system

Abstract

We provide the system. [Solution] A means of acquiring video data of a person, A means for analyzing the aforementioned video data to recognize facial expressions and body gestures, A means for converting the feedback generated based on the aforementioned recognition result into audio data and outputting it, A system that includes this.

Description

Technical Field

[0005] , ,

[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] In remote meetings and online conversations, there is a problem that it is difficult for people with physical visual constraints to understand the emotions and intentions of their counterparts. Therefore, it is difficult to grasp non-verbal information during conversations, which is a factor reducing the quality of communication. <00000,26>

Means for Solving the Problems

[0005] To solve this problem, the present invention provides a system that processes video data acquired using a camera by a specific analysis device and automatically recognizes facial expressions and body gestures. By presenting this recognized non-verbal information as an audio message in real time, it adopts a means that enables people with visual constraints to grasp the emotional state of their conversation partners. <,

[0006] "Person" refers to an individual participating in a specific remote conversation or meeting, and taking part in or both of the conversation.

[0007] "Video data" refers to visual information, including a person's face and body movements, captured through a camera.

[0008] "Analysis" refers to the process of recognizing specific patterns or features based on acquired video data and extracting related information.

[0009] "Facial expression" refers to the visual characteristics formed by the movement of facial muscles to indicate a person's emotions or intentions.

[0010] "Body gestures" refer to movements of the hands, arms, upper body, etc., that a person makes to complement conversation or communication.

[0011] "Recognition" refers to the act of identifying and understanding information from data acquired through a specific algorithm.

[0012] "Feedback" refers to the information that a system generates and provides to a user based on input.

[0013] "Audio data" refers to information expressed in a format that can be transmitted through voice, and is usually composed of digital audio format.

[0014] "Output" refers to the act of a system sending its processing results to a location or device where they are needed. [Brief explanation of the drawing]

[0015] [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] This is a conceptual diagram showing an example of the essential 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] 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

[0016] 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.

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

[0018] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0019] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0020] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] 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."

[0023] [First Embodiment]

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

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0030] 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.

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

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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".

[0036] This invention provides a system that enables individuals to understand the emotions and intentions of other conversation participants by analyzing nonverbal conversational information acquired through video and providing audio feedback. The system consists of a camera and a communication module mounted on a specific terminal, and operates by transmitting the data to a server via the internet. The server analyzes the received video data and recognizes facial expressions and body gestures.

[0037] First, the device acquires video of the conversation participants via its camera. The device's software processes this video and then sends the data to the server. The server processes this data using a dedicated analysis model to detect facial features and body movements. The analysis provides information related to the conversation participants' facial expressions and gestures, and the server generates appropriate feedback based on this. This feedback is converted into speech using natural language processing technology and sent back to the device.

[0038] One example is when a user participates in a remote meeting via a device. The device's camera captures video of other participants in the meeting and sends this data to a server in real time. The server analyzes nonverbal signals from the video, such as smiles and nods, and generates audio feedback indicating that "the participant is interested in the presentation." This feedback is provided to the user from the device, making it easier for the user to understand the flow and atmosphere of the conversation.

[0039] Thus, the present invention provides a means to improve the quality of communication in remote meetings even when there are visual constraints, and enables support for more effective dialogue.

[0040] The following describes the processing flow.

[0041] Step 1:

[0042] The device activates its camera for the meeting and acquires video data of the person in front of it in real time. The device encodes the video stream into a digital format and processes it into a format suitable for analysis.

[0043] Step 2:

[0044] The device transmits the acquired video data to the server via the internet using a secure protocol. The data is encrypted during transmission, ensuring privacy and security.

[0045] Step 3:

[0046] The server sends the received video data to an analysis process, which is then input into a specially trained AI model. This model automatically recognizes facial expressions and body gestures and generates analysis results.

[0047] Step 4:

[0048] The server extracts information indicating emotions and intentions from the analysis results and generates feedback based on that information. For example, it might convert it into a message such as, "The person you're talking to is understanding."

[0049] Step 5:

[0050] The server generates feedback, which is then converted into an audio format using natural language processing technology, creating audio data in a way that is easy for the user to understand.

[0051] Step 6:

[0052] The server sends the audio data back to the terminal. The terminal receives the audio data and provides audio feedback to the user through speakers or earphones.

[0053] Step 7:

[0054] The user listens to voice feedback, which helps them understand the other person's emotions and intentions while continuing to communicate.

[0055] (Example 1)

[0056] 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."

[0057] Conventional communication systems have struggled to adequately grasp nonverbal expressions (facial expressions and gestures) through visual information during remote meetings and online interactions. This has made it difficult for participants in remote locations to accurately understand each other's emotions and intentions, sometimes hindering smooth communication. There is a need for a solution that addresses these challenges and enables smooth communication by efficiently analyzing visual information and providing feedback.

