For example, the user uses sign language to say "Hello, how are you?"

[0319] Step 2:

[0320] The device captures video data and emotional data.

[0321] Input and output: The device's camera captures the user's sign language and facial expressions (input), and generates emotional data (output) through the video data and emotion engine.

[0322] Specifically, a high-resolution camera captures sign language movements and facial expressions, and an emotion engine (e.g., OpenCV or Emotion AI SDK) analyzes emotions from the user's facial expressions.

[0323] Step 3:

[0324] The terminal transmits the video data and emotion data to the server.

[0325] Input and Output: The captured video data and emotion data (input) are sent to the server (output) using a secure communication protocol.

[0326] Specifically, the terminal transmits data in real time using a secure communication protocol such as HTTPS.

[0327] Step 4:

[0328] The server analyzes the video data.

[0329] Input and Output: Received video data (input) is fed into the neural network model and converted into sign language words or phrases (output).

[0330] Specifically, the server uses an AI model using TensorFlow and PyTorch to analyze sign language movements and extract the corresponding sign language words and phrases.

[0331] Step 5:

[0332] The server analyzes the emotion data.

[0333] Input and output: Analyze the received emotional data (input) and reflect the emotional information (output) in the text generated from the sign language.

[0334] Specifically, the emotion engine adds the analyzed emotion information to the sign language text. For example, if the sign language is "Hello, how are you?", the emotion information "with a smile" is added.

[0335] Step 6:

[0336] The server translates the text data into multiple languages.

[0337] Input and output: The generated text data (input) is input into the Google Translate API or a proprietary translation AI model to obtain translated text data (output).

[0338] Specifically, the server uses natural language processing (NLP) technology to translate text, for example from English to Japanese.

[0339] Step 7:

[0340] The server inputs the translated text data into a sign language generation model.

[0341] Input and output: The translated text data (input) is input into the sign language generation model to generate sign language text in other countries (output).

[0342] Specifically, the translated text data is passed through a sign language generation model to generate Japanese Sign Language or other sign language text.

[0343] Step 8:

[0344] The server transmits the generated translated sign language text to the terminal.

[0345] Input and output: The generated translated sign language text (input) is sent to the terminal, which displays it to the user (output).

[0346] Specifically, the server sends the generated sign language text to the terminal, and the terminal displays it to the user in an appropriate format. For example, Japanese Sign Language text is displayed on the terminal of a Japanese user.

[0347] (Application example 2)

[0348] 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."

[0349] When hearing-impaired people shop or use services at brick-and-mortar stores, they face the challenge of having difficulty communicating smoothly using sign language. Furthermore, because store clerks cannot understand sign language, they are unable to provide sufficient product explanations or guidance, which often causes inconvenience to hearing-impaired people. Furthermore, they are required to be able to understand and respond to the emotional content of sign language.

[0350] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0351] In this invention, the server includes means for acquiring video data for the hearing impaired to communicate using sign language, means for analyzing the acquired sign language video data and converting it into corresponding text data, means for recognizing the user's emotions, generating emotion data, and reflecting it in text, means for translating the converted text data into sign language text of another country, means for transmitting the translated sign language text, and means for displaying the acquired text data and sign language text, thereby enabling sign language communication that reflects emotions between the hearing impaired and store clerks.

[0352] "Means for acquiring video data" refers to a mechanism for capturing video including sign language movements using devices such as cameras and sensors.

[0353] "Means for analyzing sign language video data" refers to algorithms or software that process the acquired sign language video, recognize sign language actions, and convert them into corresponding text data.

[0354] The "means for converting into corresponding text data" is a processing method for converting sign language expressions into natural language text based on the analyzed sign language video data.

[0355] "Means for translating into sign language text of other countries" refers to a translation algorithm or API for converting the converted text data into sign language text in multiple languages.

[0356] "Means for transmitting sign language text" refers to a communication means for transmitting the translated sign language text to another terminal or server via a network.

[0357] The "means for recognizing user emotions" refers to algorithms or software for identifying a user's emotions and generating emotion data through facial expression recognition and motion analysis.

[0358] The "means for generating emotion data" is a process for generating data for expressing emotions based on the recognized emotion information.

[0359] The "means for reflecting in text" is a processing method for integrating the generated emotion data into text data and generating a text expression that includes emotion.

[0360] The "means for displaying text data and sign language text" is a mechanism for displaying the acquired text data and sign language text on the screen or display of the terminal.

[0361] To implement this invention, the following hardware and software environment is used. The main hardware used is a smartphone, server, camera, and display. The software used is a neural network framework (TensorFlow or PyTorch), a natural language processing (NLP) engine (spaCy, Hugging Face Transformers), an emotion recognition engine (EmotionRecognitionAPI), and a translation API (Google Translate API).

[0362] First, the user signs using the smartphone's camera. The camera captures the sign language movements, and the device's emotion engine analyzes the video data in real time. This analysis recognizes emotions from the user's facial expressions and movements, and generates emotion data. The video data and emotion data are then sent to the server.

[0363] The server analyzes the received video data using a neural network and converts the sign language movements into text data. It also reflects the corresponding emotions based on the emotional data. The generated text data is then compiled into sentences using a natural language processing (NLP) algorithm.

[0364] Next, the server translates the generated text data into other languages ​​using a multilingual translation API. This translated text data is input into a sign language generation model to generate sign language text for other languages. The translated sign language text is then sent from the server to the device.

[0365] When the terminal receives the translated sign language text from the server, it displays it to the user. The displayed text is presented in an appropriate emotional format that is easy for the user to understand. In addition, it plays animations of the sign language, allowing the user to visually understand the sign language.

[0366] Examples of specific prompts include:

[0367] "Analyze the sign language video captured by this camera and generate corresponding text and emotional information. Translate the generated text data into Japanese and convert it into Japanese Sign Language."

[0368] This invention allows hearing-impaired people to easily communicate with store staff when using services in physical stores. Because it does not use voice, it can be used effectively even in quiet environments. Furthermore, by displaying sign language that reflects emotions, it is possible to accurately convey the user's intentions and emotions.

[0369] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0370] Step 1:

[0371] The user signs into the smartphone camera.

[0372] Input: Sign language action

[0373] Output: Sign language video data

[0374] The device's camera captures the user's sign language movements and obtains high-quality video data, which is important for subsequent analysis.

[0375] Step 2:

[0376] The sign language video data acquired by the device is analyzed using an emotion engine to generate emotion data.

[0377] Input: Sign language video data

[0378] Output: Emotion data

[0379] The emotion engine built into the device processes the video data, recognizes emotions from the user's facial expressions and movements, and generates emotion data, which is then used in the translation process.

[0380] Step 3:

[0381] The device transmits sign language video data and emotion data to the server.

[0382] Input: Sign language video data, emotion data

[0383] Output: Data packet to be sent

[0384] Using a high-performance network connection, the device transmits sign language video data and emotion data in real time to a server, where the data packets are analyzed.

[0385] Step 4:

[0386] The server analyzes the sign language video data using a neural network and converts it into corresponding text data.

[0387] Input: Sign language video data

[0388] Output: Corresponding text data

[0389] The server's neural network analyzes sign language movements, identifies corresponding words and phrases, and converts them into text data, using TensorFlow and PyTorch for the process.

[0390] Step 5:

[0391] The server reflects the emotion data in the generated text data.

[0392] Input: Corresponding text data, emotion data

[0393] Output: Text data containing emotions

[0394] The server can incorporate emotional information into text data based on emotion recognition data, thereby conveying the user's intentions more accurately.

[0395] Step 6:

[0396] The server translates the generated text data into other languages ​​using natural language processing algorithms.

[0397] Input: Text data containing emotions

[0398] Output: Translated text data

[0399] The server uses a natural language processing engine (e.g., spaCy or Hugging Face Transformers) to translate the text data into other languages.

[0400] Step 7:

[0401] The server inputs the translated text data into a sign language generation model to generate sign language text for other countries.

[0402] Input: Translated text data

[0403] Output: Foreign sign language text

[0404] The server uses an AI model to generate sign language text corresponding to the sign language of other countries based on the translation data.

[0405] Step 8:

[0406] The server transmits the generated sign language text of the other country to the terminal.

[0407] Input: Foreign sign language text

[0408] Output: Data packet to be sent

[0409] The server transmits the generated sign language text data of the other country to the terminal via the network.

[0410] Step 9:

[0411] The terminal displays the sign language text received from the server to the user.

[0412] Input: Foreign sign language text

[0413] Output: Data for display

[0414] The terminal displays the received data on a display and plays visual sign language animations to help the user understand the content.

[0415] 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.

[0416] 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.

[0417] 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.

[0418] [Second embodiment]

[0419] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0420] 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.

[0421] 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).

[0422] 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.

[0423] 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.

[0424] 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).

[0425] 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.

[0426] 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.

[0427] 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.

[0428] 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.

[0429] 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.

[0430] 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."

[0431] The system of the present invention is a technology that uses terminals such as smartphones and PCs and a server. The specific usage and operation are as follows.

[0432] User operations

[0433] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, so the user can simply sign naturally without any special operations.

[0434] Device behavior

[0435] The device captures video of the user's sign language and sends the video to a server. It is important that the device application has a high-performance camera and network connection to maintain video quality. The video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2) to minimize latency.

[0436] Server Operation

[0437] The server analyzes the video data received from the device. This analysis utilizes a neural network model. The server recognizes the sign language movements and generates text data corresponding to those movements. The generated text data is then further refined through natural language processing algorithms and converted into meaningful text.

[0438] Next, the server uses its multilingual translation function to translate the generated text data into other language data. This translation uses natural language processing (NLP) technology, and sign-language-specific translation models are used to accommodate each country's sign language. Through this process, translated sign language text data is generated.

[0439] Sending and displaying translation data

[0440] The server then sends the translated sign language text back to the device. This transmission is also done in real time, and the device that receives it displays the data on its screen. By checking the displayed translated text, the user can communicate smoothly with sign language users in other countries.

[0441] Specific examples

[0442] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs, and the video is captured by the device's camera and sent to a server. The server analyzes the video data and converts it into English text data, which is then translated into Japanese text corresponding to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American Sign Language, enabling smooth communication.

[0443] The system of the present invention will enable smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing.

[0444] The processing flow will be explained below.

[0445] Step 1:

[0446] Device behavior

[0447] The user signs. The device activates the camera and captures the sign language video in real time. This video data is stored in temporary memory.

[0448] Step 2:

[0449] Device behavior

[0450] The acquired video data is sent to the server in real time, and WebSocket or HTTP / 2 is often used as the communication protocol.

[0451] Step 3:

[0452] Server Operation

[0453] The server receives the video data sent from the device and stores it in temporary storage. As soon as the data is received, it is immediately passed to the processing queue.

[0454] Step 4:

[0455] Server Operation

[0456] The received video data is input into an AI model to analyze the sign language movements, using a convolutional neural network (CNN) or a recurrent neural network (RNN).

[0457] Step 5:

[0458] Server Operation

[0459] The AI ​​model analyzes sign language movements and converts the corresponding sign language words or phrases into text data.

[0460] Step 6:

[0461] Server Operation

[0462] The generated text data is passed to a natural language processing (NLP) component, which composes it into grammatically correct sentences.

[0463] Step 7:

[0464] Server Operation

[0465] Text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[0466] Step 8:

[0467] Server Operation

[0468] The generated translated sign language text is sent to the terminal in real time, again using WebSocket or HTTP / 2 as the communication protocol.

[0469] Step 9:

[0470] Device behavior

[0471] The terminal receives the translated sign language text sent from the server, and displays the received data in an appropriate format so that the user can easily understand it.

[0472] Step 10:

[0473] User Actions

[0474] The translated sign language text data is checked. If necessary, a reply sign is made using the same system. The reply sign is also transmitted to the other party through the same steps.

[0475] Example 1

[0476] 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."

[0477] Conventional systems have had problems with delays and accuracy in sign language understanding and translation when hearing-impaired people communicate with sign language users from other countries who speak different languages. Furthermore, there was a lack of technology to efficiently analyze sign language video data, convert it into text, and translate it into multiple languages. This made it difficult to achieve smooth communication between different languages.

[0478] 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.

[0479] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data in real time and converting it into corresponding text data, means for translating the converted text data into sign language text of another country, and means for transmitting and displaying the translated sign language text, thereby enabling sign language users to communicate between different languages ​​in real time with high accuracy.

[0480] "Hearing impaired" refers to a person who has a reduced or absent ability to hear sound.

[0481] "Sign language" is a visual language that conveys meaning using movements of the hands, fingers, body and face.

[0482] "Video data" is digital data that includes visual information captured by a photographic device such as a camera.

[0483] "Real-time" refers to a state in which processing or reaction occurs immediately without delay.

[0484] "Analysis" is the process of breaking down data and understanding its structure and meaning.

[0485] "Text data" refers to digital data that contains text information.

[0486] "Translation" refers to the conversion of information written in one language into another language.

[0487] A "neural network" is an algorithm that mimics the neural circuits of the human brain, and is an artificial intelligence technology that is particularly used for pattern recognition and learning.

[0488] "Natural Language Processing (NLP)" is a technology that allows computers to understand, interpret, and generate human language.

[0489] A "server" refers to a computer system that provides services to other computers over a network.

[0490] "Terminal" refers to a device that is directly operated by a user and operates as part of a computer network.

[0491] "Transmission" refers to the act of moving data from one place to another.

[0492] "Display" refers to outputting data in a form that is visually recognizable to humans.

[0493] MODE FOR CARRYING OUT THE INVENTION

[0494] The system of the present invention is a technology that enables hearing-impaired people to communicate using sign language, and is a system that uses terminals such as smartphones and PCs and a server. Specific usage and operation are described below.

[0495] User operations

[0496] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, allowing the user to sign naturally without any special operations.

[0497] Device behavior

[0498] The device captures the user's sign language video using a high-performance camera and transmits the video data to a server in real time. Communication protocols such as WebSocket and HTTP / 2 are used to transmit the video data, minimizing latency. Specifically, the device utilizes the smartphone's built-in camera, an external HD camera, Wi-Fi, or 4G / 5G networks to ensure video quality and stable communication.

[0499] Server Operation

[0500] The server uses a deep learning-based neural network model to analyze the video data received from the device. Specifically, a model using TensorFlow and PyTorch asynchronously analyzes sign language movements frame by frame and extracts their features. Based on these features, the server recognizes the sign language movements and generates text data corresponding to those movements.

