system

Abstract

We provide the system. [Solution] A means of acquiring visual data using communication technology with video acquisition equipment, A means for processing the received visual data using a generative computer program and outputting emotion data and action data, A means for transmitting the processing result to the user as audio using an audio output device, A means of using voice to enable a robot to assist a visually impaired user in interacting with others within the home, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a need to provide an effective means of information transmission in a remote environment for users who have difficulty communicating relying on visual information. However, in the existing technologies, there is no established means to intuitively understand expressions and gestures from the videos obtained in an online communication environment and provide feedback by voice, so there is a problem that it is difficult for users to communicate smoothly.

Means for Solving the Problems

[0005] This invention provides a system that embodies emotional and gesture information by acquiring image data in real time via online communication technology using an image acquisition device and analyzing that data using a generative computer model. By providing the analysis results to the user in real time as audio feedback using an audio output device, smooth communication is possible even in situations where visual information is insufficient.

[0006] A "video acquisition device" is a hardware device that uses online communication technology to acquire image data in real time.

[0007] "Online communication technology" refers to all technologies used to send and receive data via the internet or communication networks.

[0008] A "generative computing model" refers to an algorithm or model that extracts features from input data based on machine learning and data analysis, and generates a specific output.

[0009] "Emotional information" refers to information about human emotions and psychological states that is analyzed from image data.

[0010] "Gesture information" refers to information about body movements and gestures that is analyzed from image data.

[0011] A "voice output device" refers to a hardware device that provides analyzed information to the user as audio.

[0012] "User" refers to anyone who uses this system to obtain visual information as audio feedback. [Brief explanation of the drawing]

[0013] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2]It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] [[ID=2^4]]It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

MODE FOR CARRYING OUT THE INVENTION

[0014] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a processor with a reference numeral (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0017] In the following embodiments, a RAM (Random Access Memory) with a reference numeral is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0018] In the following embodiments, a storage with a reference numeral is one or more non-volatile storage devices that store various programs, various parameters, and the like. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0019] In the following embodiments, a communication I / F (Interface) with a reference numeral is an interface including a communication processor, an antenna, and the like. 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), or Bluetooth (registered trademark).

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0021] [First Embodiment]

[0022] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0023] As shown in Figure 1, the 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.

[0024] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0026] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0027] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0029] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0033] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0034] In this embodiment of the invention, a video acquisition device, a generative computer model, and an audio output device work together to construct a system that effectively converts visual information into audio information and provides it.

[0035] First, the user initiates a meeting or discussion using online communication technology. The device captures the video of the meeting and acquires image data. This image data consists of frame-by-frame data, including the facial expressions and gestures of the participants, and is sent to the server.

[0036] The server inputs the received image data into a generative computer model. This model has been pre-trained on a large amount of data on facial expressions and gestures, and analyzes human emotional and gesture information with high accuracy. Through this analysis, emotional labels such as "happy" or "confused" and gesture labels such as "nodding" are obtained.

[0037] Next, the server converts these analysis results into natural language and generates audio data using speech synthesis technology. This audio data is structured to be easily understood by the user.

[0038] The device receives the generated audio data and plays it back in real time through the user's audio output device, thereby transmitting visual information in the form of text and audio. This makes it easier for the user to grasp the emotions and intentions conveyed by the visual information, enabling smoother communication.

[0039] As a concrete example, consider a scenario where a user is a presenter during a meeting, and one of the participants shows a confused expression. A video capture device captures this expression in real time, and a server analyzes it as "confused." Based on this result, the user receives audio feedback saying, "The other person is confused, please consider providing additional explanations." This feedback allows the user to immediately adjust their response.

[0040] This system facilitates the understanding of nonverbal communication even when direct visual information is unavailable, supporting smoother conversations, especially for users with visual impairments.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The user initiates a meeting via online communication technology. This sets up the device to activate its camera and begin capturing video.

[0044] Step 2:

[0045] The device acquires video data from the camera in real time. The acquired video data is temporarily stored in memory as individual frames. The video data is also converted to an appropriate format (e.g., JPEG, PNG).

[0046] Step 3:

[0047] The device sends the captured image data to the server. This transmission is performed via a protocol that enables efficient and low-latency communication (e.g., WebSocket).

[0048] Step 4:

[0049] The server inputs the received image data into a generative computer model. The model analyzes the image frames, classifies facial expressions and gestures, and labels them as emotional and gesture information.

[0050] Step 5:

[0051] The server converts the analyzed emotion and gesture information into natural language text. Here, the information is expressed in language that is intuitively easy for the user to understand.

[0052] Step 6:

[0053] The server passes natural language text to a speech synthesis engine, which generates speech data. The speech data is then adjusted to facilitate user comprehension.

[0054] Step 7:

[0055] The server sends the generated audio data back to the terminal. To minimize communication delays during this process, an appropriate protocol is used.

[0056] Step 8:

[0057] The device plays back audio data through an audio output device. This allows the user to receive analyzed facial and gesture information as audio feedback.

[0058] Step 9:

[0059] The conversation flow is adjusted based on the user's voice feedback. For example, if the emotion label is "confused," additional explanations or follow-ups are considered.

[0060] (Example 1)

[0061] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0062] There is a need to effectively convey nonverbal communication to users who cannot directly access visual information. Furthermore, in online meetings and discussions, it is difficult to grasp participants' emotions and intentions in real time, posing a barrier to smooth communication. Conventional technologies have been unable to adequately support user understanding by relying solely on audio information to supplement this visual information.

[0063] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0064] In this invention, the server includes means for acquiring video information using digital information transmission technology with an image acquisition device, means for analyzing the received video information using a generative data processing model and outputting emotion labels and action labels, and means for providing the analysis results to the user as audio information using a voice generation device. This enables the user to grasp the emotions and intentions of participants in real time during online meetings and conversations, and to communicate smoothly.

[0065] An "image acquisition device" is a device used to acquire visual information using digital information transmission technology.

[0066] A "generative data processing model" is a machine learning-based model used to analyze received video information and output emotion labels and action labels.

[0067] A "voice generation device" is a device that converts analysis results into voice information and provides it to the user.

[0068] "Visual information" refers to visual data expressed in digital format, including images and video data acquired during online meetings and discussions.

[0069] An "emotional label" is an identifier extracted from a person's facial expressions and gestures that indicates a specific emotional state.

[0070] A "movement label" is an identifier extracted to represent human movement or gesture.

[0071] "Audio information" refers to data that expresses text data or analysis results as audio.

[0072] This invention constructs a system for converting visual information into audio information, and is primarily implemented using an image acquisition device, a generative data processing model, and a speech generation device. These elements work together to provide visual information as audio in real time, thereby supporting users who cannot directly acquire visual information.

[0073] Users initiate a meeting using online communication technology. The terminal captures video of the meeting using connected image acquisition devices such as cameras. The video information includes facial expressions and gestures of the participants frame by frame, and this is sent to the server.

[0074] The server inputs the received video information into a generative data processing model. This model is based on a pre-trained dataset and extracts emotion and behavior labels with high accuracy. Specifically, the model utilizes machine learning and deep learning technologies.

[0075] The obtained emotion and action labels are converted into natural language by a speech generator on the server. This converts the emotion and action labels into voice messages that are understandable to the user.

[0076] The device receives the audio data and plays it back in real time through an audio output device. Visual information is presented to the user as audio information via speakers or headphones.

[0077] As a concrete example, consider a situation in a meeting where the presenter cannot visually confirm the confused expressions of the participants. The system extracts the confused label and delivers an audio message to the presenter saying, "The participant is confused; please consider providing additional explanations." In this way, the presenter can immediately adjust their response based on the participants' reactions.

[0078] An example of a prompt might be: "Describe a system that analyzes the facial expressions of meeting participants and provides audio feedback on emotions such as confusion and joy." This would provide guidance for understanding and implementing the system.

[0079] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0080] Step 1:

[0081] The user initiates a meeting using online communication technology. This prepares the system for capturing real-time video of meeting participants. The device uses its connected camera to acquire video information in real time. It receives the meeting video feed as input and generates frame-by-frame image data as output.

[0082] Step 2:

[0083] The terminal sends the image data for each frame it acquires to the server. The frame data includes the person's facial expressions and movements. Specifically, the terminal packets the image data over the network and sends it to the server. The input here is the image data for each frame, and the output is the image packets that arrive at the server.

[0084] Step 3:

[0085] The server inputs the received image data into a generative data processing model. This model has been pre-trained on various facial expressions and actions, and analyzes the image data to output emotion labels and action labels. Specifically, the model processes the pixel information of each frame to extract features and generates labels based on them. The input is the image data received by the server, and the output is emotion labels and action labels.

[0086] Step 4:

[0087] The server performs a process of converting the generated emotion and behavior labels into natural language text. Next, it uses speech synthesis technology to convert the text into speech data. Specifically, it converts emotion labels into text and applies a speech synthesis algorithm to generate a speech file. The input is emotion and behavior labels, and the output is natural language speech data.

[0088] Step 5:

[0089] The terminal receives audio data from the server and provides real-time feedback to the user through an audio playback device. Specifically, the terminal decodes the audio data and plays it back through a speaker or headphones. The input is audio data from the server, and the output is the real-time audio message heard by the user.

[0090] (Application Example 1)

[0091] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0092] People with visual impairments face challenges in communicating smoothly with others within their families. This problem stems from their difficulty in understanding nonverbal information such as facial expressions and gestures. Therefore, it is necessary to reduce communication barriers by converting visual information into auditory information and effectively communicating it to people with visual impairments.

[0093] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0094] In this invention, the server includes means for acquiring visual data using communication technology, means for processing the received visual data with a generative computer program and outputting emotion data and motion data, and means for transmitting the processing results to the user as audio using an audio output device. This makes it easier for visually impaired users to understand the emotions and gestures of others, and enables smooth communication within the home.

[0095] A "video acquisition device" is a device that acquires visual data using communication technology.

[0096] "Communication technology" refers to technologies for exchanging data over long distances.

[0097] "Visual data" refers to a collection of information that is represented as images or videos.

[0098] A "generative computer program" is a computer program that processes visual data and generates emotional and behavioral data.

[0099] "Emotional data" refers to information about human emotions that is analyzed from visual data.

[0100] "Motion data" refers to information about human movement that is analyzed from visual data.

[0101] An "audio output device" is a device used to transmit audio to a user.

[0102] A "user" is a person who receives support through this system.

[0103] "Visual impairment" refers to a condition in which a person has difficulty recognizing visual information.

[0104] "Within the home" refers to the living space centered around the residence.

[0105] This invention is a communication support system for visually impaired users that utilizes communication technology. This system operates in conjunction with a video acquisition device, a generative computer program, and an audio output device.

[0106] The server receives visual data acquired in real time by video acquisition equipment. This visual data includes the speaker's facial expressions and gestures. This data is sent to a generative computer program on the server and analyzed into emotion data and motion data. The generative computer uses a pre-trained dataset, enabling rapid and highly accurate facial expression recognition and gesture classification.

[0107] The server converts the generated emotion and behavior data into natural language and then uses speech synthesis software to create audio data. This audio data is then provided to visually impaired users within their homes via audio output devices.

[0108] As a concrete example, when interacting with family members, the robot's camera captures the other person's visual information and conveys the change in their facial expression to the user as an audio message such as, "The other person is smiling." This compensates for the lack of visual information and facilitates smoother communication.