[0058] 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.

[0059] In this invention, the server includes means for acquiring visual information, means for analyzing the visual information and recognizing features, and means for converting the response generated based on the recognition results into audio information and outputting it. This makes it possible to understand the nonverbal expressions of remote participants in real time and provide feedback generated based on them.

[0060] "Visual information" refers to images and video data acquired through cameras and other imaging devices.

[0061] "Features" refer to identifiable information extracted from visual information, such as facial expressions and body gestures.

[0062] "Response" refers to information or feedback generated based on recognized features.

[0063] "Audio information" refers to data obtained by converting text or other information into speech.

[0064] "Nonverbal expression" refers to physical expressions such as facial expressions and gestures that accompany verbal communication.

[0065] "Feedback" refers to a response or reaction provided based on perceived information.

[0066] This invention is a system for recognizing a person's nonverbal expressions remotely and providing voice feedback based on these expressions. Specifically, it involves the coordinated operation of three elements: a terminal, a server, and a user.

[0067] First, the device uses a camera to acquire the user's visual information in real time. The device is equipped with image processing capabilities and is responsible for compressing the acquired video data and sending it to the server. As a specific example of use, the device is equipped with a high-resolution camera and a secure communication chipset, and compresses and transmits data using the H.264 video codec.

[0068] Next, the server uses an artificial intelligence model to analyze the received visual information. The analysis engine installed on the server applies deep learning technology to recognize facial features and gestures. Specific examples include machine learning frameworks such as TENSORFLOW® and PyTorch. Based on the analysis results, the server generates a response using natural language processing technology and converts it into audio information. At this stage, the server uses a speech synthesis API to convert the text into audio data.

[0069] Finally, the user receives audio feedback transmitted from the server through their device. This feedback helps the user understand the emotions and intentions of other remote participants. The device plays the audio data through a speaker or headphones.

[0070] As a concrete example, this system can be applied to remote meetings. For instance, the system could recognize when a participant smiles during a presentation and generate audio feedback such as, "The participant is interested in the presentation."

[0071] An example of a prompt message would be: "Explain how to analyze emotions from camera footage in real time and generate the results as audio feedback."

[0072] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0073] Step 1:

[0074] The device uses its camera to acquire visual information. The input is real-time video data, and the output is this data converted into an appropriate format. Specifically, the device's camera captures video at 60 frames per second and compresses the data using the H.264 video codec.

[0075] Step 2:

[0076] The terminal sends the acquired video data to the server. The input is compressed video data from the terminal, and the output is the video stream that reaches the server via the internet. Specifically, the terminal sends data using the HTTP or WebSocket protocol and performs efficient packet management to minimize latency.

[0077] Step 3:

[0078] The server analyzes the visual information it receives. The input is a video stream sent from the terminal, and the output is analysis data showing facial expressions and body gestures. Specifically, the server runs a deep learning model using TensorFlow or PyTorch and uses the OpenCV library to detect facial landmarks.

[0079] Step 4:

[0080] The server generates a response based on the analysis results. The input is feature data obtained from the analysis, and the output is a text message indicating feedback. Specifically, the server uses natural language processing techniques to generate text that expresses emotion, such as "I'm interested."

[0081] Step 5:

[0082] The server generates a text message, converts it into audio information, and sends it to the terminal. The input is the generated text message, and the output is audio data. Specifically, a speech synthesis API is used to convert the text into audio, and the data is sent back to the terminal.

[0083] Step 6:

[0084] The device plays and provides audio feedback to the user. The input is audio data sent from the server, and the output is audio that the user can hear. Specifically, the device outputs audio through speakers or earphones, delivering feedback to the user in real time.

[0085] (Application Example 1)

[0086] 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."

[0087] The problem that this invention aims to solve is to construct a means of effectively acquiring information in real time to understand the emotions and intentions of those receiving care and to provide appropriate care to caregivers in the field of elderly care, and to provide this information to caregivers.

[0088] 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.

[0089] In this invention, the server includes a device for acquiring video data of a person, a function for analyzing the video data to recognize facial expressions and body gestures, and a function for converting feedback generated based on the recognition results into audio data and outputting it. This makes it possible for caregivers to acquire information in real time to understand the emotions and intentions of the person being cared for and to provide appropriate care.

[0090] A "device for acquiring video data of a person" is a device that uses cameras, sensors, etc., to acquire video information of a person receiving care in a care environment in real time.