[0501] The generated text data is then translated into other languages ​​using natural language processing (NLP) technology. For example, the generated text is automatically translated using the Google Translate API or DeepL API. Furthermore, a translation model specialized for sign language is used to accurately convert the unique expressions and meanings of sign language into other languages.

[0502] Sending and displaying translation data

[0503] The server sends the generated translated sign language text to the device in real time. The device then displays the text data on the screen so that the user can confirm it. This display method can be an application UI or a pop-up message. The text displayed on the screen is displayed in a large font for easy viewing.

[0504] Specific examples

[0505] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user makes a sign to say "hello" in sign language, and the video of this is captured by the device's camera and sent to the server. The server analyzes the video data and converts it into text data for the English word "hello," which is then translated into the Japanese text "hello" that corresponds to the Japanese sign language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American sign language and achieve smooth communication.

[0506] Prompt Sentence Examples

[0507] "Please translate American Sign Language into Japanese text and display it."

[0508] This enables the system of the present invention to translate sign language in real time with high accuracy, supporting international communication.

[0509] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0510] Step 1:

[0511] The user begins signing.

[0512] Input: The user's sign language action.

[0513] Action: The user signs naturally towards the camera, specifically to express a greeting such as "hello."

[0514] Output: Video of a user signing.

[0515] Step 2:

[0516] The device captures the sign language video.

[0517] Input: Signing actions of the user seen on camera.

[0518] Actions: The device's camera records the user's sign language movements with high precision. The video is captured frame by frame and stored as digital video data.

[0519] Output: Captured video data of the user's sign language.

[0520] Step 3:

[0521] The terminal transmits the video data to the server.

[0522] Input: Captured video data of the user's sign language.

[0523] How it works: The device uses WebSocket or HTTP / 2 to send video data to the server with minimal latency, using Wi-Fi or 4G / 5G networks.

[0524] Output: Video data sent to the server.

[0525] Step 4:

[0526] The server analyzes the video data.

[0527] Input: Sign language video data sent from the device.

[0528] How it works: The server uses a deep learning-based neural network to analyze sign language movements frame by frame. Specifically, a model using TensorFlow and PyTorch extracts features.

[0529] Output: Analysis results of sign language movements (feature data).

[0530] Step 5:

[0531] The server converts the sign language into text.

[0532] Input: Analysis results of sign language movements (feature data).

[0533] Operation: The server recognizes sign language based on the feature data and generates the corresponding text. For example, the sign language for "hello" is converted into the text data "Hello."

[0534] Output: Generated English text data.

[0535] Step 6:

[0536] The server translates the text into other languages.

[0537] Input: Generated English text data.

[0538] How it works: The server uses natural language processing (NLP) techniques and sign language-specific translation models to translate text into other languages. Specifically, it uses the Google Translate API and DeepL API to convert "Hello" into "Hello."

[0539] Output: Translated Japanese text data.

[0540] Step 7:

[0541] The server sends the translated text to the device.

[0542] Input: Translated Japanese text data.

[0543] How it works: The server uses WebSocket or HTTP / 2 to send translation data to the device in real time.

[0544] Output: Japanese text data sent to the terminal.

[0545] Step 8:

[0546] The device displays the translated text.

[0547] Input: Japanese text data sent to the terminal.

[0548] How it works: Your device will display the translated text on-screen, either in the application UI or as a pop-up message, in a large, easy-to-read font.

[0549] Output: Japanese text "Hello" displayed on the screen.

[0550] Step 9:

[0551] The user checks the translated text.

[0552] Input: Japanese text "Hello" displayed on the screen.

[0553] How it works: The user reads the translated text displayed on the device and understands the sign language of the other country, enabling smooth communication.

[0554] Output: A user who understands the sign language content.

[0555] (Application example 1)

[0556] 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."

[0557] The present invention aims to provide a means for hearing-impaired people to work efficiently in factories. With conventional systems, it is difficult for hearing-impaired people to use sign language to give instructions to work machines and robots, resulting in problems with smooth communication and work efficiency. By solving this problem, the present invention aims to achieve smooth communication between hearing-impaired people working in factories and work machines, thereby improving work efficiency.

[0558] 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.

[0559] In this invention, the server includes means for analyzing the acquired sign language video data and converting it into corresponding text data, means for converting the text data into sign language text in another language using a multilingual translation algorithm, and means for receiving the analyzed sign language data and operating factory work machines. This enables hearing-impaired people to issue instructions to work machines in real time using sign language, facilitating smooth communication within the factory and improving work efficiency.

[0560] "Video data" is digital information captured by a camera of hand movements and gestures used in sign language.

[0561] "Means" refers to a method or device for achieving a specific purpose.

[0562] "Means for acquiring sign language video data" refers to a function that uses cameras, sensors, etc. to record the movements of a person signing and saves them as digital video data.

[0563] The "means for analyzing sign language video data" refers to a technology that processes captured video data, recognizes specific sign language movements, and converts them into corresponding text data.

[0564] "Text data" is character information corresponding to sign language actions, and is data used to communicate intentions.

[0565] "Translation means" is a function for converting text data expressed in one language into text data in another language.

[0566] "Means for sending" refers to the function of sending converted or translated data to other devices or servers.

[0567] "Means of operating factory machinery" means the ability to receive instructions given in sign language to control machinery in a factory and perform specific tasks.

[0568] A "neural network" is a computational algorithm modeled on the neural circuits of the human brain, and is a technology that learns and recognizes specific patterns.

[0569] "Natural language processing algorithms" are technologies that allow computers to understand, generate, and translate human language.

[0570] A system based on an application example of the present invention enables hearing-impaired people to operate work machines in factories using sign language. This system involves a series of processes: acquiring video data of a person signing, analyzing it, converting it into text data necessary for machine control, and operating the work machine based on that text data.

[0571] System Components

[0572] Hardware

[0573] 1. Device camera: Using a device equipped with a high-performance camera, the actions of the person signing are captured in real time.

[0574] 2. Factory machinery: Factory machinery connected to the system, such as robot arms and transport devices.

[0575] software

[0576] 1. Video Analysis Module: A neural network is used to analyze sign language video data and convert it into corresponding text data. This model is built using libraries such as TensorFlow and Keras.

[0577] 2. Natural Language Processing (NLP) Module: This module uses natural language processing algorithms to refine the analyzed text data and convert it into a form suitable for operational instructions. This module includes a natural language processing library with multilingual translation capabilities.

[0578] 3. Robot control module: Includes an API that receives the translated operating instructions and controls the work machines in the factory.

[0579] Operational flow

[0580] User operations

[0581] When using sign language, users simply move their hands toward the device's camera, allowing them to sign naturally without any special operations.

[0582] Device behavior

[0583] The device captures the user's sign language movements in real time through a camera and transmits the video data to a server. This process uses communication protocols (e.g., WebSocket and HTTP / 2) to maintain video quality and minimize latency.

[0584] Server Operation

[0585] The server analyzes the video data received from the device using a neural network to recognize sign language movements. The recognized sign language movements are then converted into text data. This text data is further refined through a natural language processing module and converted into appropriate instructions for controlling factory machinery. These instructions are sent to the factory machinery in real time, and the machinery performs the operations according to the user's instructions.

[0586] Specific examples

[0587] For example, if a hearing-impaired person in a factory signs "go forward," this action is captured by the device's camera and sent as video data to a server. The server analyzes the video data, recognizes it as "go forward," and converts it into corresponding text data. This text data is then converted into an operational command for "go forward" and sent to a work machine (e.g., a transport robot). The transport robot receives this command and moves forward.

[0588] Example of input prompt for generative AI model

[0589] "Specific sign language actions are captured by the camera. Analyze them in real time and translate them into work instructions. For example, if the sign recognizes 'move forward,' give the robot the command to move forward."

[0590] This will enable smooth communication between hearing-impaired people and machines within the factory, improving work efficiency.

[0591] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0592] Step 1:

[0593] The user signs.

[0594] First, the user signs in front of the device's camera. At this time, the user does not need to perform any special operations; they simply sign naturally. The image data input is the user's hand movements.

[0595] Step 2:

[0596] The terminal acquires the video data.

[0597] The device uses a camera to capture the user's sign language movements. The camera acquires video data in real time and processes it to maintain high quality. The input on the device is the video data captured by the camera, and the output is the same video data.

[0598] Step 3:

[0599] The terminal transmits the video data to the server.

[0600] The captured video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2). The device manages the transmission process to minimize the delay of the video data. The input is the captured video data, and the output is the data sent to the server.

[0601] Step 4:

[0602] The server analyzes the video data.

[0603] The server analyzes the received video data using a neural network model to recognize sign language movements. Specifically, the video frames are input into the neural network model, and the sign language movements are converted into text data. The input is the video data sent from the device, and the output is the analyzed text data.

[0604] Step 5:

[0605] The server translates the text data.

[0606] The server uses natural language processing algorithms to translate the acquired text data into sign language text in other languages, and then converts it into appropriate operating instructions for the factory's work machines. The input is the analyzed text data, and the output is the translated operating instructions.

[0607] Step 6:

[0608] The server transmits the operation instruction data.

[0609] The server then transmits the translated operation instruction data to the factory machine using a communication protocol. This process also takes place in real time. The input is the translated operation instruction data, and the output is the data sent to the factory machine.

[0610] Step 7:

[0611] The factory's work machines carry out the operation instructions.

[0612] The factory's work machines receive the operation instruction data sent from the server and perform specific operations based on it. For example, if a transport robot receives the instruction to "move forward," it will move forward. The input is the operation instruction data sent from the server, and the output is the corresponding machine operation.

[0613] This enables users to control factory machinery in real time using sign language, enabling smooth communication between hearing-impaired people and factory machinery.

[0614] 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.

[0615] The system of the present invention is a technology that combines a terminal such as a smartphone or PC, a server, and an emotion engine that recognizes the user's emotions. The specific implementation method and operation flow are as follows.

[0616] User operations

[0617] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures this sign language movement. The device is equipped with an emotion engine that recognizes emotions from the user's facial expressions and movements. This video data is analyzed in real time and sent to a server.

[0618] Device behavior

[0619] The device captures video of the user's sign language and sends the video data to the server. It is important that the device application has a high-performance camera and network connection to maintain video quality. In addition, an emotion engine analyzes the user's facial expressions and movements to extract emotional data. This emotional data is also sent to the server along with the video data.

[0620] Server Operation

[0621] The server processes the video and emotion data received from the device. The video data is input into an AI model (e.g., a neural network) that analyzes the sign language movements and converts the corresponding sign language words or phrases into text data. Regarding the emotion data, the emotion engine reflects the analyzed information in the text data and generates text that expresses the emotion.

[0622] The generated text data is converted into text using a natural language processing (NLP) algorithm. The server then uses a multilingual translation function to translate the generated text data into other languages. This translation is performed using the Google Translate API or a proprietary translation AI model. The translated text data is then input back into the sign language generation model to generate sign language text corresponding to the sign language of the other country.

[0623] Sending and displaying translation data

[0624] The server sends the generated translated sign language text to the terminal. The terminal receives the translated sign language text from the server and displays it to the user. The displayed translated text is provided in an appropriate format that is easy for the user to understand.

[0625] Specific examples

[0626] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs with emotion, and the video and emotional data are captured by the device's camera and sent to the server. The server analyzes the video data and converts it into English text data, and the emotion engine reflects the analyzed emotional information in the text. The English text data is then translated into Japanese text that corresponds to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the content of the American sign language and its emotions, achieving smooth communication.

[0627] The system of the present invention enables smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing, while reflecting the user's emotions.

[0628] The processing flow will be explained below.

[0629] Step 1:

[0630] Device behavior

[0631] The user signs into the device's camera. The device's built-in camera captures the sign language video data, and the emotion engine simultaneously analyzes the user's facial expressions and movements to obtain emotional data. The video data and emotional data are temporarily stored in memory.

[0632] Step 2:

[0633] Device behavior

[0634] The device transmits the captured video data and emotion data to the server in real time using communication protocols such as WebSocket and HTTP / 2, and appropriate compression techniques are used to minimize delays during data transmission.

[0635] Step 3:

[0636] Server Operation

[0637] The server receives the video and emotion data sent from the device and stores it in temporary storage. The received data is immediately passed to the processing queue, where analysis begins.

[0638] Step 4:

[0639] Server Operation

[0640] The video data is fed into an AI model that analyzes the sign language movements. The AI ​​model used here is typically a neural network (e.g., CNN or RNN) that recognizes each sign movement and identifies the corresponding sign word or phrase.

[0641] Step 5:

[0642] Server Operation

[0643] After identifying the sign language actions, they are converted into corresponding text data, which is then fed into an NLP algorithm to be reconstructed into natural-sounding sentences.

[0644] Step 6:

[0645] Server Operation

[0646] Emotional data is analyzed to identify the user's emotions. NLP algorithms are used to reflect the emotional data in the text, so that the generated text contains the user's emotional information.

[0647] Step 7:

[0648] Server Operation

[0649] The text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[0650] Step 8:

[0651] Server Operation

[0652] The translated sign language text data is sent to the device, again using a communication protocol (WebSocket or HTTP / 2) to transfer the data in real time.

[0653] Step 9:

[0654] Device behavior

[0655] The terminal receives the translated sign language text data sent from the server and displays it to the user in an appropriate format, which also reflects the original user's emotions.

[0656] Step 10:

[0657] User Actions

[0658] The received translated sign language text data is checked, and a reply is made in sign language if necessary. The reply sign language is also transmitted to the other party through the same steps.

[0659] Specific examples

[0660] For example, consider the case where a hearing-impaired person in the United States (User A) communicates with a hearing-impaired person in Japan (User B). User A uses sign language, and the emotion engine analyzes his or her emotions. This video and emotion data is sent from the device to the server. The server analyzes the data, converts the sign language movements into English text data, and also reflects the emotion data. Next, the English text is translated into Japanese text and Japanese Sign Language. The translation result is sent to User B's device and displayed. This allows User B to understand the content of User A's sign language, including his or her emotions, achieving smooth communication.

[0661] This system enables communication through sign language that reflects the user's emotions, realizing smooth dialogue between hearing-impaired people in different countries and between hearing-impaired people and hearing-speaking people.