[0109] An example of a prompt for facial expression analysis using a generative AI model is as follows: "Analyze the facial expression of the person captured by the camera and describe their emotion in voice."

[0110] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0111] Step 1:

[0112] The terminal captures visual data in real time using video acquisition equipment. This data includes the facial expressions and gestures of the person being spoken to. The input is video data from the camera, and the output is visual data sent to the server. The terminal sends this data to the server for the next processing step.

[0113] Step 2:

[0114] The server inputs the received visual data into a generative AI model. The generative AI model analyzes emotion data and behavior data using a pre-trained dataset. The input is visual data, and the output is the analyzed emotion data and behavior data. Based on this data processing, the server decomposes the visual information into emotion information and gesture information.

[0115] Step 3:

[0116] The server converts the analyzed emotion and behavior data into natural language. This conversion process prepares the information to be conveyed to the user into easily understandable language. The input is emotion and behavior data, and the output is the language expression as text data. The server then prepares this text for conversion into speech.

[0117] Step 4:

[0118] The server converts text data into speech data using speech synthesis technology. Specifically, it uses speech synthesis software to construct natural-sounding speech. The input is text data converted into natural language, and the output is the generated speech data. The server sends this speech data to the terminal.

[0119] Step 5:

[0120] The device transmits received audio data to the user using an audio output device. This allows the user to understand emotions and actions based on the other person's facial expressions and gestures through audio. The input is audio data, and the output is the audio the user hears. The device plays the audio at the appropriate time to support communication.

[0121] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0122] In embodiments of the present invention, a video acquisition device, a generative computer model, an audio output device, and an emotion engine are combined to construct a system that analyzes the user's nonverbal information in real time and provides it to the user through audio feedback.

[0123] First, users initiate a meeting or conversation using online communication technology. The device uses its camera to capture video of this conversation in real time, acquiring it as image data. This image data is then sent to a server for detailed analysis.

[0124] On the server, a generative computing model analyzes the received image data. This model is pre-trained on a large amount of facial expression and gesture data, and can extract emotional and gesture information from the data. Next, an emotion engine refines the results of the generative computing model, generating more detailed and accurate emotional information. This engine tracks and analyzes multiple emotional states in real time, providing personalized and optimized feedback.

[0125] The server then converts the analyzed emotion and gesture information into natural language text and generates easily understandable audio data through a speech synthesis engine. This audio data is sent to the terminal and provided to the user in real time through an audio output device.

[0126] As a concrete example, imagine a user presenting in a meeting who observes that one of the participants is showing a confused expression and restless gestures. The server's emotion engine then analyzes this information with high accuracy and generates and provides the user with audio feedback stating, "The participant appears confused and may need further explanation." This allows the user to immediately adjust their response and be considerate of the participant.

[0127] This system plays a role in facilitating the understanding of visual and nonverbal information, especially in situations where visual information is difficult to obtain, and supporting smooth and effective communication with users.

[0128] The following describes the processing flow.

[0129] Step 1:

[0130] The user initiates a meeting via online communication software. This enables the device to activate its camera and prepare for video capture.

[0131] Step 2:

[0132] The device continuously captures video of the conversation using its camera, collecting image data in real time. This data is then converted to an appropriate format and temporarily stored.

[0133] Step 3:

[0134] The device processes the collected image data and sends it to the server for analysis. The transmission process is optimized to maintain stable communication.

[0135] Step 4:

[0136] The server analyzes image data using a generative computing model. Based on a pre-trained dataset, the model detects facial expressions and gestures and extracts emotional and gesture information.

[0137] Step 5:

[0138] The server applies an emotion engine to the extracted emotion information. This engine performs a more detailed emotion analysis, evaluates multiple emotional states in real time, and tracks dynamic changes in emotions.

[0139] Step 6:

[0140] The server converts the detailed information obtained by the emotion engine into natural language and generates audio data using a speech synthesis engine. This audio data is then formatted to be easily understood by the user.

[0141] Step 7:

[0142] The server generates audio data, which is then sent to the terminal and played back to the user in real time via an audio output device.

[0143] Step 8:

[0144] Users receive voice feedback, which helps to adjust the meeting's progress and the flow of conversation. For example, if a participant seems confused, additional explanations can be provided to improve communication.

[0145] (Example 2)

[0146] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0147] In online meetings and discussions, accurately understanding participants' emotions and actions, and facilitating smooth and effective communication, presents a challenge. In particular, analyzing nonverbal information is complex, and there is a need for technology that provides real-time feedback to users.

[0148] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0149] In this invention, the server includes means for acquiring image data in real time using online communication technology, means for analyzing emotional information and behavioral information from the image data using generative computing technology, and means for generating natural language text based on the analyzed emotional information and behavioral information. This makes it possible to efficiently and accurately analyze complex non-verbal information and provide real-time feedback to the user.

[0150] "Online communication technology" refers to technology that uses the internet or networks to send and receive information in real time.

[0151] "Image data" refers to data that represents visual information acquired using devices such as cameras in a digital format.

[0152] "Generative computing technology" refers to techniques that use artificial intelligence and machine learning models to generate and analyze new information from input data.

[0153] "Emotional information" refers to information that identifies and classifies an individual's emotional state, indicating their psychological state as inferred from facial expressions, body movements, and other factors.

[0154] "Motion information" refers to information obtained by analyzing a person's physical movements, such as gestures and body language.

[0155] "Natural language text" refers to text written in a language that humans use in everyday life, and is a machine-generated, easily interpretable linguistic expression.

[0156] "Speech generation technology" is a technology that converts text data into speech and synthesizes it to sound like a human voice.

[0157] "Providing in real time" means delivering information and data to users instantly and without delay.

[0158] This invention provides an effective communication support system using online communication. Specific embodiments for carrying out the invention are described below.

[0159] Users utilize this system, for example, in online meetings or conversations. The terminal is equipped with a camera, allowing for real-time capture of the user's facial expressions and gestures. This video data is then transmitted from the terminal to the server.

[0160] The server analyzes the received video data using a generative AI model. This generative AI model is pre-trained on a large dataset collected in the past and is designed to extract emotional and motion information by performing facial expression classification and motion analysis.

[0161] Next, the server uses this information to generate natural language text. This text is converted into audio data via a speech output device and sent to the terminal. This allows the user to receive audio feedback in real time.

[0162] For example, if a participant shows a confused expression while a user is giving a presentation, the generative AI model analyzes that emotion and provides the user with audio feedback such as, "It appears the participant has some questions about the explanation." In this way, the user can immediately adjust the content of the presentation and achieve smoother communication.

[0163] An example of a prompt message is, "Show the user how to get real-time emotional feedback," which can elicit a response from the system.

[0164] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0165] Step 1:

[0166] The user starts an online meeting, and the device acquires video data in real time via its camera. The input here is the video of the user and participants during the meeting, and the output is the acquisition of video data as frames. Specifically, the camera is activated and consecutive frames are captured.

[0167] Step 2:

[0168] The terminal compresses the acquired video data and sends it to the server using a transmission protocol. The input is frame data captured in real time, and the output is compressed video data. Specifically, a data compression algorithm is applied to efficiently transfer the data to the server.

[0169] Step 3:

[0170] The server receives the transmitted video data and analyzes it using a generative AI model. The input for analysis is compressed video data, and the output is extracted emotion and motion information. Specifically, the AI ​​model evaluates each frame and identifies facial expressions and movements.

[0171] Step 4:

[0172] The server uses an emotion engine to generate detailed natural language text based on the results of the generative AI model. The input consists of emotion and behavior information, and the output is a natural language text representation. Specifically, a text generation algorithm tailored to the emotional state is executed.

[0173] Step 5:

[0174] The server converts the generated natural language text into audio data using speech generation technology and sends it to the terminal. The input is the generated natural language text, and the output is the synthesized audio data. The specific operation includes generating an audio file using a text-to-speech engine.

[0175] Step 6:

[0176] The terminal provides the user with the received audio data through an audio output device. The input is synthesized audio data, and the output is real-time audio feedback. Specifically, the audio file is decoded and played back through the speaker.

[0177] (Application Example 2)

[0178] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0179] In modern elderly care settings, understanding the emotions and psychological state of residents is crucial. However, it is difficult for busy care staff to respond immediately to the emotional changes of each resident. Furthermore, nonverbal information is easily overlooked using traditional methods, requiring greater ingenuity to improve resident satisfaction and a sense of security. Therefore, there is a need to analyze and report residents' emotional states in real time, reducing the workload of care staff while providing meticulous care.

[0180] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0181] In this invention, the server includes means for acquiring image data using a video acquisition device, means for analyzing the image data using a generative computer model to obtain emotional information, and means for converting emotional feedback into speech using a speech synthesis device and providing it to the user. This enables the acquisition of nonverbal emotional information of residents in real time in care settings, allowing care staff to respond quickly and appropriately to changes in their emotions.

[0182] A "video acquisition device" is a device that acquires image data in real time using online communication technology.

[0183] A "generative computer model" is a computer model that analyzes received image data based on a pre-trained dataset and extracts and outputs emotional information and gesture information.

[0184] An "emotion engine" is an engine that refines the analysis results of generative computer models and generates personalized emotional feedback.

[0185] A "speech synthesis device" is a device that converts analyzed emotional feedback into speech data and provides it to the user in an easily understandable format.

[0186] A "care support device" is a device that includes a system that provides voice data from a speech synthesis device to care staff and reports the emotional state of residents in real time.

[0187] "Emotional feedback" is feedback information that uses non-verbal information acquired to convey the emotional state of residents to care staff and others via voice.

[0188] The system that realizes this application example includes a video acquisition device, a generative computer model, an emotion engine, a speech synthesizer, and a care support device. The server uses real-time image data obtained via the video acquisition device and utilizes the generative computer model to analyze the data and extract nonverbal emotion and gesture information. Because this analysis uses a pre-trained dataset, highly accurate emotion classification is possible.

[0189] The server further refines the results of the generative computer model using an emotion engine to generate personalized, detailed emotional feedback, which is then converted into speech data by a speech synthesizer. This speech data is sent to a care support device and provided to care staff in real time.

[0190] As a concrete example, if a resident suddenly shows signs of restlessness, the server's emotion engine analyzes the information and delivers an audio feedback message to the care staff via the care support device stating, "The resident may be feeling anxious." This allows the staff to respond quickly and reassure the resident.

[0191] An example of a prompt message is, "Design an app that analyzes the facial expressions and movements of residents and provides voice feedback on their emotions."

[0192] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0193] Step 1:

[0194] The server receives image data acquired in real time from the video acquisition device via online communication technology. This input image data includes non-verbal information, such as the residents' facial expressions and gestures. The server then transmits this data directly to a generative computer model.

[0195] Step 2:

[0196] The server analyzes the received image data using a generative computing model. This process identifies facial expressions and other nonverbal cues based on the input image data, and extracts emotional and gesture information. The output is data indicating the analyzed emotional state.

[0197] Step 3:

[0198] The server uses an emotion engine to refine the results of the generative computer model's analysis. In this step, it generates more detailed and personalized emotion feedback based on the analyzed data. The input is the results of the generative computer model's analysis, and the output is the refined emotion feedback data.

[0199] Step 4:

[0200] The server uses a speech synthesizer to convert emotional feedback into audio data. This process synthesizes the input into an easily understandable audio format and converts it in real time. The output becomes audio data usable by care support devices.