[0091] The "function to recognize facial expressions and body gestures" is a function that analyzes and recognizes facial and body movements that indicate the emotions and intentions of the person being cared for, based on acquired video data.

[0092] The "function to convert feedback into audio data and output it" is a function that converts the information generated based on the recognition results into an audio format that can be easily understood by caregivers and outputs it.

[0093] The "function that provides caregivers with real-time information to understand the emotions and intentions of those receiving care" is a function that enables appropriate responses on-site by quickly providing caregivers with audio information indicating the condition of those receiving care.

[0094] The system for realizing this invention mainly consists of an image acquisition device, an analysis server, and a feedback provision means.

[0095] First, the terminal uses a camera as a video acquisition device to acquire video data of the person being cared for in real time. This data is transmitted to a server via the internet. The server uses a dedicated model to analyze the video data. This model recognizes facial expressions and body movements and extracts the emotions and intentions of the person being cared for. For this analysis, computer vision libraries such as OpenCV can be used.

[0096] Based on the analysis results obtained from the server, feedback is generated. This feedback is provided in audio format so that the user can understand it immediately. The TextToSpeech library is used to generate the audio data. The generated feedback is provided to caregivers in real time via the terminal, helping them to immediately grasp the condition of the person being cared for. This enables caregivers to respond more appropriately.

[0097] As a concrete example, consider a scenario where a caregiver wears smart glasses in a care facility. The smart glasses' camera captures the care recipient's facial expressions, and this data is analyzed. The caregiver can then receive real-time voice feedback such as, "The care recipient is relaxed," which allows them to proceed with other tasks with peace of mind.

[0098] An example of a prompt for a generative AI model is, "Analyze the emotions from the care recipient's facial expressions and provide real-time feedback." Using this prompt clearly instructs the library or software on how it should respond.

[0099] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0100] Step 1:

[0101] The device acquires real-time video data of the person being cared for through its camera. The input is video data from the camera, and the output is a digital video file for analysis. The camera records the facial movements and gestures of the person being cared for and transmits this data to the next processing step.

[0102] Step 2:

[0103] The terminal transmits the acquired video data to the server via the internet. The input is the digital video file obtained in step 1, and the output is a data stream on the server. The video data is then prepared for secure and rapid processing on the server.

[0104] Step 3:

[0105] The server analyzes the received video data using a dedicated model to recognize facial expressions and body gestures. The input is a data stream on the server, and the output is a dataset containing the recognition results. The server performs image processing using computer vision libraries such as OpenCV to extract information about the person's facial expressions and movements.

[0106] Step 4:

[0107] The server generates feedback that reflects the emotions and intentions of the person being cared for, based on the recognition results. The input is a dataset containing the recognition results obtained in step 3, and the output is a text-based feedback message. The server generates the feedback text using a generative AI model based on the analysis results.

[0108] Step 5:

[0109] The server converts the generated feedback into audio data and sends it to the terminal. The input is the text-formatted feedback message obtained in step 4, and the output is an audio file. The server uses the TextToSpeech library to create the audio feedback.

[0110] Step 6:

[0111] The terminal provides the caregiver with audio feedback transmitted from the server. The input is the audio file obtained in step 5, and the output is real-time audio notification. The terminal plays the audio feedback to the caregiver through its speaker, informing them of the care recipient's condition.

[0112] 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.

[0113] This invention relates to a system that recognizes a person's facial expressions and gestures through video analysis during remote meetings and conversations, and further understands the user's emotional state using an emotion engine. This system is implemented using multiple components: a terminal, a server, and an emotion engine.

[0114] First, the device uses its camera to acquire video data of the conversation participants in real time. This video data serves as primary data for analyzing nonverbal signals during the conversation. The device converts the video data into an optimized digital data format and transmits it to a server via the internet.

[0115] Next, the server acquires this video data and uses a dedicated analysis engine to recognize facial expressions and body gestures. The analysis results obtained in this process become important elements necessary for generating feedback. Furthermore, the server uses an emotion engine to analyze and evaluate the user's emotional state from the video data. This emotion engine detects subtle changes in the user's facial expressions and voice tone, and evaluates the emotional state with high accuracy.

[0116] Based on the analyzed information, the server generates voice feedback that is easy for the user to understand. The feedback is dynamically adjusted to take into account not only the emotions and intentions of the person being spoken to, but also the user's own emotional state.

[0117] As a concrete example, consider a scenario where a user participates in an online meeting and gives a presentation. The device acquires video from both the user and the person they are talking to and sends it to the server. The server analyzes this information, and the emotion engine detects whether the person they are talking to is interested and whether the user is nervous. Based on this, the user receives personalized feedback in audio form, such as, "The participants are interested in your presentation; you should try to relax and speak more." In this way, the present invention provides support to improve the quality of communication.