[0662] Example 2

[0663] 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."

[0664] There is a lack of efficient means to accurately convey sign language and emotions in communication between hearing-impaired and hearing-disabled people. Conventional sign language translation systems have difficulty conveying rich communication, including emotions, and are inadequate for translating sign language between multiple languages. This makes it difficult for hearing-impaired people to communicate smoothly with users of different cultures and languages.

[0665] 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.

[0666] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data and user emotion data and converting them into corresponding text data, means for analyzing the converted text data including the emotion data and translating it into sign language text for another country, and means for transmitting the translated sign language text. This enables multilingual translation of sign language including emotion, enabling smooth communication between hearing-impaired people and those without hearing.

[0667] "Hearing impaired" refers to a person who has difficulty communicating verbally and uses sign language or other visual methods to communicate.

[0668] "Sign language" is a method of expressing language using visual gestures, hand shapes, and facial expressions, and refers to a means of communication primarily used by people with hearing impairments.

[0669] "Video data" refers to digital data that includes visual information such as a user's sign language actions and facial expressions.

[0670] "Emotion data" refers to information about emotions analyzed from the user's facial expressions and movements.

[0671] "Analysis" refers to the process of analyzing collected data to extract useful information.

[0672] "Text data" refers to digital data that expresses linguistic information as characters.

[0673] "Translation" refers to the process of converting content expressed in one language into another language.

[0674] A "neural network" is an algorithm that mimics the neural circuits of living organisms, and refers specifically to a machine learning model used for image and sound recognition.

[0675] An "emotion engine" refers to a software or hardware system that automatically analyzes emotions from a user's facial expressions and voice.

[0676] "Natural language processing algorithms" refer to computational methods for analyzing and generating language data, and are particularly used for text translation and sentence generation.

[0677] A "sign language generation model" refers to a machine learning model or algorithm for converting text data into sign language.

[0678] A "server" refers to a computer system that provides services or functions to other computers over a network.

[0679] A "terminal" refers to a device that a user directly operates to input and display data, such as a smartphone or PC.

[0680] The system for implementing this invention operates in cooperation with users, terminals, and servers. Its purpose is to acquire video data for hearing-impaired people to communicate using sign language, analyze that data, convert it into corresponding text data, translate it into multiple languages, and then transmit the translated sign language text.

[0681] Hardware and software configuration

[0682] Device: A device that is directly operated by the user, such as a smartphone or PC. The device is equipped with a high-performance camera that captures the user's sign language movements and facial expressions.

[0683] Camera: A camera built into the device, designed to capture high-resolution, real-time video.

[0684] Emotion engine: Installed on the device, it uses technologies such as OpenCV and Emotion AI SDK to recognize emotions from the user's facial expressions and movements.

[0685] Server: A computer system that performs key processes such as data analysis, sign language to text conversion, multilingual translation, etc. The server analyzes the data using neural networks (e.g., models using TensorFlow or PyTorch) or natural language processing algorithms (e.g., spaCy or BERT models).

[0686] System Operation

[0687] This system operates in the following manner.

[0688] 1. User operations

[0689] The user expresses their feelings by signing, for example, "Hello, how are you?"

[0690] 2. Video and Emotion Data Capture

[0691] The device's camera captures the user's sign language and facial expressions, and the emotion engine analyzes the user's emotions from their facial expressions, generating video data of the sign language movements and emotional data.

[0692] 3. Sending data to the server

[0693] The generated video data and emotion data are transmitted to a server in real time using a secure communication protocol (e.g., HTTPS).

[0694] 4. Analysis of video data

[0695] The server inputs the received video data into a neural network, analyzes the sign language movements, and converts them into sign language words and phrases.

[0696] 5. Emotion Data Analysis

[0697] The server reflects emotional information in the text data generated from the sign language based on the emotional data from the emotion engine. For example, it adds the annotation "with a smile" to "Hello, how are you?"

[0698] 6. Multilingual Translation of Text Data

[0699] The server translates the generated text data using the Google Translate API or a proprietary translation AI model. For example, English text is translated into Japanese.

[0700] 7. Sign Language Text Generation

[0701] The translated text data is input into a sign language generation model to generate corresponding sign language text, for example, translated Japanese sign language text.

[0702] 8. Sending and displaying translation data to your device

[0703] The server sends the translated sign language text to the terminal. The terminal displays the received sign language text to the user in an easy-to-understand format. For example, the Japanese Sign Language text is displayed on the terminal of a Japanese user.

[0704] Specific examples

[0705] For example, if a deaf person in the United States wants to sign to a deaf person in Japan, "Hello, how are you?", they can use the following prompt:

[0706] Example prompt: "A deaf person in the United States signs 'Hello, how are you?' with emotion. Please translate this sign into Japanese sign language. Please also reflect the emotion."

[0707] As described above, this system enables multilingual translation of sign language, including emotional expressions, enabling smooth communication between hearing-impaired and hearing-disabled people.

[0708] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0709] Step 1:

[0710] The user signs.

[0711] As a specific action, the user expresses his / her feelings by using sign language. For example, the user uses sign language to say "Hello, how are you?"

[0712] Step 2:

[0713] The device captures video data and emotional data.

[0714] Input and output: The device's camera captures the user's sign language and facial expressions (input), and generates emotional data (output) through the video data and emotion engine.

[0715] Specifically, a high-resolution camera captures sign language movements and facial expressions, and an emotion engine (e.g., OpenCV or Emotion AI SDK) analyzes emotions from the user's facial expressions.

[0716] Step 3:

[0717] The terminal transmits the video data and emotion data to the server.

[0718] Input and Output: The captured video data and emotion data (input) are sent to the server (output) using a secure communication protocol.

[0719] Specifically, the terminal transmits data in real time using a secure communication protocol such as HTTPS.

[0720] Step 4:

[0721] The server analyzes the video data.

[0722] Input and Output: Received video data (input) is fed into the neural network model and converted into sign language words or phrases (output).

[0723] Specifically, the server uses an AI model using TensorFlow and PyTorch to analyze sign language movements and extract the corresponding sign language words and phrases.

[0724] Step 5:

[0725] The server analyzes the emotion data.

[0726] Input and output: Analyze the received emotional data (input) and reflect the emotional information (output) in the text generated from the sign language.

[0727] Specifically, the emotion engine adds the analyzed emotion information to the sign language text. For example, if the sign language is "Hello, how are you?", the emotion information "with a smile" is added.

[0728] Step 6:

[0729] The server translates the text data into multiple languages.

[0730] Input and output: The generated text data (input) is input into the Google Translate API or a proprietary translation AI model to obtain translated text data (output).

[0731] Specifically, the server uses natural language processing (NLP) technology to translate text, for example from English to Japanese.

[0732] Step 7:

[0733] The server inputs the translated text data into a sign language generation model.

[0734] Input and output: The translated text data (input) is input into the sign language generation model to generate sign language text in other countries (output).

[0735] Specifically, the translated text data is passed through a sign language generation model to generate Japanese Sign Language or other sign language text.

[0736] Step 8:

[0737] The server transmits the generated translated sign language text to the terminal.

[0738] Input and output: The generated translated sign language text (input) is sent to the terminal, which displays it to the user (output).

[0739] Specifically, the server sends the generated sign language text to the terminal, and the terminal displays it to the user in an appropriate format. For example, Japanese Sign Language text is displayed on the terminal of a Japanese user.

[0740] (Application example 2)

[0741] 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."

[0742] When hearing-impaired people shop or use services at brick-and-mortar stores, they face the challenge of having difficulty communicating smoothly using sign language. Furthermore, because store clerks cannot understand sign language, they are unable to provide sufficient product explanations or guidance, which often causes inconvenience to hearing-impaired people. Furthermore, they are required to be able to understand and respond to the emotional content of sign language.

[0743] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0744] In this invention, the server includes means for acquiring video data for the hearing impaired to communicate using sign language, means for analyzing the acquired sign language video data and converting it into corresponding text data, means for recognizing the user's emotions, generating emotion data, and reflecting it in text, means for translating the converted text data into sign language text of another country, means for transmitting the translated sign language text, and means for displaying the acquired text data and sign language text, thereby enabling sign language communication that reflects emotions between the hearing impaired and store clerks.

[0745] "Means for acquiring video data" refers to a mechanism for capturing video including sign language movements using devices such as cameras and sensors.

[0746] "Means for analyzing sign language video data" refers to algorithms or software that process the acquired sign language video, recognize sign language actions, and convert them into corresponding text data.

[0747] The "means for converting into corresponding text data" is a processing method for converting sign language expressions into natural language text based on the analyzed sign language video data.

[0748] "Means for translating into sign language text of other countries" refers to a translation algorithm or API for converting the converted text data into sign language text in multiple languages.

[0749] "Means for transmitting sign language text" refers to a communication means for transmitting the translated sign language text to another terminal or server via a network.

[0750] The "means for recognizing user emotions" refers to algorithms or software for identifying a user's emotions and generating emotion data through facial expression recognition and motion analysis.

[0751] The "means for generating emotion data" is a process for generating data for expressing emotions based on the recognized emotion information.

[0752] The "means for reflecting in text" is a processing method for integrating the generated emotion data into text data and generating a text expression that includes emotion.

[0753] The "means for displaying text data and sign language text" is a mechanism for displaying the acquired text data and sign language text on the screen or display of the terminal.

[0754] To implement this invention, the following hardware and software environment is used. The main hardware used is a smartphone, server, camera, and display. The software used is a neural network framework (TensorFlow or PyTorch), a natural language processing (NLP) engine (spaCy, Hugging Face Transformers), an emotion recognition engine (EmotionRecognitionAPI), and a translation API (Google Translate API).

[0755] First, the user signs using the smartphone's camera. The camera captures the sign language movements, and the device's emotion engine analyzes the video data in real time. This analysis recognizes emotions from the user's facial expressions and movements, and generates emotion data. The video data and emotion data are then sent to the server.

[0756] The server analyzes the received video data using a neural network and converts the sign language movements into text data. It also reflects the corresponding emotions based on the emotional data. The generated text data is then compiled into sentences using a natural language processing (NLP) algorithm.

[0757] Next, the server translates the generated text data into other languages ​​using a multilingual translation API. This translated text data is input into a sign language generation model to generate sign language text for other languages. The translated sign language text is then sent from the server to the device.

[0758] When the terminal receives the translated sign language text from the server, it displays it to the user. The displayed text is presented in an appropriate emotional format that is easy for the user to understand. In addition, it plays animations of the sign language, allowing the user to visually understand the sign language.

[0759] Examples of specific prompts include:

[0760] "Analyze the sign language video captured by this camera and generate corresponding text and emotional information. Translate the generated text data into Japanese and convert it into Japanese Sign Language."

[0761] This invention allows hearing-impaired people to easily communicate with store staff when using services in physical stores. Because it does not use voice, it can be used effectively even in quiet environments. Furthermore, by displaying sign language that reflects emotions, it is possible to accurately convey the user's intentions and emotions.

[0762] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0763] Step 1:

[0764] The user signs into the smartphone camera.

[0765] Input: Sign language action

[0766] Output: Sign language video data

[0767] The device's camera captures the user's sign language movements and obtains high-quality video data, which is important for subsequent analysis.

[0768] Step 2:

[0769] The sign language video data acquired by the device is analyzed using an emotion engine to generate emotion data.

[0770] Input: Sign language video data

[0771] Output: Emotion data

[0772] The emotion engine built into the device processes the video data, recognizes emotions from the user's facial expressions and movements, and generates emotion data, which is then used in the translation process.

[0773] Step 3:

[0774] The device transmits sign language video data and emotion data to the server.

[0775] Input: Sign language video data, emotion data

[0776] Output: Data packet to be sent

[0777] Using a high-performance network connection, the device transmits sign language video data and emotion data in real time to a server, where the data packets are analyzed.

[0778] Step 4:

[0779] The server analyzes the sign language video data using a neural network and converts it into corresponding text data.

[0780] Input: Sign language video data

[0781] Output: Corresponding text data

[0782] The server's neural network analyzes sign language movements, identifies corresponding words and phrases, and converts them into text data, using TensorFlow and PyTorch for the process.

[0783] Step 5:

[0784] The server reflects the emotion data in the generated text data.

[0785] Input: Corresponding text data, emotion data

[0786] Output: Text data containing emotions

[0787] The server can incorporate emotional information into text data based on emotion recognition data, thereby conveying the user's intentions more accurately.

[0788] Step 6:

[0789] The server translates the generated text data into other languages ​​using natural language processing algorithms.

[0790] Input: Text data containing emotions

[0791] Output: Translated text data

[0792] The server uses a natural language processing engine (e.g., spaCy or Hugging Face Transformers) to translate the text data into other languages.

[0793] Step 7:

[0794] The server inputs the translated text data into a sign language generation model to generate sign language text for other countries.

[0795] Input: Translated text data

[0796] Output: Foreign sign language text

[0797] The server uses an AI model to generate sign language text corresponding to the sign language of other countries based on the translation data.

[0798] Step 8:

[0799] The server transmits the generated sign language text of the other country to the terminal.

[0800] Input: Foreign sign language text

[0801] Output: Data packet to be sent

[0802] The server transmits the generated sign language text data of the other country to the terminal via the network.

[0803] Step 9:

[0804] The terminal displays the sign language text received from the server to the user.

[0805] Input: Foreign sign language text

[0806] Output: Data for display

[0807] The terminal displays the received data on a display and plays visual sign language animations to help the user understand the content.

[0808] 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.

[0809] 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.

[0810] 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.

[0811] [Third embodiment]

[0812] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0813] 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.

[0814] 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).

[0815] 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.

[0816] 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.

[0817] 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).

[0818] 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.

[0819] 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.

[0820] 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.

[0821] 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.

[0822] 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.

[0823] 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."

[0824] The system of the present invention is a technology that uses terminals such as smartphones and PCs and a server. The specific usage and operation are as follows.

[0825] User operations

[0826] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, so the user can simply sign naturally without any special operations.

[0827] Device behavior

[0828] The device captures video of the user's sign language and sends the video to a server. It is important that the device application has a high-performance camera and network connection to maintain video quality. The video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2) to minimize latency.

[0829] Server Operation

[0830] The server analyzes the video data received from the device. This analysis utilizes a neural network model. The server recognizes the sign language movements and generates text data corresponding to those movements. The generated text data is then further refined through natural language processing algorithms and converted into meaningful text.