[0201] Step 5:

[0202] The terminal receives audio data from the server and provides real-time audio feedback to care staff through the care support device. In this step, information about the resident's emotions is immediately transmitted to the care staff, and information is output to enable them to take appropriate action.

[0203] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0204] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0205] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0206] [Second Embodiment]

[0207] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0208] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0209] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0210] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0211] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0212] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0213] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0214] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0215] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0216] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0217] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0218] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0219] In this embodiment of the invention, a video acquisition device, a generative computer model, and an audio output device work together to construct a system that effectively converts visual information into audio information and provides it.

[0220] First, the user initiates a meeting or discussion using online communication technology. The device captures the video of the meeting and acquires image data. This image data consists of frame-by-frame data, including the facial expressions and gestures of the participants, and is sent to the server.

[0221] The server inputs the received image data into a generative computer model. This model has been pre-trained on a large amount of data on facial expressions and gestures, and analyzes human emotional and gesture information with high accuracy. Through this analysis, emotional labels such as "happy" or "confused" and gesture labels such as "nodding" are obtained.

[0222] Next, the server converts these analysis results into natural language and generates audio data using speech synthesis technology. This audio data is structured to be easily understood by the user.

[0223] The device receives the generated audio data and plays it back in real time through the user's audio output device, thereby transmitting visual information in the form of text and audio. This makes it easier for the user to grasp the emotions and intentions conveyed by the visual information, enabling smoother communication.

[0224] As a concrete example, consider a scenario where a user is a presenter during a meeting, and one of the participants shows a confused expression. A video capture device captures this expression in real time, and a server analyzes it as "confused." Based on this result, the user receives audio feedback saying, "The other person is confused, please consider providing additional explanations." This feedback allows the user to immediately adjust their response.

[0225] This system facilitates the understanding of nonverbal communication even when direct visual information is unavailable, supporting smoother conversations, especially for users with visual impairments.

[0226] The following describes the processing flow.

[0227] Step 1:

[0228] The user initiates a meeting via online communication technology. This sets up the device to activate its camera and begin capturing video.

[0229] Step 2:

[0230] The device acquires video data from the camera in real time. The acquired video data is temporarily stored in memory as individual frames. The video data is also converted to an appropriate format (e.g., JPEG, PNG).

[0231] Step 3:

[0232] The device sends the captured image data to the server. This transmission is performed via a protocol that enables efficient and low-latency communication (e.g., WebSocket).

[0233] Step 4:

[0234] The server inputs the received image data into a generative computer model. The model analyzes the image frames, classifies facial expressions and gestures, and labels them as emotional and gesture information.

[0235] Step 5:

[0236] The server converts the analyzed emotion and gesture information into natural language text. Here, the information is expressed in language that is intuitively easy for the user to understand.

[0237] Step 6:

[0238] The server passes natural language text to a speech synthesis engine, which generates speech data. The speech data is then adjusted to facilitate user comprehension.

[0239] Step 7:

[0240] The server sends the generated audio data back to the terminal. To minimize communication delays during this process, an appropriate protocol is used.

[0241] Step 8:

[0242] The device plays back audio data through an audio output device. This allows the user to receive analyzed facial and gesture information as audio feedback.

[0243] Step 9:

[0244] The conversation flow is adjusted based on the user's voice feedback. For example, if the emotion label is "confused," additional explanations or follow-ups are considered.

[0245] (Example 1)

[0246] Next, we will describe Example 1. 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."

[0247] There is a need to effectively convey nonverbal communication to users who cannot directly access visual information. Furthermore, in online meetings and discussions, it is difficult to grasp participants' emotions and intentions in real time, posing a barrier to smooth communication. Conventional technologies have been unable to adequately support user understanding by relying solely on audio information to supplement this visual information.

[0248] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0249] In this invention, the server includes means for acquiring video information using digital information transmission technology with an image acquisition device, means for analyzing the received video information using a generative data processing model and outputting emotion labels and action labels, and means for providing the analysis results to the user as audio information using a voice generation device. This enables the user to grasp the emotions and intentions of participants in real time during online meetings and conversations, and to communicate smoothly.

[0250] An "image acquisition device" is a device used to acquire visual information using digital information transmission technology.

[0251] A "generative data processing model" is a machine learning-based model used to analyze received video information and output emotion labels and action labels.

[0252] A "voice generation device" is a device that converts analysis results into voice information and provides it to the user.

[0253] "Visual information" refers to visual data expressed in digital format, including images and video data acquired during online meetings and discussions.

[0254] An "emotional label" is an identifier extracted from a person's facial expressions and gestures that indicates a specific emotional state.

[0255] A "movement label" is an identifier extracted to represent human movement or gesture.

[0256] "Audio information" refers to data that expresses text data or analysis results as audio.

[0257] This invention constructs a system for converting visual information into audio information, and is primarily implemented using an image acquisition device, a generative data processing model, and a speech generation device. These elements work together to provide visual information as audio in real time, thereby supporting users who cannot directly acquire visual information.

[0258] Users initiate a meeting using online communication technology. The terminal captures video of the meeting using connected image acquisition devices such as cameras. The video information includes facial expressions and gestures of the participants frame by frame, and this is sent to the server.

[0259] The server inputs the received video information into a generative data processing model. This model is based on a pre-trained dataset and extracts emotion and behavior labels with high accuracy. Specifically, the model utilizes machine learning and deep learning technologies.

[0260] The obtained emotion and action labels are converted into natural language by a speech generator on the server. This converts the emotion and action labels into voice messages that are understandable to the user.

[0261] The device receives the audio data and plays it back in real time through an audio output device. Visual information is presented to the user as audio information via speakers or headphones.

[0262] As a concrete example, consider a situation in a meeting where the presenter cannot visually confirm the confused expressions of the participants. The system extracts the confused label and delivers an audio message to the presenter saying, "The participant is confused; please consider providing additional explanations." In this way, the presenter can immediately adjust their response based on the participants' reactions.

[0263] An example of a prompt might be: "Describe a system that analyzes the facial expressions of meeting participants and provides audio feedback on emotions such as confusion and joy." This would provide guidance for understanding and implementing the system.

[0264] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0265] Step 1:

[0266] The user initiates a meeting using online communication technology. This prepares the system for capturing real-time video of meeting participants. The device uses its connected camera to acquire video information in real time. It receives the meeting video feed as input and generates frame-by-frame image data as output.

[0267] Step 2:

[0268] The terminal sends the image data for each frame it acquires to the server. The frame data includes the person's facial expressions and movements. Specifically, the terminal packets the image data over the network and sends it to the server. The input here is the image data for each frame, and the output is the image packets that arrive at the server.

[0269] Step 3:

[0270] The server inputs the received image data into a generative data processing model. This model has been pre-trained on various facial expressions and actions, and analyzes the image data to output emotion labels and action labels. Specifically, the model processes the pixel information of each frame to extract features and generates labels based on them. The input is the image data received by the server, and the output is emotion labels and action labels.

[0271] Step 4:

[0272] The server performs a process of converting the generated emotion and behavior labels into natural language text. Next, it uses speech synthesis technology to convert the text into speech data. Specifically, it converts emotion labels into text and applies a speech synthesis algorithm to generate a speech file. The input is emotion and behavior labels, and the output is natural language speech data.

[0273] Step 5:

[0274] The terminal receives audio data from the server and provides real-time feedback to the user through an audio playback device. Specifically, the terminal decodes the audio data and plays it back through a speaker or headphones. The input is audio data from the server, and the output is the real-time audio message heard by the user.

[0275] (Application Example 1)

[0276] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0277] People with visual impairments face challenges in communicating smoothly with others within their families. This problem stems from their difficulty in understanding nonverbal information such as facial expressions and gestures. Therefore, it is necessary to reduce communication barriers by converting visual information into auditory information and effectively communicating it to people with visual impairments.

[0278] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0279] In this invention, the server includes means for acquiring visual data using communication technology, means for processing the received visual data with a generative computer program and outputting emotion data and motion data, and means for transmitting the processing results to the user as audio using an audio output device. This makes it easier for visually impaired users to understand the emotions and gestures of others, and enables smooth communication within the home.

[0280] A "video acquisition device" is a device that acquires visual data using communication technology.

[0281] "Communication technology" refers to technologies for exchanging data over long distances.

[0282] "Visual data" refers to a collection of information that is represented as images or videos.

[0283] A "generative computer program" is a computer program that processes visual data and generates emotional and behavioral data.

[0284] "Emotional data" refers to information about human emotions that is analyzed from visual data.

[0285] "Motion data" refers to information on human motion analyzed from visual data.

[0286] "Voice output device" refers to a device for transmitting voice to the user.

[0287] "User" refers to a person who receives assistance from this system.

[0288] "Visual impairment" refers to a state in which it is difficult to recognize visual information.

[0289] "Indoor" refers to a living space centered around a residence.

[0290] The present invention is a communication support system for users with visual impairments using communication technology. This system operates in cooperation with a video acquisition device, a generative computer program, and a voice output device.

[0291] The server receives visual data acquired in real time by the video acquisition device. The visual data includes the expressions and gestures of the speaker. This data is sent to a generative computer program on the server and analyzed into emotion data and motion data. A pre-learned data group is used in the generative computer, and expression judgment and gesture classification are performed quickly and with high accuracy.

[0292] The server converts the generated emotion data and motion data into natural language and makes voice data using voice synthesis software. This voice data is provided to the user with visual impairment indoors through a voice output device.

[0293] As a specific example, when interacting with family members, the camera of the robot captures the visual information of the other party, and the change in its expression is transmitted to the user as a voice message "The other party is smiling." This compensates for the lack of visual information and makes communication smoother.

[0294] An example of a prompt for facial expression analysis using a generative AI model is as follows: "Analyze the facial expression of the person captured by the camera and describe their emotion in voice."

[0295] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0296] Step 1:

[0297] The terminal captures visual data in real time using video acquisition equipment. This data includes the facial expressions and gestures of the person being spoken to. The input is video data from the camera, and the output is visual data sent to the server. The terminal sends this data to the server for the next processing step.

[0298] Step 2:

[0299] The server inputs the received visual data into a generative AI model. The generative AI model analyzes emotion data and behavior data using a pre-trained dataset. The input is visual data, and the output is the analyzed emotion data and behavior data. Based on this data processing, the server decomposes the visual information into emotion information and gesture information.

[0300] Step 3:

[0301] The server converts the analyzed emotion and behavior data into natural language. This conversion process prepares the information to be conveyed to the user into easily understandable language. The input is emotion and behavior data, and the output is the language expression as text data. The server then prepares this text for conversion into speech.

[0302] Step 4:

[0303] The server converts text data into voice data using voice synthesis technology. Specifically, it uses voice synthesis software to compose it as natural voice. The input is text data converted into natural language, and the output is the generated voice data. The server transmits this voice data to the terminal.

[0304] Step 5:

[0305] The terminal transmits the received voice data to the user using a voice output device. As a result, the user can understand the emotions and motion information based on the expression and body movements of the other party through voice. The input is voice data, and the output is the voice that the user hears. The terminal plays the voice at an appropriate timing to assist communication.

[0306] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[0307] In an embodiment of the present invention, a system is constructed in which a video acquisition device, a generative computer model, a voice output device, and an emotion engine are combined to analyze the user's non-verbal information in real time and provide it to the user through voice feedback.