[0831] Next, the server uses its multilingual translation function to translate the generated text data into other language data. This translation uses natural language processing (NLP) technology, and sign-language-specific translation models are used to accommodate each country's sign language. Through this process, translated sign language text data is generated.

[0832] Sending and displaying translation data

[0833] The server then sends the translated sign language text back to the device. This transmission is also done in real time, and the device that receives it displays the data on its screen. By checking the displayed translated text, the user can communicate smoothly with sign language users in other countries.

[0834] Specific examples

[0835] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs, and the video is captured by the device's camera and sent to a server. The server analyzes the video data and converts it into English text data, which is then translated into Japanese text corresponding to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American Sign Language, enabling smooth communication.

[0836] The system of the present invention will enable smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing.

[0837] The processing flow will be explained below.

[0838] Step 1:

[0839] Device behavior

[0840] The user signs. The device activates the camera and captures the sign language video in real time. This video data is stored in temporary memory.

[0841] Step 2:

[0842] Device behavior

[0843] The acquired video data is sent to the server in real time, and WebSocket or HTTP / 2 is often used as the communication protocol.

[0844] Step 3:

[0845] Server Operation

[0846] The server receives the video data sent from the device and stores it in temporary storage. As soon as the data is received, it is immediately passed to the processing queue.

[0847] Step 4:

[0848] Server Operation

[0849] The received video data is input into an AI model to analyze the sign language movements, using a convolutional neural network (CNN) or a recurrent neural network (RNN).

[0850] Step 5:

[0851] Server Operation

[0852] The AI ​​model analyzes sign language movements and converts the corresponding sign language words or phrases into text data.

[0853] Step 6:

[0854] Server Operation

[0855] The generated text data is passed to a natural language processing (NLP) component, which composes it into grammatically correct sentences.

[0856] Step 7:

[0857] Server Operation

[0858] Text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[0859] Step 8:

[0860] Server Operation

[0861] The generated translated sign language text is sent to the terminal in real time, again using WebSocket or HTTP / 2 as the communication protocol.

[0862] Step 9:

[0863] Device behavior

[0864] The terminal receives the translated sign language text sent from the server, and displays the received data in an appropriate format so that the user can easily understand it.

[0865] Step 10:

[0866] User Actions

[0867] The translated sign language text data is checked. If necessary, a reply sign is made using the same system. The reply sign is also transmitted to the other party through the same steps.

[0868] Example 1

[0869] 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."

[0870] Conventional systems have had problems with delays and accuracy in sign language understanding and translation when hearing-impaired people communicate with sign language users from other countries who speak different languages. Furthermore, there was a lack of technology to efficiently analyze sign language video data, convert it into text, and translate it into multiple languages. This made it difficult to achieve smooth communication between different languages.

[0871] 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.

[0872] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data in real time and converting it into corresponding text data, means for translating the converted text data into sign language text of another country, and means for transmitting and displaying the translated sign language text, thereby enabling sign language users to communicate between different languages ​​in real time with high accuracy.

[0873] "Hearing impaired" refers to a person who has a reduced or absent ability to hear sound.

[0874] "Sign language" is a visual language that conveys meaning using movements of the hands, fingers, body and face.

[0875] "Video data" is digital data that includes visual information captured by a photographic device such as a camera.

[0876] "Real-time" refers to a state in which processing or reaction occurs immediately without delay.

[0877] "Analysis" is the process of breaking down data and understanding its structure and meaning.

[0878] "Text data" refers to digital data that contains text information.

[0879] "Translation" refers to the conversion of information written in one language into another language.

[0880] A "neural network" is an algorithm that mimics the neural circuits of the human brain, and is an artificial intelligence technology that is particularly used for pattern recognition and learning.

[0881] "Natural Language Processing (NLP)" is a technology that allows computers to understand, interpret, and generate human language.

[0882] A "server" refers to a computer system that provides services to other computers over a network.

[0883] "Terminal" refers to a device that is directly operated by a user and operates as part of a computer network.

[0884] "Transmission" refers to the act of moving data from one place to another.

[0885] "Display" refers to outputting data in a form that is visually recognizable to humans.

[0886] MODE FOR CARRYING OUT THE INVENTION

[0887] The system of the present invention is a technology that enables hearing-impaired people to communicate using sign language, and is a system that uses terminals such as smartphones and PCs and a server. Specific usage and operation are described below.

[0888] User operations

[0889] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, allowing the user to sign naturally without any special operations.

[0890] Device behavior

[0891] The device captures the user's sign language video using a high-performance camera and transmits the video data to a server in real time. Communication protocols such as WebSocket and HTTP / 2 are used to transmit the video data, minimizing latency. Specifically, the device utilizes the smartphone's built-in camera, an external HD camera, Wi-Fi, or 4G / 5G networks to ensure video quality and stable communication.

[0892] Server Operation

[0893] The server uses a deep learning-based neural network model to analyze the video data received from the device. Specifically, a model using TensorFlow and PyTorch asynchronously analyzes sign language movements frame by frame and extracts their features. Based on these features, the server recognizes the sign language movements and generates text data corresponding to those movements.

[0894] The generated text data is then translated into other languages ​​using natural language processing (NLP) technology. For example, the generated text is automatically translated using the Google Translate API or DeepL API. Furthermore, a translation model specialized for sign language is used to accurately convert the unique expressions and meanings of sign language into other languages.

[0895] Sending and displaying translation data

[0896] The server sends the generated translated sign language text to the device in real time. The device then displays the text data on the screen so that the user can confirm it. This display method can be an application UI or a pop-up message. The text displayed on the screen is displayed in a large font for easy viewing.

[0897] Specific examples

[0898] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user makes a sign to say "hello" in sign language, and the video of this is captured by the device's camera and sent to the server. The server analyzes the video data and converts it into text data for the English word "hello," which is then translated into the Japanese text "hello" that corresponds to the Japanese sign language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American sign language and achieve smooth communication.

[0899] Prompt Sentence Examples

[0900] "Please translate American Sign Language into Japanese text and display it."

[0901] This enables the system of the present invention to translate sign language in real time with high accuracy, supporting international communication.

[0902] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0903] Step 1:

[0904] The user begins signing.

[0905] Input: The user's sign language action.

[0906] Action: The user signs naturally towards the camera, specifically to express a greeting such as "hello."

[0907] Output: Video of a user signing.

[0908] Step 2:

[0909] The device captures the sign language video.

[0910] Input: Signing actions of the user seen on camera.

[0911] Actions: The device's camera records the user's sign language movements with high precision. The video is captured frame by frame and stored as digital video data.

[0912] Output: Captured video data of the user's sign language.

[0913] Step 3:

[0914] The terminal transmits the video data to the server.

[0915] Input: Captured video data of the user's sign language.

[0916] How it works: The device uses WebSocket or HTTP / 2 to send video data to the server with minimal latency, using Wi-Fi or 4G / 5G networks.

[0917] Output: Video data sent to the server.

[0918] Step 4:

[0919] The server analyzes the video data.

[0920] Input: Sign language video data sent from the device.

[0921] How it works: The server uses a deep learning-based neural network to analyze sign language movements frame by frame. Specifically, a model using TensorFlow and PyTorch extracts features.

[0922] Output: Analysis results of sign language movements (feature data).

[0923] Step 5:

[0924] The server converts the sign language into text.

[0925] Input: Analysis results of sign language movements (feature data).

[0926] Operation: The server recognizes sign language based on the feature data and generates the corresponding text. For example, the sign language for "hello" is converted into the text data "Hello."

[0927] Output: Generated English text data.

[0928] Step 6:

[0929] The server translates the text into other languages.

[0930] Input: Generated English text data.

[0931] How it works: The server uses natural language processing (NLP) techniques and sign language-specific translation models to translate text into other languages. Specifically, it uses the Google Translate API and DeepL API to convert "Hello" into "Hello."

[0932] Output: Translated Japanese text data.

[0933] Step 7:

[0934] The server sends the translated text to the device.

[0935] Input: Translated Japanese text data.

[0936] How it works: The server uses WebSocket or HTTP / 2 to send translation data to the device in real time.

[0937] Output: Japanese text data sent to the terminal.

[0938] Step 8:

[0939] The device displays the translated text.

[0940] Input: Japanese text data sent to the terminal.

[0941] How it works: Your device will display the translated text on-screen, either in the application UI or as a pop-up message, in a large, easy-to-read font.

[0942] Output: Japanese text "Hello" displayed on the screen.

[0943] Step 9:

[0944] The user checks the translated text.

[0945] Input: Japanese text "Hello" displayed on the screen.

[0946] How it works: The user reads the translated text displayed on the device and understands the sign language of the other country, enabling smooth communication.

[0947] Output: A user who understands the sign language content.

[0948] (Application example 1)

[0949] 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."

[0950] The present invention aims to provide a means for hearing-impaired people to work efficiently in factories. With conventional systems, it is difficult for hearing-impaired people to use sign language to give instructions to work machines and robots, resulting in problems with smooth communication and work efficiency. By solving this problem, the present invention aims to achieve smooth communication between hearing-impaired people working in factories and work machines, thereby improving work efficiency.

[0951] 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.

[0952] In this invention, the server includes means for analyzing the acquired sign language video data and converting it into corresponding text data, means for converting the text data into sign language text in another language using a multilingual translation algorithm, and means for receiving the analyzed sign language data and operating factory work machines. This enables hearing-impaired people to issue instructions to work machines in real time using sign language, facilitating smooth communication within the factory and improving work efficiency.

[0953] "Video data" is digital information captured by a camera of hand movements and gestures used in sign language.

[0954] "Means" refers to a method or device for achieving a specific purpose.

[0955] "Means for acquiring sign language video data" refers to a function that uses cameras, sensors, etc. to record the movements of a person signing and saves them as digital video data.

[0956] The "means for analyzing sign language video data" refers to a technology that processes captured video data, recognizes specific sign language movements, and converts them into corresponding text data.

[0957] "Text data" is character information corresponding to sign language actions, and is data used to communicate intentions.

[0958] "Translation means" is a function for converting text data expressed in one language into text data in another language.

[0959] "Means for sending" refers to the function of sending converted or translated data to other devices or servers.

[0960] "Means of operating factory machinery" means the ability to receive instructions given in sign language to control machinery in a factory and perform specific tasks.

[0961] A "neural network" is a computational algorithm modeled on the neural circuits of the human brain, and is a technology that learns and recognizes specific patterns.

[0962] "Natural language processing algorithms" are technologies that allow computers to understand, generate, and translate human language.

[0963] A system based on an application example of the present invention enables hearing-impaired people to operate work machines in factories using sign language. This system involves a series of processes: acquiring video data of a person signing, analyzing it, converting it into text data necessary for machine control, and operating the work machine based on that text data.

[0964] System Components

[0965] Hardware

[0966] 1. Device camera: Using a device equipped with a high-performance camera, the actions of the person signing are captured in real time.

[0967] 2. Factory machinery: Factory machinery connected to the system, such as robot arms and transport devices.

[0968] software

[0969] 1. Video Analysis Module: A neural network is used to analyze sign language video data and convert it into corresponding text data. This model is built using libraries such as TensorFlow and Keras.

[0970] 2. Natural Language Processing (NLP) Module: This module uses natural language processing algorithms to refine the analyzed text data and convert it into a form suitable for operational instructions. This module includes a natural language processing library with multilingual translation capabilities.

[0971] 3. Robot control module: Includes an API that receives the translated operating instructions and controls the work machines in the factory.

[0972] Operational flow

[0973] User operations

[0974] When using sign language, users simply move their hands toward the device's camera, allowing them to sign naturally without any special operations.

[0975] Device behavior

[0976] The device captures the user's sign language movements in real time through a camera and transmits the video data to a server. This process uses communication protocols (e.g., WebSocket and HTTP / 2) to maintain video quality and minimize latency.

[0977] Server Operation

[0978] The server analyzes the video data received from the device using a neural network to recognize sign language movements. The recognized sign language movements are then converted into text data. This text data is further refined through a natural language processing module and converted into appropriate instructions for controlling factory machinery. These instructions are sent to the factory machinery in real time, and the machinery performs the operations according to the user's instructions.

[0979] Specific examples

[0980] For example, if a hearing-impaired person in a factory signs "go forward," this action is captured by the device's camera and sent as video data to a server. The server analyzes the video data, recognizes it as "go forward," and converts it into corresponding text data. This text data is then converted into an operational command for "go forward" and sent to a work machine (e.g., a transport robot). The transport robot receives this command and moves forward.

[0981] Example of input prompt for generative AI model

[0982] "Specific sign language actions are captured by the camera. Analyze them in real time and translate them into work instructions. For example, if the sign recognizes 'move forward,' give the robot the command to move forward."

[0983] This will enable smooth communication between hearing-impaired people and machines within the factory, improving work efficiency.

[0984] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0985] Step 1:

[0986] The user signs.

[0987] First, the user signs in front of the device's camera. At this time, the user does not need to perform any special operations; they simply sign naturally. The image data input is the user's hand movements.

[0988] Step 2:

[0989] The terminal acquires the video data.

[0990] The device uses a camera to capture the user's sign language movements. The camera acquires video data in real time and processes it to maintain high quality. The input on the device is the video data captured by the camera, and the output is the same video data.

[0991] Step 3:

[0992] The terminal transmits the video data to the server.

[0993] The captured video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2). The device manages the transmission process to minimize the delay of the video data. The input is the captured video data, and the output is the data sent to the server.

[0994] Step 4:

[0995] The server analyzes the video data.

[0996] The server analyzes the received video data using a neural network model to recognize sign language movements. Specifically, the video frames are input into the neural network model, and the sign language movements are converted into text data. The input is the video data sent from the device, and the output is the analyzed text data.

[0997] Step 5:

[0998] The server translates the text data.

[0999] The server uses natural language processing algorithms to translate the acquired text data into sign language text in other languages, and then converts it into appropriate operating instructions for the factory's work machines. The input is the analyzed text data, and the output is the translated operating instructions.

[1000] Step 6:

[1001] The server transmits the operation instruction data.

[1002] The server then transmits the translated operation instruction data to the factory machine using a communication protocol. This process also takes place in real time. The input is the translated operation instruction data, and the output is the data sent to the factory machine.

[1003] Step 7:

[1004] The factory's work machines carry out the operation instructions.

[1005] The factory's work machines receive the operation instruction data sent from the server and perform specific operations based on it. For example, if a transport robot receives the instruction to "move forward," it will move forward. The input is the operation instruction data sent from the server, and the output is the corresponding machine operation.

[1006] This enables users to control factory machinery in real time using sign language, enabling smooth communication between hearing-impaired people and factory machinery.

[1007] 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.

[1008] The system of the present invention is a technology that combines a terminal such as a smartphone or PC, a server, and an emotion engine that recognizes the user's emotions. The specific implementation method and operation flow are as follows.

[1009] User operations

[1010] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures this sign language movement. The device is equipped with an emotion engine that recognizes emotions from the user's facial expressions and movements. This video data is analyzed in real time and sent to a server.

[1011] Device behavior

[1012] The device captures video of the user's sign language and sends the video data to the server. It is important that the device application has a high-performance camera and network connection to maintain video quality. In addition, an emotion engine analyzes the user's facial expressions and movements to extract emotional data. This emotional data is also sent to the server along with the video data.

[1013] Server Operation

[1014] The server processes the video and emotion data received from the device. The video data is input into an AI model (e.g., a neural network) that analyzes the sign language movements and converts the corresponding sign language words or phrases into text data. Regarding the emotion data, the emotion engine reflects the analyzed information in the text data and generates text that expresses the emotion.

[1015] The generated text data is converted into text using a natural language processing (NLP) algorithm. The server then uses a multilingual translation function to translate the generated text data into other languages. This translation is performed using the Google Translate API or a proprietary translation AI model. The translated text data is then input back into the sign language generation model to generate sign language text corresponding to the sign language of the other country.

[1016] Sending and displaying translation data

[1017] The server sends the generated translated sign language text to the terminal. The terminal receives the translated sign language text from the server and displays it to the user. The displayed translated text is provided in an appropriate format that is easy for the user to understand.

[1018] Specific examples

[1019] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs with emotion, and the video and emotional data are captured by the device's camera and sent to the server. The server analyzes the video data and converts it into English text data, and the emotion engine reflects the analyzed emotional information in the text. The English text data is then translated into Japanese text that corresponds to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the content of the American sign language and its emotions, achieving smooth communication.

[1020] The system of the present invention enables smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing, while reflecting the user's emotions.

[1021] The processing flow will be explained below.

[1022] Step 1:

[1023] Device behavior

[1024] The user signs into the device's camera. The device's built-in camera captures the sign language video data, and the emotion engine simultaneously analyzes the user's facial expressions and movements to obtain emotional data. The video data and emotional data are temporarily stored in memory.

[1025] Step 2:

[1026] Device behavior

[1027] The device transmits the captured video data and emotion data to the server in real time using communication protocols such as WebSocket and HTTP / 2, and appropriate compression techniques are used to minimize delays during data transmission.

[1028] Step 3:

[1029] Server Operation

[1030] The server receives the video and emotion data sent from the device and stores it in temporary storage. The received data is immediately passed to the processing queue, where analysis begins.

[1031] Step 4:

[1032] Server Operation

[1033] The video data is fed into an AI model that analyzes the sign language movements. The AI ​​model used here is typically a neural network (e.g., CNN or RNN) that recognizes each sign movement and identifies the corresponding sign word or phrase.

[1034] Step 5:

[1035] Server Operation

[1036] After identifying the sign language actions, they are converted into corresponding text data, which is then fed into an NLP algorithm to be reconstructed into natural-sounding sentences.

[1037] Step 6:

[1038] Server Operation

[1039] Emotional data is analyzed to identify the user's emotions. NLP algorithms are used to reflect the emotional data in the text, so that the generated text contains the user's emotional information.

[1040] Step 7:

[1041] Server Operation

[1042] The text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[1043] Step 8:

[1044] Server Operation

[1045] The translated sign language text data is sent to the device, again using a communication protocol (WebSocket or HTTP / 2) to transfer the data in real time.

[1046] Step 9:

[1047] Device behavior

[1048] The terminal receives the translated sign language text data sent from the server and displays it to the user in an appropriate format, which also reflects the original user's emotions.

[1049] Step 10:

[1050] User Actions

[1051] The received translated sign language text data is checked, and a reply is made in sign language if necessary. The reply sign language is also transmitted to the other party through the same steps.

[1052] Specific examples

[1053] For example, consider the case where a hearing-impaired person in the United States (User A) communicates with a hearing-impaired person in Japan (User B). User A uses sign language, and the emotion engine analyzes his or her emotions. This video and emotion data is sent from the device to the server. The server analyzes the data, converts the sign language movements into English text data, and also reflects the emotion data. Next, the English text is translated into Japanese text and Japanese Sign Language. The translation result is sent to User B's device and displayed. This allows User B to understand the content of User A's sign language, including his or her emotions, achieving smooth communication.

[1054] This system enables communication through sign language that reflects the user's emotions, realizing smooth dialogue between hearing-impaired people in different countries and between hearing-impaired people and hearing-speaking people.

[1055] Example 2

[1056] 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."

[1057] There is a lack of efficient means to accurately convey sign language and emotions in communication between hearing-impaired and hearing-disabled people. Conventional sign language translation systems have difficulty conveying rich communication, including emotions, and are inadequate for translating sign language between multiple languages. This makes it difficult for hearing-impaired people to communicate smoothly with users of different cultures and languages.

[1058] 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.

[1059] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data and user emotion data and converting them into corresponding text data, means for analyzing the converted text data including the emotion data and translating it into sign language text for another country, and means for transmitting the translated sign language text. This enables multilingual translation of sign language including emotion, enabling smooth communication between hearing-impaired people and those without hearing.

[1060] "Hearing impaired" refers to a person who has difficulty communicating verbally and uses sign language or other visual methods to communicate.

[1061] "Sign language" is a method of expressing language using visual gestures, hand shapes, and facial expressions, and refers to a means of communication primarily used by people with hearing impairments.

[1062] "Video data" refers to digital data that includes visual information such as a user's sign language actions and facial expressions.

[1063] "Emotion data" refers to information about emotions analyzed from the user's facial expressions and movements.

[1064] "Analysis" refers to the process of analyzing collected data to extract useful information.

[1065] "Text data" refers to digital data that expresses linguistic information as characters.

[1066] "Translation" refers to the process of converting content expressed in one language into another language.

[1067] A "neural network" is an algorithm that mimics the neural circuits of living organisms, and refers to a machine learning model that is particularly used for recognizing images and sounds.

[1068] An "emotion engine" refers to a software or hardware system that automatically analyzes emotions from a user's facial expressions and voice.

[1069] "Natural language processing algorithms" refer to computational methods for analyzing and generating language data, and are particularly used for text translation and sentence generation.

[1070] A "sign language generation model" refers to a machine learning model or algorithm for converting text data into sign language.

[1071] A "server" refers to a computer system that provides services or functions to other computers over a network.

[1072] A "terminal" refers to a device that a user directly operates to input and display data, such as a smartphone or PC.

[1073] The system for implementing this invention operates in cooperation with users, terminals, and servers. Its purpose is to acquire video data for hearing-impaired people to communicate using sign language, analyze that data, convert it into corresponding text data, translate it into multiple languages, and then transmit the translated sign language text.

[1074] Hardware and software configuration

[1075] Device: A device that is directly operated by the user, such as a smartphone or PC. The device is equipped with a high-performance camera that captures the user's sign language movements and facial expressions.

[1076] Camera: A camera built into the device, designed to capture high-resolution, real-time video.

[1077] Emotion engine: Installed on the device, it uses technologies such as OpenCV and Emotion AI SDK to recognize emotions from the user's facial expressions and movements.

[1078] Server: A computer system that performs key processes such as data analysis, sign language to text conversion, multilingual translation, etc. The server analyzes the data using neural networks (e.g., models using TensorFlow or PyTorch) or natural language processing algorithms (e.g., spaCy or BERT models).

[1079] System Operation

[1080] This system operates in the following manner.

[1081] 1. User operations

[1082] The user expresses their feelings by signing, for example, "Hello, how are you?"

[1083] 2. Video and Emotion Data Capture

[1084] The device's camera captures the user's sign language and facial expressions, and the emotion engine analyzes the user's emotions from their facial expressions, generating video data of the sign language movements and emotional data.

[1085] 3. Sending data to the server

[1086] The generated video data and emotion data are transmitted to a server in real time using a secure communication protocol (e.g., HTTPS).

[1087] 4. Analysis of video data

[1088] The server inputs the received video data into a neural network, analyzes the sign language movements, and converts them into sign language words and phrases.

[1089] 5. Emotion Data Analysis

[1090] The server reflects emotional information in the text data generated from the sign language based on the emotional data from the emotion engine. For example, it adds the annotation "with a smile" to "Hello, how are you?"

[1091] 6. Multilingual Translation of Text Data

[1092] The server translates the generated text data using the Google Translate API or a proprietary translation AI model. For example, English text is translated into Japanese.

[1093] 7. Sign Language Text Generation

[1094] The translated text data is input into a sign language generation model to generate corresponding sign language text, for example, translated Japanese sign language text.

[1095] 8. Sending and displaying translation data to your device

[1096] The server sends the translated sign language text to the terminal. The terminal displays the received sign language text to the user in an easy-to-understand format. For example, the Japanese Sign Language text is displayed on the terminal of a Japanese user.

[1097] Specific examples

[1098] For example, if a deaf person in the United States wants to sign to a deaf person in Japan, "Hello, how are you?", they can use the following prompt:

[1099] Example prompt: "A deaf person in the United States signs 'Hello, how are you?' with emotion. Please translate this sign into Japanese sign language. Please also reflect the emotion."

[1100] As described above, this system enables multilingual translation of sign language, including emotional expressions, enabling smooth communication between hearing-impaired and hearing-disabled people.

[1101] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1102] Step 1:

[1103] The user signs.

[1104] As a specific action, the user expresses his / her feelings by using sign language. For example, the user uses sign language to say "Hello, how are you?"

[1105] Step 2:

[1106] The device captures video data and emotional data.

[1107] Input and output: The device's camera captures the user's sign language and facial expressions (input), and generates emotional data (output) through the video data and emotion engine.

[1108] Specifically, a high-resolution camera captures sign language movements and facial expressions, and an emotion engine (e.g., OpenCV or Emotion AI SDK) analyzes emotions from the user's facial expressions.

[1109] Step 3:

[1110] The terminal transmits the video data and emotion data to the server.

[1111] Input and Output: The captured video data and emotion data (input) are sent to the server (output) using a secure communication protocol.

[1112] Specifically, the terminal transmits data in real time using a secure communication protocol such as HTTPS.

[1113] Step 4:

[1114] The server analyzes the video data.

[1115] Input and Output: Received video data (input) is fed into the neural network model and converted into sign language words or phrases (output).

[1116] Specifically, the server uses an AI model using TensorFlow and PyTorch to analyze sign language movements and extract the corresponding sign language words and phrases.

[1117] Step 5:

[1118] The server analyzes the emotion data.

[1119] Input and output: Analyze the received emotional data (input) and reflect the emotional information (output) in the text generated from the sign language.

[1120] Specifically, the emotion engine adds the analyzed emotion information to the sign language text. For example, if the sign language is "Hello, how are you?", the emotion information "with a smile" is added.

[1121] Step 6:

[1122] The server translates the text data into multiple languages.

[1123] Input and output: The generated text data (input) is input into the Google Translate API or a proprietary translation AI model to obtain translated text data (output).

[1124] Specifically, the server uses natural language processing (NLP) technology to translate text, for example from English to Japanese.

[1125] Step 7:

[1126] The server inputs the translated text data into a sign language generation model.

[1127] Input and output: The translated text data (input) is input into the sign language generation model to generate sign language text in other countries (output).

[1128] Specifically, the translated text data is passed through a sign language generation model to generate Japanese Sign Language or other sign language text.

[1129] Step 8:

[1130] The server transmits the generated translated sign language text to the terminal.

[1131] Input and output: The generated translated sign language text (input) is sent to the terminal, which displays it to the user (output).

[1132] Specifically, the server sends the generated sign language text to the terminal, and the terminal displays it to the user in an appropriate format. For example, Japanese Sign Language text is displayed on the terminal of a Japanese user.

[1133] (Application example 2)

[1134] 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."

[1135] When hearing-impaired people shop or use services at brick-and-mortar stores, they face the challenge of having difficulty communicating smoothly using sign language. Furthermore, because store clerks cannot understand sign language, they are unable to provide sufficient product explanations or guidance, which often causes inconvenience to hearing-impaired people. Furthermore, they are required to be able to understand and respond to the emotional content of sign language.

[1136] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1137] In this invention, the server includes means for acquiring video data for the hearing impaired to communicate using sign language, means for analyzing the acquired sign language video data and converting it into corresponding text data, means for recognizing the user's emotions, generating emotion data, and reflecting it in text, means for translating the converted text data into sign language text of another country, means for transmitting the translated sign language text, and means for displaying the acquired text data and sign language text, thereby enabling sign language communication that reflects emotions between the hearing impaired and store clerks.

[1138] "Means for acquiring video data" refers to a mechanism for capturing video including sign language movements using devices such as cameras and sensors.

[1139] "Means for analyzing sign language video data" refers to algorithms or software that process the acquired sign language video, recognize sign language actions, and convert them into corresponding text data.

[1140] The "means for converting into corresponding text data" is a processing method for converting sign language expressions into natural language text based on the analyzed sign language video data.

[1141] "Means for translating into sign language text of other countries" refers to a translation algorithm or API for converting the converted text data into sign language text in multiple languages.

[1142] "Means for transmitting sign language text" refers to a communication means for transmitting the translated sign language text to another terminal or server via a network.

[1143] The "means for recognizing user emotions" refers to algorithms or software for identifying a user's emotions and generating emotion data through facial expression recognition and motion analysis.

[1144] The "means for generating emotion data" is a process for generating data for expressing emotions based on the recognized emotion information.

[1145] The "means for reflecting in text" is a processing method for integrating the generated emotion data into text data and generating a text expression that includes emotion.

[1146] The "means for displaying text data and sign language text" is a mechanism for displaying the acquired text data and sign language text on the screen or display of the terminal.