[0308] First, the user starts a meeting or conversation through online communication technology. The terminal uses a camera to capture the video of this conversation in real time and obtains it as image data. This image data is transmitted to the server for detailed analysis.

[0309] On the server, a generative computing model analyzes the received image data. This model is pre-trained on a large amount of facial expression and gesture data, and can extract emotional and gesture information from the data. Next, an emotion engine refines the results of the generative computing model, generating more detailed and accurate emotional information. This engine tracks and analyzes multiple emotional states in real time, providing personalized and optimized feedback.

[0310] The server then converts the analyzed emotion and gesture information into natural language text and generates easily understandable audio data through a speech synthesis engine. This audio data is sent to the terminal and provided to the user in real time through an audio output device.

[0311] As a concrete example, imagine a user presenting in a meeting who observes that one of the participants is showing a confused expression and restless gestures. The server's emotion engine then analyzes this information with high accuracy and generates and provides the user with audio feedback stating, "The participant appears confused and may need further explanation." This allows the user to immediately adjust their response and be considerate of the participant.

[0312] This system plays a role in facilitating the understanding of visual and nonverbal information, especially in situations where visual information is difficult to obtain, and supporting smooth and effective communication with users.

[0313] The following describes the processing flow.

[0314] Step 1:

[0315] The user initiates a meeting via online communication software. This enables the device to activate its camera and prepare for video capture.

[0316] Step 2:

[0317] The device continuously captures video of the conversation using its camera, collecting image data in real time. This data is then converted to an appropriate format and temporarily stored.

[0318] Step 3:

[0319] The device processes the collected image data and sends it to the server for analysis. The transmission process is optimized to maintain stable communication.

[0320] Step 4:

[0321] The server analyzes image data using a generative computing model. Based on a pre-trained dataset, the model detects facial expressions and gestures and extracts emotional and gesture information.

[0322] Step 5:

[0323] The server applies an emotion engine to the extracted emotion information. This engine performs a more detailed emotion analysis, evaluates multiple emotional states in real time, and tracks dynamic changes in emotions.

[0324] Step 6:

[0325] The server converts the detailed information obtained by the emotion engine into natural language and generates audio data using a speech synthesis engine. This audio data is then formatted to be easily understood by the user.

[0326] Step 7:

[0327] The server generates audio data, which is then sent to the terminal and played back to the user in real time via an audio output device.

[0328] Step 8:

[0329] Users receive voice feedback, which helps to adjust the meeting's progress and the flow of conversation. For example, if a participant seems confused, additional explanations can be provided to improve communication.

[0330] (Example 2)

[0331] Next, we will describe Example 2. 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".

[0332] In online meetings and discussions, accurately understanding participants' emotions and actions, and facilitating smooth and effective communication, presents a challenge. In particular, analyzing nonverbal information is complex, and there is a need for technology that provides real-time feedback to users.

[0333] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0334] In this invention, the server includes means for acquiring image data in real time using online communication technology, means for analyzing emotional information and behavioral information from the image data using generative computing technology, and means for generating natural language text based on the analyzed emotional information and behavioral information. This makes it possible to efficiently and accurately analyze complex non-verbal information and provide real-time feedback to the user.

[0335] "Online communication technology" refers to technology that uses the internet or networks to send and receive information in real time.

[0336] "Image data" refers to data that represents visual information acquired using devices such as cameras in a digital format.

[0337] "Generative computing technology" refers to techniques that use artificial intelligence and machine learning models to generate and analyze new information from input data.

[0338] "Emotional information" refers to information that identifies and classifies an individual's emotional state, indicating their psychological state as inferred from facial expressions, body movements, and other factors.

[0339] "Motion information" refers to information obtained by analyzing a person's physical movements, such as gestures and body language.

[0340] "Natural language text" refers to text written in a language that humans use in everyday life, and is a machine-generated, easily interpretable linguistic expression.

[0341] "Speech generation technology" is a technology that converts text data into speech and synthesizes it to sound like a human voice.

[0342] "Providing in real time" means delivering information and data to users instantly and without delay.

[0343] This invention provides an effective communication support system using online communication. Specific embodiments for carrying out the invention are described below.

[0344] Users utilize this system, for example, in online meetings or conversations. The terminal is equipped with a camera, allowing for real-time capture of the user's facial expressions and gestures. This video data is then transmitted from the terminal to the server.

[0345] The server analyzes the received video data using a generative AI model. This generative AI model is pre-trained on a large dataset collected in the past and is designed to extract emotional and motion information by performing facial expression classification and motion analysis.

[0346] Next, the server uses this information to generate natural language text. This text is converted into audio data via a speech output device and sent to the terminal. This allows the user to receive audio feedback in real time.

[0347] For example, if a participant shows a confused expression while a user is giving a presentation, the generative AI model analyzes that emotion and provides the user with audio feedback such as, "It appears the participant has some questions about the explanation." In this way, the user can immediately adjust the content of the presentation and achieve smoother communication.

[0348] An example of a prompt message is, "Show the user how to get real-time emotional feedback," which can elicit a response from the system.

[0349] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0350] Step 1:

[0351] The user starts an online meeting, and the device acquires video data in real time via its camera. The input here is the video of the user and participants during the meeting, and the output is the acquisition of video data as frames. Specifically, the camera is activated and consecutive frames are captured.

[0352] Step 2:

[0353] The terminal compresses the acquired video data and sends it to the server using a transmission protocol. The input is frame data captured in real time, and the output is compressed video data. Specifically, a data compression algorithm is applied to efficiently transfer the data to the server.

[0354] Step 3:

[0355] The server receives the transmitted video data and analyzes it using a generative AI model. The input for analysis is compressed video data, and the output is extracted emotion and motion information. Specifically, the AI ​​model evaluates each frame and identifies facial expressions and movements.

[0356] Step 4:

[0357] The server uses an emotion engine to generate detailed natural language text based on the results of the generative AI model. The input consists of emotion and behavior information, and the output is a natural language text representation. Specifically, a text generation algorithm tailored to the emotional state is executed.

[0358] Step 5:

[0359] The server converts the generated natural language text into audio data using speech generation technology and sends it to the terminal. The input is the generated natural language text, and the output is the synthesized audio data. The specific operation includes generating an audio file using a text-to-speech engine.

[0360] Step 6:

[0361] The terminal provides the user with the received audio data through an audio output device. The input is synthesized audio data, and the output is real-time audio feedback. Specifically, the audio file is decoded and played back through the speaker.

[0362] (Application Example 2)

[0363] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0364] In modern elderly care settings, understanding the emotions and psychological state of residents is crucial. However, it is difficult for busy care staff to respond immediately to the emotional changes of each resident. Furthermore, nonverbal information is easily overlooked using traditional methods, requiring greater ingenuity to improve resident satisfaction and a sense of security. Therefore, there is a need to analyze and report residents' emotional states in real time, reducing the workload of care staff while providing meticulous care.

[0365] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0366] In this invention, the server includes means for acquiring image data using a video acquisition device, means for analyzing the image data using a generative computer model to obtain emotional information, and means for converting emotional feedback into speech using a speech synthesis device and providing it to the user. This enables the acquisition of nonverbal emotional information of residents in real time in care settings, allowing care staff to respond quickly and appropriately to changes in their emotions.

[0367] A "video acquisition device" is a device that acquires image data in real time using online communication technology.

[0368] A "generative computer model" is a computer model that analyzes received image data based on a pre-trained dataset and extracts and outputs emotional information and gesture information.

[0369] An "emotion engine" is an engine that refines the analysis results of generative computer models and generates personalized emotional feedback.

[0370] A "speech synthesis device" is a device that converts analyzed emotional feedback into speech data and provides it to the user in an easily understandable format.

[0371] A "care support device" is a device that includes a system that provides voice data from a speech synthesis device to care staff and reports the emotional state of residents in real time.

[0372] "Emotional feedback" is feedback information that uses non-verbal information acquired to convey the emotional state of residents to care staff and others via voice.

[0373] The system that realizes this application example includes a video acquisition device, a generative computer model, an emotion engine, a speech synthesizer, and a care support device. The server uses real-time image data obtained via the video acquisition device and utilizes the generative computer model to analyze the data and extract nonverbal emotion and gesture information. Because this analysis uses a pre-trained dataset, highly accurate emotion classification is possible.

[0374] The server further refines the results of the generative computer model using an emotion engine to generate personalized, detailed emotional feedback, which is then converted into speech data by a speech synthesizer. This speech data is sent to a care support device and provided to care staff in real time.

[0375] As a concrete example, if a resident suddenly shows signs of restlessness, the server's emotion engine analyzes the information and delivers an audio feedback message to the care staff via the care support device stating, "The resident may be feeling anxious." This allows the staff to respond quickly and reassure the resident.

[0376] An example of a prompt message is, "Design an app that analyzes the facial expressions and movements of residents and provides voice feedback on their emotions."

[0377] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0378] Step 1:

[0379] The server receives image data acquired in real time from the video acquisition device via online communication technology. This input image data includes non-verbal information, such as the residents' facial expressions and gestures. The server then transmits this data directly to a generative computer model.

[0380] Step 2:

[0381] The server analyzes the received image data using a generative computing model. This process identifies facial expressions and other nonverbal cues based on the input image data, and extracts emotional and gesture information. The output is data indicating the analyzed emotional state.

[0382] Step 3:

[0383] The server uses an emotion engine to refine the results of the generative computer model's analysis. In this step, it generates more detailed and personalized emotion feedback based on the analyzed data. The input is the results of the generative computer model's analysis, and the output is the refined emotion feedback data.

[0384] Step 4:

[0385] The server uses a speech synthesizer to convert emotional feedback into audio data. This process synthesizes the input into an easily understandable audio format and converts it in real time. The output becomes audio data usable by care support devices.

[0386] Step 5:

[0387] The terminal receives audio data from the server and provides real-time audio feedback to care staff through the care support device. In this step, information about the resident's emotions is immediately transmitted to the care staff, and information is output to enable them to take appropriate action.

[0388] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0389] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0390] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0391] [Third Embodiment]

[0392] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0393] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0394] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0395] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0396] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0397] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0398] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0399] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0400] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0401] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0402] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0403] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0404] In this embodiment of the invention, a video acquisition device, a generative computer model, and an audio output device work together to construct a system that effectively converts visual information into audio information and provides it.

[0405] First, the user initiates a meeting or discussion using online communication technology. The device captures the video of the meeting and acquires image data. This image data consists of frame-by-frame data, including the facial expressions and gestures of the participants, and is sent to the server.

[0406] The server inputs the received image data into a generative computer model. This model has been pre-trained on a large amount of data on facial expressions and gestures, and analyzes human emotional and gesture information with high accuracy. Through this analysis, emotional labels such as "happy" or "confused" and gesture labels such as "nodding" are obtained.

[0407] Next, the server converts these analysis results into natural language and generates audio data using speech synthesis technology. This audio data is structured to be easily understood by the user.

[0408] The device receives the generated audio data and plays it back in real time through the user's audio output device, thereby transmitting visual information in the form of text and audio. This makes it easier for the user to grasp the emotions and intentions conveyed by the visual information, enabling smoother communication.

[0409] As a concrete example, consider a scenario where a user is a presenter during a meeting, and one of the participants shows a confused expression. A video capture device captures this expression in real time, and a server analyzes it as "confused." Based on this result, the user receives audio feedback saying, "The other person is confused, please consider providing additional explanations." This feedback allows the user to immediately adjust their response.

[0410] This system facilitates the understanding of nonverbal communication even when direct visual information is unavailable, supporting smoother conversations, especially for users with visual impairments.

[0411] The following describes the processing flow.

[0412] Step 1:

[0413] The user initiates a meeting via online communication technology. This sets up the device to activate its camera and begin capturing video.

[0414] Step 2:

[0415] The device acquires video data from the camera in real time. The acquired video data is temporarily stored in memory as individual frames. The video data is also converted to an appropriate format (e.g., JPEG, PNG).

[0416] Step 3:

[0417] The device sends the captured image data to the server. This transmission is performed via a protocol that enables efficient and low-latency communication (e.g., WebSocket).

[0418] Step 4:

[0419] The server inputs the received image data into a generative computer model. The model analyzes the image frames, classifies facial expressions and gestures, and labels them as emotional and gesture information.

[0420] Step 5:

[0421] The server converts the analyzed emotion and gesture information into natural language text. Here, the information is expressed in language that is intuitively easy for the user to understand.

[0422] Step 6:

[0423] The server passes natural language text to a speech synthesis engine, which generates speech data. The speech data is then adjusted to facilitate user comprehension.

[0424] Step 7:

[0425] The server sends the generated audio data back to the terminal. To minimize communication delays during this process, an appropriate protocol is used.

[0426] Step 8:

[0427] The device plays back audio data through an audio output device. This allows the user to receive analyzed facial and gesture information as audio feedback.

[0428] Step 9:

[0429] The conversation flow is adjusted based on the user's voice feedback. For example, if the emotion label is "confused," additional explanations or follow-ups are considered.

[0430] (Example 1)

[0431] Next, we will describe Example 1. 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."

[0432] There is a need to effectively convey nonverbal communication to users who cannot directly access visual information. Furthermore, in online meetings and discussions, it is difficult to grasp participants' emotions and intentions in real time, posing a barrier to smooth communication. Conventional technologies have been unable to adequately support user understanding by relying solely on audio information to supplement this visual information.

[0433] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0434] In this invention, the server includes means for acquiring video information using digital information transmission technology with an image acquisition device, means for analyzing the received video information using a generative data processing model and outputting emotion labels and action labels, and means for providing the analysis results to the user as audio information using a voice generation device. This enables the user to grasp the emotions and intentions of participants in real time during online meetings and conversations, and to communicate smoothly.

[0435] An "image acquisition device" is a device used to acquire visual information using digital information transmission technology.

[0436] A "generative data processing model" is a machine learning-based model used to analyze received video information and output emotion labels and action labels.

[0437] A "voice generation device" is a device that converts analysis results into voice information and provides it to the user.

[0438] "Visual information" refers to visual data expressed in digital format, including images and video data acquired during online meetings and discussions.

[0439] An "emotional label" is an identifier extracted from a person's facial expressions and gestures that indicates a specific emotional state.

[0440] A "movement label" is an identifier extracted to represent human movement or gesture.

[0441] "Audio information" refers to data that expresses text data or analysis results as audio.

[0442] This invention constructs a system for converting visual information into audio information, and is primarily implemented using an image acquisition device, a generative data processing model, and a speech generation device. These elements work together to provide visual information as audio in real time, thereby supporting users who cannot directly acquire visual information.

[0443] Users initiate a meeting using online communication technology. The terminal captures video of the meeting using connected image acquisition devices such as cameras. The video information includes facial expressions and gestures of the participants frame by frame, and this is sent to the server.

[0444] The server inputs the received video information into a generative data processing model. This model is based on a pre-trained dataset and extracts emotion and behavior labels with high accuracy. Specifically, the model utilizes machine learning and deep learning technologies.

[0445] The obtained emotion and action labels are converted into natural language by a speech generator on the server. This converts the emotion and action labels into voice messages that are understandable to the user.

[0446] The device receives the audio data and plays it back in real time through an audio output device. Visual information is presented to the user as audio information via speakers or headphones.

[0447] As a concrete example, consider a situation in a meeting where the presenter cannot visually confirm the confused expressions of the participants. The system extracts the confused label and delivers an audio message to the presenter saying, "The participant is confused; please consider providing additional explanations." In this way, the presenter can immediately adjust their response based on the participants' reactions.

[0448] An example of a prompt might be: "Describe a system that analyzes the facial expressions of meeting participants and provides audio feedback on emotions such as confusion and joy." This would provide guidance for understanding and implementing the system.

[0449] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0450] Step 1:

[0451] The user initiates a meeting using online communication technology. This prepares the system for capturing real-time video of meeting participants. The device uses its connected camera to acquire video information in real time. It receives the meeting video feed as input and generates frame-by-frame image data as output.

[0452] Step 2:

[0453] The terminal sends the image data for each frame it acquires to the server. The frame data includes the person's facial expressions and movements. Specifically, the terminal packets the image data over the network and sends it to the server. The input here is the image data for each frame, and the output is the image packets that arrive at the server.

[0454] Step 3:

[0455] The server inputs the received image data into a generative data processing model. This model has been pre-trained on various facial expressions and actions, and analyzes the image data to output emotion labels and action labels. Specifically, the model processes the pixel information of each frame to extract features and generates labels based on them. The input is the image data received by the server, and the output is emotion labels and action labels.

[0456] Step 4:

[0457] The server performs a process of converting the generated emotion and behavior labels into natural language text. Next, it uses speech synthesis technology to convert the text into speech data. Specifically, it converts emotion labels into text and applies a speech synthesis algorithm to generate a speech file. The input is emotion and behavior labels, and the output is natural language speech data.

[0458] Step 5:

[0459] The terminal receives audio data from the server and provides real-time feedback to the user through an audio playback device. Specifically, the terminal decodes the audio data and plays it back through a speaker or headphones. The input is audio data from the server, and the output is the real-time audio message heard by the user.

[0460] (Application Example 1)

[0461] Next, we will explain Application Example 1. In the following explanation, 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."

[0462] People with visual impairments face challenges in communicating smoothly with others within their families. This problem stems from their difficulty in understanding nonverbal information such as facial expressions and gestures. Therefore, it is necessary to reduce communication barriers by converting visual information into auditory information and effectively communicating it to people with visual impairments.

[0463] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0464] In this invention, the server includes means for acquiring visual data using communication technology, means for processing the received visual data with a generative computer program and outputting emotion data and motion data, and means for transmitting the processing results to the user as audio using an audio output device. This makes it easier for visually impaired users to understand the emotions and gestures of others, and enables smooth communication within the home.

[0465] A "video acquisition device" is a device that acquires visual data using communication technology.

[0466] "Communication technology" refers to technologies for exchanging data over long distances.

[0467] "Visual data" refers to a collection of information that is represented as images or videos.

[0468] A "generative computer program" is a computer program that processes visual data and generates emotional and behavioral data.

[0469] "Emotional data" refers to information about human emotions that is analyzed from visual data.

[0470] "Motion data" refers to information about human movement that is analyzed from visual data.

[0471] An "audio output device" is a device used to transmit audio to a user.

[0472] A "user" is a person who receives support through this system.

[0473] "Visual impairment" refers to a condition in which a person has difficulty recognizing visual information.

[0474] "Within the home" refers to the living space centered around the residence.

[0475] This invention is a communication support system for visually impaired users that utilizes communication technology. This system operates in conjunction with a video acquisition device, a generative computer program, and an audio output device.

[0476] The server receives visual data acquired in real time by video acquisition equipment. This visual data includes the speaker's facial expressions and gestures. This data is sent to a generative computer program on the server and analyzed into emotion data and motion data. The generative computer uses a pre-trained dataset, enabling rapid and highly accurate facial expression recognition and gesture classification.

[0477] The server converts the generated emotion and behavior data into natural language and then uses speech synthesis software to create audio data. This audio data is then provided to visually impaired users within their homes via audio output devices.

[0478] As a concrete example, when interacting with family members, the robot's camera captures the other person's visual information and conveys the change in their facial expression to the user as an audio message such as, "The other person is smiling." This compensates for the lack of visual information and facilitates smoother communication.

[0479] An example of a prompt for facial expression analysis using a generative AI model is as follows: "Analyze the facial expression of the person captured by the camera and describe their emotion in voice."

[0480] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0481] Step 1:

[0482] The terminal captures visual data in real time using video acquisition equipment. This data includes the facial expressions and gestures of the person being spoken to. The input is video data from the camera, and the output is visual data sent to the server. The terminal sends this data to the server for the next processing step.

[0483] Step 2:

[0484] The server inputs the received visual data into a generative AI model. The generative AI model analyzes emotion data and behavior data using a pre-trained dataset. The input is visual data, and the output is the analyzed emotion data and behavior data. Based on this data processing, the server decomposes the visual information into emotion information and gesture information.

[0485] Step 3:

[0486] The server converts the analyzed emotion and behavior data into natural language. This conversion process prepares the information to be conveyed to the user into easily understandable language. The input is emotion and behavior data, and the output is the language expression as text data. The server then prepares this text for conversion into speech.

[0487] Step 4:

[0488] The server converts text data into speech data using speech synthesis technology. Specifically, it uses speech synthesis software to construct natural-sounding speech. The input is text data converted into natural language, and the output is the generated speech data. The server sends this speech data to the terminal.

[0489] Step 5:

[0490] The device transmits received audio data to the user using an audio output device. This allows the user to understand emotions and actions based on the other person's facial expressions and gestures through audio. The input is audio data, and the output is the audio the user hears. The device plays the audio at the appropriate time to support communication.

[0491] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0492] In embodiments of the present invention, a video acquisition device, a generative computer model, an audio output device, and an emotion engine are combined to construct a system that analyzes the user's nonverbal information in real time and provides it to the user through audio feedback.

[0493] First, users initiate a meeting or conversation using online communication technology. The device uses its camera to capture video of this conversation in real time, acquiring it as image data. This image data is then sent to a server for detailed analysis.

[0494] On the server, a generative computing model analyzes the received image data. This model is pre-trained on a large amount of facial expression and gesture data, and can extract emotional and gesture information from the data. Next, an emotion engine refines the results of the generative computing model, generating more detailed and accurate emotional information. This engine tracks and analyzes multiple emotional states in real time, providing personalized and optimized feedback.

[0495] The server then converts the analyzed emotion and gesture information into natural language text and generates easily understandable audio data through a speech synthesis engine. This audio data is sent to the terminal and provided to the user in real time through an audio output device.

[0496] As a concrete example, imagine a user presenting in a meeting who observes that one of the participants is showing a confused expression and restless gestures. The server's emotion engine then analyzes this information with high accuracy and generates and provides the user with audio feedback stating, "The participant appears confused and may need further explanation." This allows the user to immediately adjust their response and be considerate of the participant.

[0497] This system plays a role in facilitating the understanding of visual and nonverbal information, especially in situations where visual information is difficult to obtain, and supporting smooth and effective communication with users.

[0498] The following describes the processing flow.

[0499] Step 1:

[0500] The user initiates a meeting via online communication software. This enables the device to activate its camera and prepare for video capture.

[0501] Step 2:

[0502] The device continuously captures video of the conversation using its camera, collecting image data in real time. This data is then converted to an appropriate format and temporarily stored.