[1147] To implement this invention, the following hardware and software environment is used. The main hardware used is a smartphone, server, camera, and display. The software used is a neural network framework (TensorFlow or PyTorch), a natural language processing (NLP) engine (spaCy, Hugging Face Transformers), an emotion recognition engine (EmotionRecognitionAPI), and a translation API (Google Translate API).

[1148] First, the user signs using the smartphone's camera. The camera captures the sign language movements, and the device's emotion engine analyzes the video data in real time. This analysis recognizes emotions from the user's facial expressions and movements, and generates emotion data. The video data and emotion data are then sent to the server.

[1149] The server analyzes the received video data using a neural network and converts the sign language movements into text data. It also reflects the corresponding emotions based on the emotional data. The generated text data is then compiled into sentences using a natural language processing (NLP) algorithm.

[1150] Next, the server translates the generated text data into other languages ​​using a multilingual translation API. This translated text data is input into a sign language generation model to generate sign language text for other languages. The translated sign language text is then sent from the server to the device.

[1151] When the terminal receives the translated sign language text from the server, it displays it to the user. The displayed text is presented in an appropriate emotional format that is easy for the user to understand. In addition, it plays animations of the sign language, allowing the user to visually understand the sign language.

[1152] Examples of specific prompts include:

[1153] "Analyze the sign language video captured by this camera and generate corresponding text and emotional information. Translate the generated text data into Japanese and convert it into Japanese Sign Language."

[1154] This invention allows hearing-impaired people to easily communicate with store staff when using services in physical stores. Because it does not use voice, it can be used effectively even in quiet environments. Furthermore, by displaying sign language that reflects emotions, it is possible to accurately convey the user's intentions and emotions.

[1155] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1156] Step 1:

[1157] The user signs into the smartphone camera.

[1158] Input: Sign language action

[1159] Output: Sign language video data

[1160] The device's camera captures the user's sign language movements and obtains high-quality video data, which is important for subsequent analysis.

[1161] Step 2:

[1162] The sign language video data acquired by the device is analyzed using an emotion engine to generate emotion data.

[1163] Input: Sign language video data

[1164] Output: Emotion data

[1165] The emotion engine built into the device processes the video data, recognizes emotions from the user's facial expressions and movements, and generates emotion data, which is then used in the translation process.

[1166] Step 3:

[1167] The device transmits sign language video data and emotion data to the server.

[1168] Input: Sign language video data, emotion data

[1169] Output: Data packet to be sent

[1170] Using a high-performance network connection, the device transmits sign language video data and emotion data in real time to a server, where the data packets are analyzed.

[1171] Step 4:

[1172] The server analyzes the sign language video data using a neural network and converts it into corresponding text data.

[1173] Input: Sign language video data

[1174] Output: Corresponding text data

[1175] The server's neural network analyzes sign language movements, identifies corresponding words and phrases, and converts them into text data, using TensorFlow and PyTorch for the process.

[1176] Step 5:

[1177] The server reflects the emotion data in the generated text data.

[1178] Input: Corresponding text data, emotion data

[1179] Output: Text data containing emotions

[1180] The server can incorporate emotional information into text data based on emotion recognition data, thereby conveying the user's intentions more accurately.

[1181] Step 6:

[1182] The server translates the generated text data into other languages ​​using natural language processing algorithms.

[1183] Input: Text data containing emotions

[1184] Output: Translated text data

[1185] The server uses a natural language processing engine (e.g., spaCy or Hugging Face Transformers) to translate the text data into other languages.

[1186] Step 7:

[1187] The server inputs the translated text data into a sign language generation model to generate sign language text for other countries.

[1188] Input: Translated text data

[1189] Output: Foreign sign language text

[1190] The server uses an AI model to generate sign language text corresponding to the sign language of other countries based on the translation data.

[1191] Step 8:

[1192] The server transmits the generated sign language text of the other country to the terminal.

[1193] Input: Foreign sign language text

[1194] Output: Data packet to be sent

[1195] The server transmits the generated sign language text data of the other country to the terminal via the network.

[1196] Step 9:

[1197] The terminal displays the sign language text received from the server to the user.

[1198] Input: Foreign sign language text

[1199] Output: Data for display

[1200] The terminal displays the received data on a display and plays visual sign language animations to help the user understand the content.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] [Fourth embodiment]

[1205] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1206] 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.

[1207] 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).

[1208] 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.

[1209] 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.

[1210] 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).

[1211] 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.

[1212] 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.

[1213] 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.

[1214] 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.

[1215] 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.

[1216] 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.

[1217] 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."

[1218] The system of the present invention is a technology that uses terminals such as smartphones and PCs and a server. The specific usage and operation are as follows.

[1219] User operations

[1220] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, so the user can simply sign naturally without any special operations.

[1221] Device behavior

[1222] The device captures video of the user's sign language and sends the video to a server. It is important that the device application has a high-performance camera and network connection to maintain video quality. The video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2) to minimize latency.

[1223] Server Operation

[1224] The server analyzes the video data received from the device. This analysis utilizes a neural network model. The server recognizes the sign language movements and generates text data corresponding to those movements. The generated text data is then further refined through natural language processing algorithms and converted into meaningful text.

[1225] Next, the server uses its multilingual translation function to translate the generated text data into other language data. This translation uses natural language processing (NLP) technology, and sign-language-specific translation models are used to accommodate each country's sign language. Through this process, translated sign language text data is generated.

[1226] Sending and displaying translation data

[1227] The server then sends the translated sign language text back to the device. This transmission is also done in real time, and the device that receives it displays the data on its screen. By checking the displayed translated text, the user can communicate smoothly with sign language users in other countries.

[1228] Specific examples

[1229] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs, and the video is captured by the device's camera and sent to a server. The server analyzes the video data and converts it into English text data, which is then translated into Japanese text corresponding to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American Sign Language, enabling smooth communication.

[1230] The system of the present invention will enable smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing.

[1231] The processing flow will be explained below.

[1232] Step 1:

[1233] Device behavior

[1234] The user signs. The device activates the camera and captures the sign language video in real time. This video data is stored in temporary memory.

[1235] Step 2:

[1236] Device behavior

[1237] The acquired video data is sent to the server in real time, and WebSocket or HTTP / 2 is often used as the communication protocol.

[1238] Step 3:

[1239] Server Operation

[1240] The server receives the video data sent from the device and stores it in temporary storage. As soon as the data is received, it is immediately passed to the processing queue.

[1241] Step 4:

[1242] Server Operation

[1243] The received video data is input into an AI model to analyze the sign language movements, using a convolutional neural network (CNN) or a recurrent neural network (RNN).

[1244] Step 5:

[1245] Server Operation

[1246] The AI ​​model analyzes sign language movements and converts the corresponding sign language words or phrases into text data.

[1247] Step 6:

[1248] Server Operation

[1249] The generated text data is passed to a natural language processing (NLP) component, which composes it into grammatically correct sentences.

[1250] Step 7:

[1251] Server Operation

[1252] Text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[1253] Step 8:

[1254] Server Operation

[1255] The generated translated sign language text is sent to the terminal in real time, again using WebSocket or HTTP / 2 as the communication protocol.

[1256] Step 9:

[1257] Device behavior

[1258] The terminal receives the translated sign language text sent from the server, and displays the received data in an appropriate format so that the user can easily understand it.

[1259] Step 10:

[1260] User Actions

[1261] The translated sign language text data is checked. If necessary, a reply sign is made using the same system. The reply sign is also transmitted to the other party through the same steps.

[1262] Example 1

[1263] 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."

[1264] Conventional systems have had problems with delays and accuracy in sign language understanding and translation when hearing-impaired people communicate with sign language users from other countries who speak different languages. Furthermore, there was a lack of technology to efficiently analyze sign language video data, convert it into text, and translate it into multiple languages. This made it difficult to achieve smooth communication between different languages.

[1265] 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.

[1266] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data in real time and converting it into corresponding text data, means for translating the converted text data into sign language text of another country, and means for transmitting and displaying the translated sign language text, thereby enabling sign language users to communicate between different languages ​​in real time with high accuracy.

[1267] "Hearing impaired" refers to a person who has a reduced or absent ability to hear sound.

[1268] "Sign language" is a visual language that conveys meaning using movements of the hands, fingers, body and face.

[1269] "Video data" is digital data that includes visual information captured by a photographic device such as a camera.

[1270] "Real-time" refers to a state in which processing or reaction occurs immediately without delay.

[1271] "Analysis" is the process of breaking down data and understanding its structure and meaning.

[1272] "Text data" refers to digital data that contains text information.

[1273] "Translation" refers to the conversion of information written in one language into another language.

[1274] A "neural network" is an algorithm that mimics the neural circuits of the human brain, and is an artificial intelligence technology that is particularly used for pattern recognition and learning.

[1275] "Natural Language Processing (NLP)" is a technology that allows computers to understand, interpret, and generate human language.

[1276] A "server" refers to a computer system that provides services to other computers over a network.

[1277] "Terminal" refers to a device that is directly operated by a user and operates as part of a computer network.

[1278] "Transmission" refers to the act of moving data from one place to another.

[1279] "Display" refers to outputting data in a form that is visually recognizable to humans.

[1280] MODE FOR CARRYING OUT THE INVENTION

[1281] The system of the present invention is a technology that enables hearing-impaired people to communicate using sign language, and is a system that uses terminals such as smartphones and PCs and a server. Specific usage and operation are described below.

[1282] User operations

[1283] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures the sign movements. The video data is analyzed in real time, allowing the user to sign naturally without any special operations.

[1284] Device behavior

[1285] The device captures the user's sign language video using a high-performance camera and transmits the video data to a server in real time. Communication protocols such as WebSocket and HTTP / 2 are used to transmit the video data, minimizing latency. Specifically, the device utilizes the smartphone's built-in camera, an external HD camera, Wi-Fi, or 4G / 5G networks to ensure video quality and stable communication.

[1286] Server Operation

[1287] The server uses a deep learning-based neural network model to analyze the video data received from the device. Specifically, a model using TensorFlow and PyTorch asynchronously analyzes sign language movements frame by frame and extracts their features. Based on these features, the server recognizes the sign language movements and generates text data corresponding to those movements.

[1288] The generated text data is then translated into other languages ​​using natural language processing (NLP) technology. For example, the generated text is automatically translated using the Google Translate API or DeepL API. Furthermore, a translation model specialized for sign language is used to accurately convert the unique expressions and meanings of sign language into other languages.

[1289] Sending and displaying translation data

[1290] The server sends the generated translated sign language text to the device in real time. The device then displays the text data on the screen so that the user can confirm it. This display method can be an application UI or a pop-up message. The text displayed on the screen is displayed in a large font for easy viewing.

[1291] Specific examples

[1292] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user makes a sign to say "hello" in sign language, and the video of this is captured by the device's camera and sent to the server. The server analyzes the video data and converts it into text data for the English word "hello," which is then translated into the Japanese text "hello" that corresponds to the Japanese sign language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the American sign language and achieve smooth communication.

[1293] Prompt Sentence Examples

[1294] "Please translate American Sign Language into Japanese text and display it."

[1295] This enables the system of the present invention to translate sign language in real time with high accuracy, supporting international communication.

[1296] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1297] Step 1:

[1298] The user begins signing.

[1299] Input: The user's sign language action.

[1300] Action: The user signs naturally towards the camera, specifically to express a greeting such as "hello."

[1301] Output: Video of a user signing.

[1302] Step 2:

[1303] The device captures the sign language video.

[1304] Input: Signing actions of the user seen on camera.

[1305] Actions: The device's camera records the user's sign language movements with high precision. The video is captured frame by frame and stored as digital video data.

[1306] Output: Captured video data of the user's sign language.

[1307] Step 3:

[1308] The terminal transmits the video data to the server.

[1309] Input: Captured video data of the user's sign language.

[1310] How it works: The device uses WebSocket or HTTP / 2 to send video data to the server with minimal latency, using Wi-Fi or 4G / 5G networks.

[1311] Output: Video data sent to the server.

[1312] Step 4:

[1313] The server analyzes the video data.

[1314] Input: Sign language video data sent from the device.

[1315] How it works: The server uses a deep learning-based neural network to analyze sign language movements frame by frame. Specifically, a model using TensorFlow and PyTorch extracts features.

[1316] Output: Analysis results of sign language movements (feature data).

[1317] Step 5:

[1318] The server converts the sign language into text.

[1319] Input: Analysis results of sign language movements (feature data).

[1320] Operation: The server recognizes sign language based on the feature data and generates the corresponding text. For example, the sign language for "hello" is converted into the text data "Hello."

[1321] Output: Generated English text data.

[1322] Step 6:

[1323] The server translates the text into other languages.

[1324] Input: Generated English text data.

[1325] How it works: The server uses natural language processing (NLP) techniques and sign language-specific translation models to translate text into other languages. Specifically, it uses the Google Translate API and DeepL API to convert "Hello" into "Hello."

[1326] Output: Translated Japanese text data.

[1327] Step 7:

[1328] The server sends the translated text to the device.

[1329] Input: Translated Japanese text data.

[1330] How it works: The server uses WebSocket or HTTP / 2 to send translation data to the device in real time.

[1331] Output: Japanese text data sent to the terminal.

[1332] Step 8:

[1333] The device displays the translated text.

[1334] Input: Japanese text data sent to the terminal.

[1335] How it works: Your device will display the translated text on-screen, either in the application UI or as a pop-up message, in a large, easy-to-read font.

[1336] Output: Japanese text "Hello" displayed on the screen.

[1337] Step 9:

[1338] The user checks the translated text.

[1339] Input: Japanese text "Hello" displayed on the screen.

[1340] How it works: The user reads the translated text displayed on the device and understands the sign language of the other country, enabling smooth communication.

[1341] Output: A user who understands the sign language content.

[1342] (Application example 1)

[1343] 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."

[1344] The present invention aims to provide a means for hearing-impaired people to work efficiently in factories. With conventional systems, it is difficult for hearing-impaired people to use sign language to give instructions to work machines and robots, resulting in problems with smooth communication and work efficiency. By solving this problem, the present invention aims to achieve smooth communication between hearing-impaired people working in factories and work machines, thereby improving work efficiency.

[1345] 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.

[1346] In this invention, the server includes means for analyzing the acquired sign language video data and converting it into corresponding text data, means for converting the text data into sign language text in another language using a multilingual translation algorithm, and means for receiving the analyzed sign language data and operating factory work machines. This enables hearing-impaired people to issue instructions to work machines in real time using sign language, facilitating smooth communication within the factory and improving work efficiency.