[0503] Step 3:

[0504] The device processes the collected image data and sends it to the server for analysis. The transmission process is optimized to maintain stable communication.

[0505] Step 4:

[0506] The server analyzes image data using a generative computing model. Based on a pre-trained dataset, the model detects facial expressions and gestures and extracts emotional and gesture information.

[0507] Step 5:

[0508] The server applies an emotion engine to the extracted emotion information. This engine performs a more detailed emotion analysis, evaluates multiple emotional states in real time, and tracks dynamic changes in emotions.

[0509] Step 6:

[0510] The server converts the detailed information obtained by the emotion engine into natural language and generates audio data using a speech synthesis engine. This audio data is then formatted to be easily understood by the user.

[0511] Step 7:

[0512] The server generates audio data, which is then sent to the terminal and played back to the user in real time via an audio output device.

[0513] Step 8:

[0514] Users receive voice feedback, which helps to adjust the meeting's progress and the flow of conversation. For example, if a participant seems confused, additional explanations can be provided to improve communication.

[0515] (Example 2)

[0516] Next, we will describe Example 2. 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."

[0517] In online meetings and discussions, accurately understanding participants' emotions and actions, and facilitating smooth and effective communication, presents a challenge. In particular, analyzing nonverbal information is complex, and there is a need for technology that provides real-time feedback to users.

[0518] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0519] In this invention, the server includes means for acquiring image data in real time using online communication technology, means for analyzing emotional information and behavioral information from the image data using generative computing technology, and means for generating natural language text based on the analyzed emotional information and behavioral information. This makes it possible to efficiently and accurately analyze complex non-verbal information and provide real-time feedback to the user.

[0520] "Online communication technology" refers to technology that uses the internet or networks to send and receive information in real time.

[0521] "Image data" refers to data that represents visual information acquired using devices such as cameras in a digital format.

[0522] "Generative computing technology" refers to techniques that use artificial intelligence and machine learning models to generate and analyze new information from input data.

[0523] "Emotional information" refers to information that identifies and classifies an individual's emotional state, indicating their psychological state as inferred from facial expressions, body movements, and other factors.

[0524] "Motion information" refers to information obtained by analyzing a person's physical movements, such as gestures and body language.

[0525] "Natural language text" refers to text written in a language that humans use in everyday life, and is a machine-generated, easily interpretable linguistic expression.

[0526] "Speech generation technology" is a technology that converts text data into speech and synthesizes it to sound like a human voice.

[0527] "Providing in real time" means delivering information and data to users instantly and without delay.

[0528] This invention provides an effective communication support system using online communication. Specific embodiments for carrying out the invention are described below.

[0529] Users utilize this system, for example, in online meetings or conversations. The terminal is equipped with a camera, allowing for real-time capture of the user's facial expressions and gestures. This video data is then transmitted from the terminal to the server.

[0530] The server analyzes the received video data using a generative AI model. This generative AI model is pre-trained on a large dataset collected in the past and is designed to extract emotional and motion information by performing facial expression classification and motion analysis.

[0531] Next, the server uses this information to generate natural language text. This text is converted into audio data via a speech output device and sent to the terminal. This allows the user to receive audio feedback in real time.

[0532] For example, if a participant shows a confused expression while a user is giving a presentation, the generative AI model analyzes that emotion and provides the user with audio feedback such as, "It appears the participant has some questions about the explanation." In this way, the user can immediately adjust the content of the presentation and achieve smoother communication.

[0533] An example of a prompt message is, "Show the user how to get real-time emotional feedback," which can elicit a response from the system.

[0534] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0535] Step 1:

[0536] The user starts an online meeting, and the device acquires video data in real time via its camera. The input here is the video of the user and participants during the meeting, and the output is the acquisition of video data as frames. Specifically, the camera is activated and consecutive frames are captured.

[0537] Step 2:

[0538] The terminal compresses the acquired video data and sends it to the server using a transmission protocol. The input is frame data captured in real time, and the output is compressed video data. Specifically, a data compression algorithm is applied to efficiently transfer the data to the server.

[0539] Step 3:

[0540] The server receives the transmitted video data and analyzes it using a generative AI model. The input for analysis is compressed video data, and the output is extracted emotion and motion information. Specifically, the AI ​​model evaluates each frame and identifies facial expressions and movements.

[0541] Step 4:

[0542] The server uses an emotion engine to generate detailed natural language text based on the results of the generative AI model. The input consists of emotion and behavior information, and the output is a natural language text representation. Specifically, a text generation algorithm tailored to the emotional state is executed.

[0543] Step 5:

[0544] The server converts the generated natural language text into audio data using speech generation technology and sends it to the terminal. The input is the generated natural language text, and the output is the synthesized audio data. The specific operation includes generating an audio file using a text-to-speech engine.

[0545] Step 6:

[0546] The terminal provides the user with the received audio data through an audio output device. The input is synthesized audio data, and the output is real-time audio feedback. Specifically, the audio file is decoded and played back through the speaker.

[0547] (Application Example 2)

[0548] Next, we will explain Application Example 2. In the following explanation, 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."

[0549] In modern elderly care settings, understanding the emotions and psychological state of residents is crucial. However, it is difficult for busy care staff to respond immediately to the emotional changes of each resident. Furthermore, nonverbal information is easily overlooked using traditional methods, requiring greater ingenuity to improve resident satisfaction and a sense of security. Therefore, there is a need to analyze and report residents' emotional states in real time, reducing the workload of care staff while providing meticulous care.

[0550] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0551] In this invention, the server includes means for acquiring image data using a video acquisition device, means for analyzing the image data using a generative computer model to obtain emotional information, and means for converting emotional feedback into speech using a speech synthesis device and providing it to the user. This enables the acquisition of nonverbal emotional information of residents in real time in care settings, allowing care staff to respond quickly and appropriately to changes in their emotions.

[0552] A "video acquisition device" is a device that acquires image data in real time using online communication technology.

[0553] A "generative computer model" is a computer model that analyzes received image data based on a pre-trained dataset and extracts and outputs emotional information and gesture information.

[0554] An "emotion engine" is an engine that refines the analysis results of generative computer models and generates personalized emotional feedback.

[0555] A "speech synthesis device" is a device that converts analyzed emotional feedback into speech data and provides it to the user in an easily understandable format.

[0556] A "care support device" is a device that includes a system that provides voice data from a speech synthesis device to care staff and reports the emotional state of residents in real time.

[0557] "Emotional feedback" is feedback information that uses non-verbal information acquired to convey the emotional state of residents to care staff and others via voice.

[0558] The system that realizes this application example includes a video acquisition device, a generative computer model, an emotion engine, a speech synthesizer, and a care support device. The server uses real-time image data obtained via the video acquisition device and utilizes the generative computer model to analyze the data and extract nonverbal emotion and gesture information. Because this analysis uses a pre-trained dataset, highly accurate emotion classification is possible.

[0559] The server further refines the results of the generative computer model using an emotion engine to generate personalized, detailed emotional feedback, which is then converted into speech data by a speech synthesizer. This speech data is sent to a care support device and provided to care staff in real time.

[0560] As a concrete example, if a resident suddenly shows signs of restlessness, the server's emotion engine analyzes the information and delivers an audio feedback message to the care staff via the care support device stating, "The resident may be feeling anxious." This allows the staff to respond quickly and reassure the resident.

[0561] An example of a prompt message is, "Design an app that analyzes the facial expressions and movements of residents and provides voice feedback on their emotions."

[0562] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0563] Step 1:

[0564] The server receives image data acquired in real time from the video acquisition device via online communication technology. This input image data includes non-verbal information, such as the residents' facial expressions and gestures. The server then transmits this data directly to a generative computer model.

[0565] Step 2:

[0566] The server analyzes the received image data using a generative computing model. This process identifies facial expressions and other nonverbal cues based on the input image data, and extracts emotional and gesture information. The output is data indicating the analyzed emotional state.

[0567] Step 3:

[0568] The server uses an emotion engine to refine the results of the generative computer model's analysis. In this step, it generates more detailed and personalized emotion feedback based on the analyzed data. The input is the results of the generative computer model's analysis, and the output is the refined emotion feedback data.

[0569] Step 4:

[0570] The server uses a speech synthesizer to convert emotional feedback into audio data. This process synthesizes the input into an easily understandable audio format and converts it in real time. The output becomes audio data usable by care support devices.

[0571] Step 5:

[0572] The terminal receives audio data from the server and provides real-time audio feedback to care staff through the care support device. In this step, information about the resident's emotions is immediately transmitted to the care staff, and information is output to enable them to take appropriate action.

[0573] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0574] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0575] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0576] [Fourth Embodiment]

[0577] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0578] As shown in Figure 7, the 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.

[0579] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0580] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0581] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0582] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0583] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0584] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0585] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0586] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0587] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0588] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0589] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0590] In this embodiment of the invention, a video acquisition device, a generative computer model, and an audio output device work together to construct a system that effectively converts visual information into audio information and provides it.

[0591] First, the user initiates a meeting or discussion using online communication technology. The device captures the video of the meeting and acquires image data. This image data consists of frame-by-frame data, including the facial expressions and gestures of the participants, and is sent to the server.

[0592] The server inputs the received image data into a generative computer model. This model has been pre-trained on a large amount of data on facial expressions and gestures, and analyzes human emotional and gesture information with high accuracy. Through this analysis, emotional labels such as "happy" or "confused" and gesture labels such as "nodding" are obtained.

[0593] Next, the server converts these analysis results into natural language and generates audio data using speech synthesis technology. This audio data is structured to be easily understood by the user.

[0594] The device receives the generated audio data and plays it back in real time through the user's audio output device, thereby transmitting visual information in the form of text and audio. This makes it easier for the user to grasp the emotions and intentions conveyed by the visual information, enabling smoother communication.

[0595] As a concrete example, consider a scenario where a user is a presenter during a meeting, and one of the participants shows a confused expression. A video capture device captures this expression in real time, and a server analyzes it as "confused." Based on this result, the user receives audio feedback saying, "The other person is confused, please consider providing additional explanations." This feedback allows the user to immediately adjust their response.

[0596] This system facilitates the understanding of nonverbal communication even when direct visual information is unavailable, supporting smoother conversations, especially for users with visual impairments.

[0597] The following describes the processing flow.

[0598] Step 1:

[0599] The user initiates a meeting via online communication technology. This sets up the device to activate its camera and begin capturing video.

[0600] Step 2:

[0601] The device acquires video data from the camera in real time. The acquired video data is temporarily stored in memory as individual frames. The video data is also converted to an appropriate format (e.g., JPEG, PNG).

[0602] Step 3:

[0603] The device sends the captured image data to the server. This transmission is performed via a protocol that enables efficient and low-latency communication (e.g., WebSocket).

[0604] Step 4:

[0605] The server inputs the received image data into a generative computer model. The model analyzes the image frames, classifies facial expressions and gestures, and labels them as emotional and gesture information.

[0606] Step 5:

[0607] The server converts the analyzed emotion and gesture information into natural language text. Here, the information is expressed in language that is intuitively easy for the user to understand.

[0608] Step 6:

[0609] The server passes natural language text to a speech synthesis engine, which generates speech data. The speech data is then adjusted to facilitate user comprehension.

[0610] Step 7:

[0611] The server sends the generated audio data back to the terminal. To minimize communication delays during this process, an appropriate protocol is used.