[1347] "Video data" is digital information captured by a camera of hand movements and gestures used in sign language.

[1348] "Means" refers to a method or device for achieving a specific purpose.

[1349] "Means for acquiring sign language video data" refers to a function that uses cameras, sensors, etc. to record the movements of a person signing and saves them as digital video data.

[1350] The "means for analyzing sign language video data" refers to a technology that processes captured video data, recognizes specific sign language movements, and converts them into corresponding text data.

[1351] "Text data" is character information corresponding to sign language actions, and is data used to communicate intentions.

[1352] "Translation means" is a function for converting text data expressed in one language into text data in another language.

[1353] "Means for sending" refers to the function of sending converted or translated data to other devices or servers.

[1354] "Means of operating factory machinery" means the ability to receive instructions given in sign language to control machinery in a factory and perform specific tasks.

[1355] A "neural network" is a computational algorithm modeled on the neural circuits of the human brain, and is a technology that learns and recognizes specific patterns.

[1356] "Natural language processing algorithms" are technologies that allow computers to understand, generate, and translate human language.

[1357] A system based on an application example of the present invention enables hearing-impaired people to operate work machines in factories using sign language. This system involves a series of processes: acquiring video data of a person signing, analyzing it, converting it into text data necessary for machine control, and operating the work machine based on that text data.

[1358] System Components

[1359] Hardware

[1360] 1. Device camera: Using a device equipped with a high-performance camera, the actions of the person signing are captured in real time.

[1361] 2. Factory machinery: Factory machinery connected to the system, such as robot arms and transport devices.

[1362] software

[1363] 1. Video Analysis Module: A neural network is used to analyze sign language video data and convert it into corresponding text data. This model is built using libraries such as TensorFlow and Keras.

[1364] 2. Natural Language Processing (NLP) Module: This module uses natural language processing algorithms to refine the analyzed text data and convert it into a form suitable for operational instructions. This module includes a natural language processing library with multilingual translation capabilities.

[1365] 3. Robot control module: Includes an API that receives the translated operating instructions and controls the work machines in the factory.

[1366] Operational flow

[1367] User operations

[1368] When using sign language, users simply move their hands toward the device's camera, allowing them to sign naturally without any special operations.

[1369] Device behavior

[1370] The device captures the user's sign language movements in real time through a camera and transmits the video data to a server. This process uses communication protocols (e.g., WebSocket and HTTP / 2) to maintain video quality and minimize latency.

[1371] Server Operation

[1372] The server analyzes the video data received from the device using a neural network to recognize sign language movements. The recognized sign language movements are then converted into text data. This text data is further refined through a natural language processing module and converted into appropriate instructions for controlling factory machinery. These instructions are sent to the factory machinery in real time, and the machinery performs the operations according to the user's instructions.

[1373] Specific examples

[1374] For example, if a hearing-impaired person in a factory signs "go forward," this action is captured by the device's camera and sent as video data to a server. The server analyzes the video data, recognizes it as "go forward," and converts it into corresponding text data. This text data is then converted into an operational command for "go forward" and sent to a work machine (e.g., a transport robot). The transport robot receives this command and moves forward.

[1375] Example of input prompt for generative AI model

[1376] "Specific sign language actions are captured by the camera. Analyze them in real time and translate them into work instructions. For example, if the sign recognizes 'move forward,' give the robot the command to move forward."

[1377] This will enable smooth communication between hearing-impaired people and machines within the factory, improving work efficiency.

[1378] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1379] Step 1:

[1380] The user signs.

[1381] First, the user signs in front of the device's camera. At this time, the user does not need to perform any special operations; they simply sign naturally. The image data input is the user's hand movements.

[1382] Step 2:

[1383] The terminal acquires the video data.

[1384] The device uses a camera to capture the user's sign language movements. The camera acquires video data in real time and processes it to maintain high quality. The input on the device is the video data captured by the camera, and the output is the same video data.

[1385] Step 3:

[1386] The terminal transmits the video data to the server.

[1387] The captured video data is sent to the server using a communication protocol (e.g., WebSocket or HTTP / 2). The device manages the transmission process to minimize the delay of the video data. The input is the captured video data, and the output is the data sent to the server.

[1388] Step 4:

[1389] The server analyzes the video data.

[1390] The server analyzes the received video data using a neural network model to recognize sign language movements. Specifically, the video frames are input into the neural network model, and the sign language movements are converted into text data. The input is the video data sent from the device, and the output is the analyzed text data.

[1391] Step 5:

[1392] The server translates the text data.

[1393] The server uses natural language processing algorithms to translate the acquired text data into sign language text in other languages, and then converts it into appropriate operating instructions for the factory's work machines. The input is the analyzed text data, and the output is the translated operating instructions.

[1394] Step 6:

[1395] The server transmits the operation instruction data.

[1396] The server then transmits the translated operation instruction data to the factory machine using a communication protocol. This process also takes place in real time. The input is the translated operation instruction data, and the output is the data sent to the factory machine.

[1397] Step 7:

[1398] The factory's work machines carry out the operation instructions.

[1399] The factory's work machines receive the operation instruction data sent from the server and perform specific operations based on it. For example, if a transport robot receives the instruction to "move forward," it will move forward. The input is the operation instruction data sent from the server, and the output is the corresponding machine operation.

[1400] This enables users to control factory machinery in real time using sign language, enabling smooth communication between hearing-impaired people and factory machinery.

[1401] 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.

[1402] The system of the present invention is a technology that combines a terminal such as a smartphone or PC, a server, and an emotion engine that recognizes the user's emotions. The specific implementation method and operation flow are as follows.

[1403] User operations

[1404] When a user communicates using sign language, they first sign in front of the device's camera. The camera captures this sign language movement. The device is equipped with an emotion engine that recognizes emotions from the user's facial expressions and movements. This video data is analyzed in real time and sent to a server.

[1405] Device behavior

[1406] The device captures video of the user's sign language and sends the video data to the server. It is important that the device application has a high-performance camera and network connection to maintain video quality. In addition, an emotion engine analyzes the user's facial expressions and movements to extract emotional data. This emotional data is also sent to the server along with the video data.

[1407] Server Operation

[1408] The server processes the video and emotion data received from the device. The video data is input into an AI model (e.g., a neural network) that analyzes the sign language movements and converts the corresponding sign language words or phrases into text data. Regarding the emotion data, the emotion engine reflects the analyzed information in the text data and generates text that expresses the emotion.

[1409] The generated text data is converted into text using a natural language processing (NLP) algorithm. The server then uses a multilingual translation function to translate the generated text data into other languages. This translation is performed using the Google Translate API or a proprietary translation AI model. The translated text data is then input back into the sign language generation model to generate sign language text corresponding to the sign language of the other country.

[1410] Sending and displaying translation data

[1411] The server sends the generated translated sign language text to the terminal. The terminal receives the translated sign language text from the server and displays it to the user. The displayed translated text is provided in an appropriate format that is easy for the user to understand.

[1412] Specific examples

[1413] For example, consider the case where a deaf person in the United States communicates with a deaf person in Japan. The American user signs with emotion, and the video and emotional data are captured by the device's camera and sent to the server. The server analyzes the video data and converts it into English text data, and the emotion engine reflects the analyzed emotional information in the text. The English text data is then translated into Japanese text that corresponds to Japanese Sign Language. The translated text data is sent to the device of the deaf person in Japan and displayed appropriately. This allows the Japanese user to understand the content of the American sign language and its emotions, achieving smooth communication.

[1414] The system of the present invention enables smooth international communication between hearing-impaired people and between hearing-impaired people and people without hearing, while reflecting the user's emotions.

[1415] The processing flow will be explained below.

[1416] Step 1:

[1417] Device behavior

[1418] The user signs into the device's camera. The device's built-in camera captures the sign language video data, and the emotion engine simultaneously analyzes the user's facial expressions and movements to obtain emotional data. The video data and emotional data are temporarily stored in memory.

[1419] Step 2:

[1420] Device behavior

[1421] The device transmits the captured video data and emotion data to the server in real time using communication protocols such as WebSocket and HTTP / 2, and appropriate compression techniques are used to minimize delays during data transmission.

[1422] Step 3:

[1423] Server Operation

[1424] The server receives the video and emotion data sent from the device and stores it in temporary storage. The received data is immediately passed to the processing queue, where analysis begins.

[1425] Step 4:

[1426] Server Operation

[1427] The video data is fed into an AI model that analyzes the sign language movements. The AI ​​model used here is typically a neural network (e.g., CNN or RNN) that recognizes each sign movement and identifies the corresponding sign word or phrase.

[1428] Step 5:

[1429] Server Operation

[1430] After identifying the sign language actions, they are converted into corresponding text data, which is then fed into an NLP algorithm to be reconstructed into natural-sounding sentences.

[1431] Step 6:

[1432] Server Operation

[1433] Emotional data is analyzed to identify the user's emotions. NLP algorithms are used to reflect the emotional data in the text, so that the generated text contains the user's emotional information.

[1434] Step 7:

[1435] Server Operation

[1436] The text data is input into a multilingual translation model and translated into a specified target language. The resulting text data is then input into a sign language generation model to generate sign language text in the corresponding target language.

[1437] Step 8:

[1438] Server Operation

[1439] The translated sign language text data is sent to the device, again using a communication protocol (WebSocket or HTTP / 2) to transfer the data in real time.

[1440] Step 9:

[1441] Device behavior

[1442] The terminal receives the translated sign language text data sent from the server and displays it to the user in an appropriate format, which also reflects the original user's emotions.

[1443] Step 10:

[1444] User Actions

[1445] The received translated sign language text data is checked, and a reply is made in sign language if necessary. The reply sign language is also transmitted to the other party through the same steps.

[1446] Specific examples

[1447] For example, consider the case where a hearing-impaired person in the United States (User A) communicates with a hearing-impaired person in Japan (User B). User A uses sign language, and the emotion engine analyzes his or her emotions. This video and emotion data is sent from the device to the server. The server analyzes the data, converts the sign language movements into English text data, and also reflects the emotion data. Next, the English text is translated into Japanese text and Japanese Sign Language. The translation result is sent to User B's device and displayed. This allows User B to understand the content of User A's sign language, including his or her emotions, achieving smooth communication.

[1448] This system enables communication through sign language that reflects the user's emotions, realizing smooth dialogue between hearing-impaired people in different countries and between hearing-impaired people and hearing-speaking people.

[1449] Example 2

[1450] 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."

[1451] There is a lack of efficient means to accurately convey sign language and emotions in communication between hearing-impaired and hearing-disabled people. Conventional sign language translation systems have difficulty conveying rich communication, including emotions, and are inadequate for translating sign language between multiple languages. This makes it difficult for hearing-impaired people to communicate smoothly with users of different cultures and languages.

[1452] 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.

[1453] In this invention, the server includes means for acquiring video data for hearing-impaired people to communicate using sign language, means for analyzing the acquired sign language video data and user emotion data and converting them into corresponding text data, means for analyzing the converted text data including the emotion data and translating it into sign language text for another country, and means for transmitting the translated sign language text. This enables multilingual translation of sign language including emotion, enabling smooth communication between hearing-impaired people and those without hearing.

[1454] "Hearing impaired" refers to a person who has difficulty communicating verbally and uses sign language or other visual methods to communicate.

[1455] "Sign language" is a method of expressing language using visual gestures, hand shapes, and facial expressions, and refers to a means of communication primarily used by people with hearing impairments.

[1456] "Video data" refers to digital data that includes visual information such as a user's sign language actions and facial expressions.

[1457] "Emotion data" refers to information about emotions analyzed from the user's facial expressions and movements.

[1458] "Analysis" refers to the process of analyzing collected data to extract useful information.

[1459] "Text data" refers to digital data that expresses linguistic information as characters.

[1460] "Translation" refers to the process of converting content expressed in one language into another language.

[1461] A "neural network" is an algorithm that mimics the neural circuits of living organisms, and refers specifically to a machine learning model used for image and sound recognition.

[1462] An "emotion engine" refers to a software or hardware system that automatically analyzes emotions from a user's facial expressions and voice.

[1463] "Natural language processing algorithms" refer to computational methods for analyzing and generating language data, and are particularly used for text translation and sentence generation.

[1464] A "sign language generation model" refers to a machine learning model or algorithm for converting text data into sign language.

[1465] A "server" refers to a computer system that provides services or functions to other computers over a network.

[1466] A "terminal" refers to a device that a user directly operates to input and display data, such as a smartphone or PC.

[1467] The system for implementing this invention operates in cooperation with users, terminals, and servers. Its purpose is to acquire video data for hearing-impaired people to communicate using sign language, analyze that data, convert it into corresponding text data, translate it into multiple languages, and then transmit the translated sign language text.

[1468] Hardware and software configuration

[1469] Device: A device that is directly operated by the user, such as a smartphone or PC. The device is equipped with a high-performance camera that captures the user's sign language movements and facial expressions.

[1470] Camera: A camera built into the device, designed to capture high-resolution, real-time video.

[1471] Emotion engine: Installed on the device, it uses technologies such as OpenCV and Emotion AI SDK to recognize emotions from the user's facial expressions and movements.

[1472] Server: A computer system that performs key processes such as data analysis, sign language to text conversion, multilingual translation, etc. The server analyzes the data using neural networks (e.g., models using TensorFlow or PyTorch) or natural language processing algorithms (e.g., spaCy or BERT models).

[1473] System Operation

[1474] This system operates in the following manner.

[1475] 1. User operations

[1476] The user expresses their feelings by signing, for example, "Hello, how are you?"

[1477] 2. Video and Emotion Data Capture

[1478] The device's camera captures the user's sign language and facial expressions, and the emotion engine analyzes the user's emotions from their facial expressions, generating video data of the sign language movements and emotional data.

[1479] 3. Sending data to the server

[1480] The generated video data and emotion data are transmitted to a server in real time using a secure communication protocol (e.g., HTTPS).

[1481] 4. Analysis of video data

[1482] The server inputs the received video data into a neural network, analyzes the sign language movements, and converts them into sign language words and phrases.

[1483] 5. Emotion Data Analysis

[1484] The server reflects emotional information in the text data generated from the sign language based on the emotional data from the emotion engine. For example, it adds the annotation "with a smile" to "Hello, how are you?"