[0612] Step 8:

[0613] The device plays back audio data through an audio output device. This allows the user to receive analyzed facial and gesture information as audio feedback.

[0614] Step 9:

[0615] The conversation flow is adjusted based on the user's voice feedback. For example, if the emotion label is "confused," additional explanations or follow-ups are considered.

[0616] (Example 1)

[0617] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0618] There is a need to effectively convey nonverbal communication to users who cannot directly access visual information. Furthermore, in online meetings and discussions, it is difficult to grasp participants' emotions and intentions in real time, posing a barrier to smooth communication. Conventional technologies have been unable to adequately support user understanding by relying solely on audio information to supplement this visual information.

[0619] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0620] In this invention, the server includes means for acquiring video information using digital information transmission technology with an image acquisition device, means for analyzing the received video information using a generative data processing model and outputting emotion labels and action labels, and means for providing the analysis results to the user as audio information using a voice generation device. This enables the user to grasp the emotions and intentions of participants in real time during online meetings and conversations, and to communicate smoothly.

[0621] An "image acquisition device" is a device used to acquire visual information using digital information transmission technology.

[0622] A "generative data processing model" is a machine learning-based model used to analyze received video information and output emotion labels and action labels.

[0623] A "voice generation device" is a device that converts analysis results into voice information and provides it to the user.

[0624] "Visual information" refers to visual data expressed in digital format, including images and video data acquired during online meetings and discussions.

[0625] An "emotional label" is an identifier extracted from a person's facial expressions and gestures that indicates a specific emotional state.

[0626] A "movement label" is an identifier extracted to represent human movement or gesture.

[0627] "Audio information" refers to data that expresses text data or analysis results as audio.

[0628] This invention constructs a system for converting visual information into audio information, and is primarily implemented using an image acquisition device, a generative data processing model, and a speech generation device. These elements work together to provide visual information as audio in real time, thereby supporting users who cannot directly acquire visual information.

[0629] Users initiate a meeting using online communication technology. The terminal captures video of the meeting using connected image acquisition devices such as cameras. The video information includes facial expressions and gestures of the participants frame by frame, and this is sent to the server.

[0630] The server inputs the received video information into a generative data processing model. This model is based on a pre-trained dataset and extracts emotion and behavior labels with high accuracy. Specifically, the model utilizes machine learning and deep learning technologies.

[0631] The obtained emotion and action labels are converted into natural language by a speech generator on the server. This converts the emotion and action labels into voice messages that are understandable to the user.

[0632] The device receives the audio data and plays it back in real time through an audio output device. Visual information is presented to the user as audio information via speakers or headphones.

[0633] As a concrete example, consider a situation in a meeting where the presenter cannot visually confirm the confused expressions of the participants. The system extracts the confused label and delivers an audio message to the presenter saying, "The participant is confused; please consider providing additional explanations." In this way, the presenter can immediately adjust their response based on the participants' reactions.

[0634] An example of a prompt might be: "Describe a system that analyzes the facial expressions of meeting participants and provides audio feedback on emotions such as confusion and joy." This would provide guidance for understanding and implementing the system.

[0635] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0636] Step 1:

[0637] The user initiates a meeting using online communication technology. This prepares the system for capturing real-time video of meeting participants. The device uses its connected camera to acquire video information in real time. It receives the meeting video feed as input and generates frame-by-frame image data as output.

[0638] Step 2:

[0639] The terminal sends the image data for each frame it acquires to the server. The frame data includes the person's facial expressions and movements. Specifically, the terminal packets the image data over the network and sends it to the server. The input here is the image data for each frame, and the output is the image packets that arrive at the server.

[0640] Step 3:

[0641] The server inputs the received image data into a generative data processing model. This model has been pre-trained on various facial expressions and actions, and analyzes the image data to output emotion labels and action labels. Specifically, the model processes the pixel information of each frame to extract features and generates labels based on them. The input is the image data received by the server, and the output is emotion labels and action labels.

[0642] Step 4:

[0643] The server performs a process of converting the generated emotion and behavior labels into natural language text. Next, it uses speech synthesis technology to convert the text into speech data. Specifically, it converts emotion labels into text and applies a speech synthesis algorithm to generate a speech file. The input is emotion and behavior labels, and the output is natural language speech data.

[0644] Step 5:

[0645] The terminal receives audio data from the server and provides real-time feedback to the user through an audio playback device. Specifically, the terminal decodes the audio data and plays it back through a speaker or headphones. The input is audio data from the server, and the output is the real-time audio message heard by the user.

[0646] (Application Example 1)

[0647] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0648] People with visual impairments face challenges in communicating smoothly with others within their families. This problem stems from their difficulty in understanding nonverbal information such as facial expressions and gestures. Therefore, it is necessary to reduce communication barriers by converting visual information into auditory information and effectively communicating it to people with visual impairments.

[0649] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0650] In this invention, the server includes means for acquiring visual data using communication technology, means for processing the received visual data with a generative computer program and outputting emotion data and motion data, and means for transmitting the processing results to the user as audio using an audio output device. This makes it easier for visually impaired users to understand the emotions and gestures of others, and enables smooth communication within the home.

[0651] A "video acquisition device" is a device that acquires visual data using communication technology.

[0652] "Communication technology" refers to technologies for exchanging data over long distances.

[0653] "Visual data" refers to a collection of information that is represented as images or videos.

[0654] A "generative computer program" is a computer program that processes visual data and generates emotional and behavioral data.

[0655] "Emotional data" refers to information about human emotions that is analyzed from visual data.

[0656] "Motion data" refers to information about human movement that is analyzed from visual data.

[0657] An "audio output device" is a device used to transmit audio to a user.

[0658] A "user" is a person who receives support through this system.

[0659] "Visual impairment" refers to a condition in which a person has difficulty recognizing visual information.

[0660] "Within the home" refers to the living space centered around the residence.

[0661] This invention is a communication support system for visually impaired users that utilizes communication technology. This system operates in conjunction with a video acquisition device, a generative computer program, and an audio output device.

[0662] The server receives visual data acquired in real time by video acquisition equipment. This visual data includes the speaker's facial expressions and gestures. This data is sent to a generative computer program on the server and analyzed into emotion data and motion data. The generative computer uses a pre-trained dataset, enabling rapid and highly accurate facial expression recognition and gesture classification.

[0663] The server converts the generated emotion and behavior data into natural language and then uses speech synthesis software to create audio data. This audio data is then provided to visually impaired users within their homes via audio output devices.

[0664] As a concrete example, when interacting with family members, the robot's camera captures the other person's visual information and conveys the change in their facial expression to the user as an audio message such as, "The other person is smiling." This compensates for the lack of visual information and facilitates smoother communication.

[0665] An example of a prompt for facial expression analysis using a generative AI model is as follows: "Analyze the facial expression of the person captured by the camera and describe their emotion in voice."

[0666] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0667] Step 1:

[0668] The terminal captures visual data in real time using video acquisition equipment. This data includes the facial expressions and gestures of the person being spoken to. The input is video data from the camera, and the output is visual data sent to the server. The terminal sends this data to the server for the next processing step.

[0669] Step 2:

[0670] The server inputs the received visual data into a generative AI model. The generative AI model analyzes emotion data and behavior data using a pre-trained dataset. The input is visual data, and the output is the analyzed emotion data and behavior data. Based on this data processing, the server decomposes the visual information into emotion information and gesture information.

[0671] Step 3:

[0672] The server converts the analyzed emotion and behavior data into natural language. This conversion process prepares the information to be conveyed to the user into easily understandable language. The input is emotion and behavior data, and the output is the language expression as text data. The server then prepares this text for conversion into speech.

[0673] Step 4:

[0674] The server converts text data into speech data using speech synthesis technology. Specifically, it uses speech synthesis software to construct natural-sounding speech. The input is text data converted into natural language, and the output is the generated speech data. The server sends this speech data to the terminal.

[0675] Step 5:

[0676] The device transmits received audio data to the user using an audio output device. This allows the user to understand emotions and actions based on the other person's facial expressions and gestures through audio. The input is audio data, and the output is the audio the user hears. The device plays the audio at the appropriate time to support communication.

[0677] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0678] In embodiments of the present invention, a video acquisition device, a generative computer model, an audio output device, and an emotion engine are combined to construct a system that analyzes the user's nonverbal information in real time and provides it to the user through audio feedback.

[0679] First, users initiate a meeting or conversation using online communication technology. The device uses its camera to capture video of this conversation in real time, acquiring it as image data. This image data is then sent to a server for detailed analysis.

[0680] On the server, a generative computing model analyzes the received image data. This model is pre-trained on a large amount of facial expression and gesture data, and can extract emotional and gesture information from the data. Next, an emotion engine refines the results of the generative computing model, generating more detailed and accurate emotional information. This engine tracks and analyzes multiple emotional states in real time, providing personalized and optimized feedback.

[0681] The server then converts the analyzed emotion and gesture information into natural language text and generates easily understandable audio data through a speech synthesis engine. This audio data is sent to the terminal and provided to the user in real time through an audio output device.

[0682] As a concrete example, imagine a user presenting in a meeting who observes that one of the participants is showing a confused expression and restless gestures. The server's emotion engine then analyzes this information with high accuracy and generates and provides the user with audio feedback stating, "The participant appears confused and may need further explanation." This allows the user to immediately adjust their response and be considerate of the participant.

[0683] This system plays a role in facilitating the understanding of visual and nonverbal information, especially in situations where visual information is difficult to obtain, and supporting smooth and effective communication with users.

[0684] The following describes the processing flow.

[0685] Step 1:

[0686] The user initiates a meeting via online communication software. This enables the device to activate its camera and prepare for video capture.

[0687] Step 2:

[0688] The device continuously captures video of the conversation using its camera, collecting image data in real time. This data is then converted to an appropriate format and temporarily stored.

[0689] Step 3:

[0690] The device processes the collected image data and sends it to the server for analysis. The transmission process is optimized to maintain stable communication.

[0691] Step 4:

[0692] The server analyzes image data using a generative computing model. Based on a pre-trained dataset, the model detects facial expressions and gestures and extracts emotional and gesture information.

[0693] Step 5:

[0694] The server applies an emotion engine to the extracted emotion information. This engine performs a more detailed emotion analysis, evaluates multiple emotional states in real time, and tracks dynamic changes in emotions.

[0695] Step 6:

[0696] The server converts the detailed information obtained by the emotion engine into natural language and generates audio data using a speech synthesis engine. This audio data is then formatted to be easily understood by the user.

[0697] Step 7:

[0698] The server generates audio data, which is then sent to the terminal and played back to the user in real time via an audio output device.

[0699] Step 8:

[0700] Users receive voice feedback, which helps to adjust the meeting's progress and the flow of conversation. For example, if a participant seems confused, additional explanations can be provided to improve communication.

[0701] (Example 2)

[0702] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0703] In online meetings and discussions, accurately understanding participants' emotions and actions, and facilitating smooth and effective communication, presents a challenge. In particular, analyzing nonverbal information is complex, and there is a need for technology that provides real-time feedback to users.

[0704] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0705] In this invention, the server includes means for acquiring image data in real time using online communication technology, means for analyzing emotional information and behavioral information from the image data using generative computing technology, and means for generating natural language text based on the analyzed emotional information and behavioral information. This makes it possible to efficiently and accurately analyze complex non-verbal information and provide real-time feedback to the user.

[0706] "Online communication technology" refers to technology that uses the internet or networks to send and receive information in real time.