[1485] 6. Multilingual Translation of Text Data

[1486] The server translates the generated text data using the Google Translate API or a proprietary translation AI model. For example, English text is translated into Japanese.

[1487] 7. Sign Language Text Generation

[1488] The translated text data is input into a sign language generation model to generate corresponding sign language text, for example, translated Japanese sign language text.

[1489] 8. Sending and displaying translation data to your device

[1490] The server sends the translated sign language text to the terminal. The terminal displays the received sign language text to the user in an easy-to-understand format. For example, the Japanese Sign Language text is displayed on the terminal of a Japanese user.

[1491] Specific examples

[1492] For example, if a deaf person in the United States wants to sign to a deaf person in Japan, "Hello, how are you?", they can use the following prompt:

[1493] Example prompt: "A deaf person in the United States signs 'Hello, how are you?' with emotion. Please translate this sign into Japanese sign language. Please also reflect the emotion."

[1494] As described above, this system enables multilingual translation of sign language, including emotional expressions, enabling smooth communication between hearing-impaired and hearing-disabled people.

[1495] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1496] Step 1:

[1497] The user signs.

[1498] As a specific action, the user expresses his / her feelings using sign language. For example, the user signs "Hello, how are you?"

[1499] Step 2:

[1500] The device captures video data and emotional data.

[1501] Input and output: The device's camera captures the user's sign language and facial expressions (input), and generates emotional data (output) through the video data and emotion engine.

[1502] Specifically, a high-resolution camera captures sign language movements and facial expressions, and an emotion engine (e.g., OpenCV or Emotion AI SDK) analyzes emotions from the user's facial expressions.

[1503] Step 3:

[1504] The terminal transmits the video data and emotion data to the server.

[1505] Input and Output: The captured video data and emotion data (input) are sent to the server (output) using a secure communication protocol.

[1506] Specifically, the terminal transmits data in real time using a secure communication protocol such as HTTPS.

[1507] Step 4:

[1508] The server analyzes the video data.

[1509] Input and Output: Received video data (input) is fed into the neural network model and converted into sign language words or phrases (output).

[1510] Specifically, the server uses an AI model using TensorFlow and PyTorch to analyze sign language movements and extract the corresponding sign language words and phrases.

[1511] Step 5:

[1512] The server analyzes the emotion data.

[1513] Input and output: Analyze the received emotional data (input) and reflect the emotional information (output) in the text generated from the sign language.

[1514] Specifically, the emotion engine adds the analyzed emotion information to the sign language text. For example, if the sign language is "Hello, how are you?", the emotion information "with a smile" is added.

[1515] Step 6:

[1516] The server translates the text data into multiple languages.

[1517] Input and output: The generated text data (input) is input into the Google Translate API or a proprietary translation AI model to obtain translated text data (output).

[1518] Specifically, the server uses natural language processing (NLP) technology to translate text, for example from English to Japanese.

[1519] Step 7:

[1520] The server inputs the translated text data into a sign language generation model.

[1521] Input and output: The translated text data (input) is input into the sign language generation model to generate sign language text in other countries (output).

[1522] Specifically, the translated text data is passed through a sign language generation model to generate Japanese Sign Language or other sign language text.

[1523] Step 8:

[1524] The server transmits the generated translated sign language text to the terminal.

[1525] Input and output: The generated translated sign language text (input) is sent to the terminal, which displays it to the user (output).

[1526] Specifically, the server sends the generated sign language text to the terminal, and the terminal displays it to the user in an appropriate format. For example, Japanese Sign Language text is displayed on the terminal of a Japanese user.

[1527] (Application example 2)

[1528] 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."

[1529] When hearing-impaired people shop or use services at brick-and-mortar stores, they face the challenge of having difficulty communicating smoothly using sign language. Furthermore, because store clerks cannot understand sign language, they are unable to provide sufficient product explanations or guidance, which often causes inconvenience to hearing-impaired people. Furthermore, they are required to be able to understand and respond to the emotional content of sign language.

[1530] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1531] In this invention, the server includes means for acquiring video data for the hearing impaired to communicate using sign language, means for analyzing the acquired sign language video data and converting it into corresponding text data, means for recognizing the user's emotions, generating emotion data, and reflecting it in text, means for translating the converted text data into sign language text of another country, means for transmitting the translated sign language text, and means for displaying the acquired text data and sign language text, thereby enabling sign language communication that reflects emotions between the hearing impaired and store clerks.

[1532] "Means for acquiring video data" refers to a mechanism for capturing video including sign language movements using devices such as cameras and sensors.

[1533] "Means for analyzing sign language video data" refers to algorithms or software that process the acquired sign language video, recognize sign language actions, and convert them into corresponding text data.

[1534] The "means for converting into corresponding text data" is a processing method for converting sign language expressions into natural language text based on the analyzed sign language video data.

[1535] "Means for translating into sign language text of other countries" refers to a translation algorithm or API for converting the converted text data into sign language text in multiple languages.

[1536] "Means for transmitting sign language text" refers to a communication means for transmitting the translated sign language text to another terminal or server via a network.

[1537] The "means for recognizing user emotions" refers to algorithms or software for identifying a user's emotions and generating emotion data through facial expression recognition and motion analysis.

[1538] The "means for generating emotion data" is a process for generating data for expressing emotions based on the recognized emotion information.

[1539] The "means for reflecting in text" is a processing method for integrating the generated emotion data into text data and generating a text expression that includes emotion.

[1540] The "means for displaying text data and sign language text" is a mechanism for displaying the acquired text data and sign language text on the screen or display of the terminal.

[1541] To implement this invention, the following hardware and software environment is used. The main hardware used is a smartphone, server, camera, and display. The software used is a neural network framework (TensorFlow or PyTorch), a natural language processing (NLP) engine (spaCy, Hugging Face Transformers), an emotion recognition engine (EmotionRecognitionAPI), and a translation API (Google Translate API).

[1542] First, the user signs using the smartphone's camera. The camera captures the sign language movements, and the device's emotion engine analyzes the video data in real time. This analysis recognizes emotions from the user's facial expressions and movements, and generates emotion data. The video data and emotion data are then sent to the server.

[1543] The server analyzes the received video data using a neural network and converts the sign language movements into text data. It also reflects the corresponding emotions based on the emotional data. The generated text data is then compiled into sentences using a natural language processing (NLP) algorithm.

[1544] Next, the server translates the generated text data into other languages ​​using a multilingual translation API. This translated text data is input into a sign language generation model to generate sign language text for other languages. The translated sign language text is then sent from the server to the device.

[1545] When the terminal receives the translated sign language text from the server, it displays it to the user. The displayed text is presented in an appropriate emotional format that is easy for the user to understand. In addition, it plays animations of the sign language, allowing the user to visually understand the sign language.

[1546] Examples of specific prompts include:

[1547] "Analyze the sign language video captured by this camera and generate corresponding text and emotional information. Translate the generated text data into Japanese and convert it into Japanese Sign Language."

[1548] This invention allows hearing-impaired people to easily communicate with store staff when using services in physical stores. Because it does not use voice, it can be used effectively even in quiet environments. Furthermore, by displaying sign language that reflects emotions, it is possible to accurately convey the user's intentions and emotions.

[1549] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1550] Step 1:

[1551] The user signs into the smartphone camera.

[1552] Input: Sign language action

[1553] Output: Sign language video data

[1554] The device's camera captures the user's sign language movements and obtains high-quality video data, which is important for subsequent analysis.

[1555] Step 2:

[1556] The sign language video data acquired by the device is analyzed using an emotion engine to generate emotion data.

[1557] Input: Sign language video data

[1558] Output: Emotion data

[1559] The emotion engine built into the device processes the video data, recognizes emotions from the user's facial expressions and movements, and generates emotion data, which is then used in the translation process.

[1560] Step 3:

[1561] The device transmits sign language video data and emotion data to the server.

[1562] Input: Sign language video data, emotion data

[1563] Output: Data packet to be sent

[1564] Using a high-performance network connection, the device transmits sign language video data and emotion data in real time to a server, where the data packets are analyzed.

[1565] Step 4:

[1566] The server analyzes the sign language video data using a neural network and converts it into corresponding text data.

[1567] Input: Sign language video data

[1568] Output: Corresponding text data

[1569] The server's neural network analyzes sign language movements, identifies corresponding words and phrases, and converts them into text data, using TensorFlow and PyTorch for the process.

[1570] Step 5:

[1571] The server reflects the emotion data in the generated text data.

[1572] Input: Corresponding text data, emotion data

[1573] Output: Text data containing emotions

[1574] The server can incorporate emotional information into text data based on emotion recognition data, thereby conveying the user's intentions more accurately.

[1575] Step 6:

[1576] The server translates the generated text data into other languages ​​using natural language processing algorithms.

[1577] Input: Text data containing emotions

[1578] Output: Translated text data

[1579] The server uses a natural language processing engine (e.g., spaCy or Hugging Face Transformers) to translate the text data into other languages.

[1580] Step 7:

[1581] The server inputs the translated text data into a sign language generation model to generate sign language text for other countries.

[1582] Input: Translated text data

[1583] Output: Foreign sign language text

[1584] The server uses an AI model to generate sign language text corresponding to the sign language of other countries based on the translation data.

[1585] Step 8:

[1586] The server transmits the generated sign language text of the other country to the terminal.

[1587] Input: Foreign sign language text

[1588] Output: Data packet to be sent

[1589] The server transmits the generated sign language text data of the other country to the terminal via the network.

[1590] Step 9:

[1591] The terminal displays the sign language text received from the server to the user.

[1592] Input: Foreign sign language text

[1593] Output: Data for display

[1594] The terminal displays the received data on a display and plays visual sign language animations to help the user understand the content.

[1595] 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.

[1596] 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.

[1597] 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.

[1598] 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.

[1599] FIG. 9 is a diagram illustrating 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 actions 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.

[1600] 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.

[1601] 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).

[1602] 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.

[1603] 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."

[1604] 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.

[1605] 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).

[1606] 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.

[1607] 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.

[1608] 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.

[1609] 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.

[1610] 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.

[1611] 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.

[1612] 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.

[1613] 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.

[1614] 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.

[1615] 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.

[1616] The following is further disclosed regarding the above embodiment.

[1617] (Claim 1)

[1618] A means for acquiring video data for the hearing impaired to communicate using sign language;

[1619] means for analyzing the acquired sign language video data and converting it into corresponding text data;

[1620] A means for translating the converted text data into sign language text of another country;

[1621] means for transmitting the translated sign language text;

[1622] A system including:

[1623] (Claim 2)

[1624] 2. The system according to claim 1, wherein the means for analyzing the sign language video data uses a neural network to recognize sign language actions.

[1625] (Claim 3)

[1626] 2. The system according to claim 1, wherein the multilingual translation means translates text data using a natural language processing algorithm.

[1627] "Example 1"

[1628] (Claim 1)

[1629] A means for acquiring video data for the hearing impaired to communicate using sign language;

[1630] means for analyzing the acquired sign language video data in real time and converting it into corresponding text data;

[1631] A means for translating the converted text data into sign language text of another country;

[1632] means for transmitting and displaying the translated sign language text;

[1633] A system including:

[1634] (Claim 2)

[1635] 2. The system according to claim 1, wherein the means for analyzing the sign language video data uses a neural network to recognize sign language movements and extract features.

[1636] (Claim 3)

[1637] 2. The system according to claim 1, wherein the multilingual translation means translates text data using a natural language processing algorithm and utilizes a specialized translation model.

[1638] "Application Example 1"

[1639] (Claim 1)

[1640] A means for acquiring video data for the hearing impaired to communicate using sign language;

[1641] means for analyzing the acquired sign language video data and converting it into corresponding text data;

[1642] A means for translating the converted text data into sign language text of another country;

[1643] means for transmitting the translated sign language text;

[1644] a means for receiving the analyzed sign language data and operating a factory machine;

[1645] A system including:

[1646] (Claim 2)

[1647] 2. The system according to claim 1, wherein the means for analyzing the sign language video data uses a neural network to recognize sign language actions.

[1648] (Claim 3)

[1649] 2. The system according to claim 1, wherein the multilingual translation means translates text data using a natural language processing algorithm.

[1650] "Example 2: Combining Emotion Engines"

[1651] (Claim 1)

[1652] A means for acquiring video data for the hearing impaired to communicate using sign language;

[1653] means for analyzing the acquired sign language video data and user emotion data and converting them into corresponding text data;

[1654] means for analyzing the converted text data including emotion data and translating it into sign language text of another country;

[1655] means for transmitting the translated sign language text;

[1656] A system including:

[1657] (Claim 2)

[1658] 2. The system according to claim 1, wherein the means for analyzing the sign language video data uses a neural network to recognize sign language actions and an emotion engine to analyze the user's emotions.

[1659] (Claim 3)

[1660] 2. The system according to claim 1, wherein the multilingual translation means translates text data using a natural language processing algorithm, and inputs the translated text data into a sign language generation model to generate sign language text.

[1661] "Application example 2 when combining emotion engines"

[1662] (Claim 1)

[1663] A means for acquiring video data for the hearing impaired to communicate using sign language;

[1664] means for analyzing the acquired sign language video data and converting it into corresponding text data;

[1665] A means for translating the converted text data into sign language text of another country;

[1666] means for transmitting the translated sign language text;

[1667] A means for recognizing a user's emotion, generating emotion data, and reflecting the emotion data in text;

[1668] means for displaying the acquired text data and sign language text;

[1669] A system including:

[1670] (Claim 2)

[1671] 2. The system according to claim 1, wherein the means for analyzing the sign language video data uses a neural network to recognize sign language actions.

[1672] (Claim 3)

[1673] 2. The system according to claim 1, wherein the multilingual translation means translates text data using a natural language processing algorithm. [Explanation of symbols]

[1674] 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 acquiring video data for the hearing impaired to communicate using sign language; means for analyzing the acquired sign language video data and converting it into corresponding text data; A means for translating the converted text data into sign language text of another country; means for transmitting the translated sign language text; A system including:

2. 2. The system according to claim 1, wherein the means for analyzing the sign language video data uses a neural network to recognize the sign language gestures.

3. 2. The system according to claim 1, wherein the multilingual translation means translates text data using a natural language processing algorithm.

Images