[0707] "Image data" refers to data that represents visual information acquired using devices such as cameras in a digital format.

[0708] "Generative computing technology" refers to techniques that use artificial intelligence and machine learning models to generate and analyze new information from input data.

[0709] "Emotional information" refers to information that identifies and classifies an individual's emotional state, indicating their psychological state as inferred from facial expressions, body movements, and other factors.

[0710] "Motion information" refers to information obtained by analyzing a person's physical movements, such as gestures and body language.

[0711] "Natural language text" refers to text written in a language that humans use in everyday life, and is a machine-generated, easily interpretable linguistic expression.

[0712] "Speech generation technology" is a technology that converts text data into speech and synthesizes it to sound like a human voice.

[0713] "Providing in real time" means delivering information and data to users instantly and without delay.

[0714] This invention provides an effective communication support system using online communication. Specific embodiments for carrying out the invention are described below.

[0715] Users utilize this system, for example, in online meetings or conversations. The terminal is equipped with a camera, allowing for real-time capture of the user's facial expressions and gestures. This video data is then transmitted from the terminal to the server.

[0716] The server analyzes the received video data using a generative AI model. This generative AI model is pre-trained on a large dataset collected in the past and is designed to extract emotional and motion information by performing facial expression classification and motion analysis.

[0717] Next, the server uses this information to generate natural language text. This text is converted into audio data via a speech output device and sent to the terminal. This allows the user to receive audio feedback in real time.

[0718] For example, if a participant shows a confused expression while a user is giving a presentation, the generative AI model analyzes that emotion and provides the user with audio feedback such as, "It appears the participant has some questions about the explanation." In this way, the user can immediately adjust the content of the presentation and achieve smoother communication.

[0719] An example of a prompt message is, "Show the user how to get real-time emotional feedback," which can elicit a response from the system.

[0720] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0721] Step 1:

[0722] The user starts an online meeting, and the device acquires video data in real time via its camera. The input here is the video of the user and participants during the meeting, and the output is the acquisition of video data as frames. Specifically, the camera is activated and consecutive frames are captured.

[0723] Step 2:

[0724] The terminal compresses the acquired video data and sends it to the server using a transmission protocol. The input is frame data captured in real time, and the output is compressed video data. Specifically, a data compression algorithm is applied to efficiently transfer the data to the server.

[0725] Step 3:

[0726] The server receives the transmitted video data and analyzes it using a generative AI model. The input for analysis is compressed video data, and the output is extracted emotion and motion information. Specifically, the AI ​​model evaluates each frame and identifies facial expressions and movements.

[0727] Step 4:

[0728] The server uses an emotion engine to generate detailed natural language text based on the results of the generative AI model. The input consists of emotion and behavior information, and the output is a natural language text representation. Specifically, a text generation algorithm tailored to the emotional state is executed.

[0729] Step 5:

[0730] The server converts the generated natural language text into audio data using speech generation technology and sends it to the terminal. The input is the generated natural language text, and the output is the synthesized audio data. The specific operation includes generating an audio file using a text-to-speech engine.

[0731] Step 6:

[0732] The terminal provides the user with the received audio data through an audio output device. The input is synthesized audio data, and the output is real-time audio feedback. Specifically, the audio file is decoded and played back through the speaker.

[0733] (Application Example 2)

[0734] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0735] In modern elderly care settings, understanding the emotions and psychological state of residents is crucial. However, it is difficult for busy care staff to respond immediately to the emotional changes of each resident. Furthermore, nonverbal information is easily overlooked using traditional methods, requiring greater ingenuity to improve resident satisfaction and a sense of security. Therefore, there is a need to analyze and report residents' emotional states in real time, reducing the workload of care staff while providing meticulous care.

[0736] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0737] In this invention, the server includes means for acquiring image data using a video acquisition device, means for analyzing the image data using a generative computer model to obtain emotional information, and means for converting emotional feedback into speech using a speech synthesis device and providing it to the user. This enables the acquisition of nonverbal emotional information of residents in real time in care settings, allowing care staff to respond quickly and appropriately to changes in their emotions.

[0738] A "video acquisition device" is a device that acquires image data in real time using online communication technology.

[0739] A "generative computer model" is a computer model that analyzes received image data based on a pre-trained dataset and extracts and outputs emotional information and gesture information.

[0740] An "emotion engine" is an engine that refines the analysis results of generative computer models and generates personalized emotional feedback.

[0741] A "speech synthesis device" is a device that converts analyzed emotional feedback into speech data and provides it to the user in an easily understandable format.

[0742] A "care support device" is a device that includes a system that provides voice data from a speech synthesis device to care staff and reports the emotional state of residents in real time.

[0743] "Emotional feedback" is feedback information that uses non-verbal information acquired to convey the emotional state of residents to care staff and others via voice.

[0744] The system that realizes this application example includes a video acquisition device, a generative computer model, an emotion engine, a speech synthesizer, and a care support device. The server uses real-time image data obtained via the video acquisition device and utilizes the generative computer model to analyze the data and extract nonverbal emotion and gesture information. Because this analysis uses a pre-trained dataset, highly accurate emotion classification is possible.

[0745] The server further refines the results of the generative computer model using an emotion engine to generate personalized, detailed emotional feedback, which is then converted into speech data by a speech synthesizer. This speech data is sent to a care support device and provided to care staff in real time.

[0746] As a concrete example, if a resident suddenly shows signs of restlessness, the server's emotion engine analyzes the information and delivers an audio feedback message to the care staff via the care support device stating, "The resident may be feeling anxious." This allows the staff to respond quickly and reassure the resident.

[0747] An example of a prompt message is, "Design an app that analyzes the facial expressions and movements of residents and provides voice feedback on their emotions."

[0748] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0749] Step 1:

[0750] The server receives image data acquired in real time from the video acquisition device via online communication technology. This input image data includes non-verbal information, such as the residents' facial expressions and gestures. The server then transmits this data directly to a generative computer model.

[0751] Step 2:

[0752] The server analyzes the received image data using a generative computing model. This process identifies facial expressions and other nonverbal cues based on the input image data, and extracts emotional and gesture information. The output is data indicating the analyzed emotional state.

[0753] Step 3:

[0754] The server uses an emotion engine to refine the results of the generative computer model's analysis. In this step, it generates more detailed and personalized emotion feedback based on the analyzed data. The input is the results of the generative computer model's analysis, and the output is the refined emotion feedback data.

[0755] Step 4:

[0756] The server uses a speech synthesizer to convert emotional feedback into audio data. This process synthesizes the input into an easily understandable audio format and converts it in real time. The output becomes audio data usable by care support devices.

[0757] Step 5:

[0758] The terminal receives audio data from the server and provides real-time audio feedback to care staff through the care support device. In this step, information about the resident's emotions is immediately transmitted to the care staff, and information is output to enable them to take appropriate action.

[0759] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0760] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0761] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0762] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0763] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0764] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0765] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0766] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0767] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0768] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0769] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0770] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0771] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0772] 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.

[0773] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0774] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0775] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0776] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0777] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0778] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0779] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0780] The following is further disclosed regarding the embodiments described above.

[0781] (Claim 1)

[0782] A means of acquiring image data using online communication technology with a video acquisition device,

[0783] A means for analyzing the received image data using a generative computing model and outputting emotion information and gesture information,

[0784] A means of providing the analysis results to the user as audio using an audio output device,

[0785] A system that includes this.

[0786] (Claim 2)

[0787] The system according to claim 1, wherein the video acquisition device acquires the image data in real time.

[0788] (Claim 3)

[0789] The system according to claim 1, wherein the generative computer model performs facial expression classification based on a pre-trained dataset.

[0790] "Example 1"

[0791] (Claim 1)

[0792] A means of acquiring video information using an image acquisition device and digital information transmission technology,

[0793] A means for analyzing the received video information using a generative data processing model and outputting emotion labels and action labels,

[0794] A means of providing the analysis results to the user as audio information using a voice generation device,

[0795] A means for processing video information acquired using a video acquisition device on a frame-by-frame basis,

[0796] A means for converting emotion labels obtained by a generative data processing model into text data using speech synthesis technology,

[0797] A means of presenting audio information generated in real time using an audio playback device to the user,

[0798] A system that includes this.

[0799] (Claim 2)

[0800] The system according to claim 1, characterized in that the image acquisition device acquires the video information in real time, and the video information includes facial expression information of each participant.

[0801] (Claim 3)

[0802] The system according to claim 1, characterized in that the generative data processing model performs behavioral classification based on a pre-trained data set and outputs emotion labels and behavioral labels.

[0803] "Application Example 1"

[0804] (Claim 1)

[0805] A means of acquiring visual data using communication technology with video acquisition equipment,

[0806] A means for processing the received visual data using a generative computer program and outputting emotion data and action data,

[0807] A means for transmitting the processing result to the user as audio using an audio output device,

[0808] A means of using voice to enable a robot to assist a visually impaired user in interacting with others within the home,

[0809] A system that includes this.

[0810] (Claim 2)

[0811] The system according to claim 1, wherein the video acquisition device acquires the visual data in real time.

[0812] (Claim 3)

[0813] The system according to claim 1, wherein the generative computer program performs facial expression recognition based on a pre-learned set of data.

[0814] "Example 2 of combining an emotion engine"

[0815] (Claim 1)

[0816] A device that acquires image data in real time using online communication technology,

[0817] A device for analyzing emotional information and motion information from the image data using generative computing technology,

[0818] A device that generates natural language text based on analyzed emotional and behavioral information,

[0819] A device that uses speech generation technology to convert the aforementioned natural language text into speech data,

[0820] A device that provides the aforementioned audio data to the user in real time,

[0821] A system that includes this.

[0822] (Claim 2)

[0823] The system according to claim 1, wherein the generative computation technique performs facial expression analysis based on a trained dataset.

[0824] (Claim 3)

[0825] The system according to claim 1, which tracks the aforementioned real-time emotional and behavioral information and provides personalized feedback to the user.

[0826] "Application example 2 when combining with an emotional engine"

[0827] (Claim 1)

[0828] A means of acquiring image data using online communication technology with a video acquisition device,

[0829] A means for analyzing the received image data using a generative computing model and outputting emotion information and gesture information,

[0830] A means of refining analysis results using an emotion engine and generating personalized emotional feedback,

[0831] A means for converting the emotional feedback into audio data using a speech synthesis device and providing it to the user,

[0832] A means for providing the aforementioned voice data to a care support device and reporting the user's emotional state in real time,

[0833] A system that includes this.

[0834] (Claim 2)

[0835] The system according to claim 1, wherein the video acquisition device acquires the image data in real time and provides immediate feedback to the care support device.

[0836] (Claim 3)

[0837] The system according to claim 1, wherein the generative computer model performs facial expression classification based on a pre-trained dataset and generates voice feedback appropriate to the care situation. [Explanation of Symbols]

[0838] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of acquiring visual data using communication technology with video acquisition equipment, A means for processing the received visual data using a generative computer program and outputting emotion data and action data, A means for transmitting the processing result to the user as audio using an audio output device, A means of using voice to enable a robot to assist a visually impaired user in interacting with others within the home, A system that includes this.

2. The system according to claim 1, wherein the video acquisition device acquires the visual data in real time.

3. The system according to claim 1, wherein the generative computer program performs facial expression recognition based on a pre-learned set of data.

Images