[0118] The following describes the processing flow.

[0119] Step 1:

[0120] The device activates its camera and acquires video data of the participants in the remote meeting in real time. The video data is encoded in a digital format and prepared for immediate processing.

[0121] Step 2:

[0122] The device transmits encoded video data to the server via a secure communication channel. The data is encrypted before transmission to protect it from unauthorized access.

[0123] Step 3:

[0124] The server inputs the received video data into the analysis engine, which identifies facial expressions and body gestures. The analysis engine uses a machine learning model to recognize characteristic facial patterns and movements with high accuracy.

[0125] Step 4:

[0126] The server activates the emotion engine and evaluates the person's emotional state. The emotion engine analyzes facial expression data and voice tone to detect emotional states (e.g., tension, joy) in real time.

[0127] Step 5:

[0128] The server generates feedback for the user based on the analysis results and sentiment evaluation. For example, it adjusts the feedback considering the level of interest of the person being spoken to and the user's level of relaxation.

[0129] Step 6:

[0130] The server generates feedback, which is then converted into audio data using natural language processing. The audio feedback is presented in a format that is easy for the user to understand.

[0131] Step 7:

[0132] The server sends audio data to the terminal. The terminal decodes the received audio feedback and presents it to the user through a speaker or earphones.

[0133] Step 8:

[0134] By receiving voice feedback, users can more effectively understand the emotions and intentions of their conversation partners, as well as their own emotional state, which helps them to communicate appropriately.

[0135] (Example 2)

[0136] 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".

[0137] In remote meetings and conversations, accurately recognizing participants' emotional states and providing feedback based on them is difficult. Traditional systems often fail to fully utilize information obtained from facial expressions and voice, leading to a decline in communication quality. Therefore, there is a need to improve communication quality by understanding the emotional states of both the conversation partner and the user in real time and providing dynamically adjusted feedback.

[0138] 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.

[0139] In this invention, the server includes means for acquiring video data of a person in real time using a video acquisition device, means for converting the video data into an optimized digital data format and transmitting it through a communication network, means for analyzing the received video data and recognizing facial expressions and body movements, means for evaluating the emotional state based on the recognized expressions and movements, and means for generating feedback using generative AI technology based on the emotional evaluation, converting it into audio data, and outputting it. This makes it possible to accurately grasp the emotional states of both parties in a conversation and improve the quality of communication.

[0140] A "video acquisition device" is a device used to acquire video data of participants in remote meetings or discussions in real time.

[0141] A "digital data format" is a data format optimized for transmitting acquired video data over a network.

[0142] A "communication network" is a network infrastructure used to send and receive data between a terminal and a server.

[0143] An "analysis engine" is a program that processes video data to recognize facial expressions and body movements.

[0144] "Emotional assessment" is a process that accurately determines a participant's emotional state based on recognized facial expressions and actions.

[0145] "Generative AI technology" is an artificial intelligence technology that automatically generates feedback for users based on acquired data.

[0146] "Audio data" refers to audio data used to present the generated feedback message to the user.

[0147] This invention is designed as a system that understands the emotional state of participants in remote meetings and conversations and provides dynamic feedback based on that understanding. This system is implemented using multiple components, primarily terminals, servers, and generative AI models.

[0148] First, the device uses its built-in video acquisition device to acquire video data of the conversation participants in real time. This device can primarily utilize existing hardware technologies such as webcams and smartphone cameras.

[0149] The acquired video data is converted into a digital data format optimized by the terminal. This conversion uses data compression technology and format conversion algorithms. The terminal then transmits this converted digital data to the server via the communication network.

[0150] The server inputs the received video data into an analysis engine to recognize facial expressions and body movements. This analysis engine can be built using open-source libraries or commercial image recognition software. Next, the server applies a dedicated emotion evaluation algorithm to assess emotions based on the recognition results. This algorithm analyzes the user's facial expressions and vocal characteristics to estimate their emotional state with high accuracy.

[0151] Based on the analyzed data, the server uses generative AI technology to generate specific and appropriate feedback. This employs an advanced text generation model using natural language processing. This feedback is converted into audio data and transmitted to the terminal using streaming technology.

[0152] As a concrete example, consider a scenario where a user is giving a presentation in an online meeting. In this situation, the device acquires video data of the person it is talking to and sends it to the server. The server analyzes the data and identifies participants who are listening attentively and users who are nervous. Based on this, the generative AI model generates feedback such as, "Participants are interested in your presentation. Please relax and continue." This audio feedback is then conveyed to the user through the device.

[0153] An example of a prompt is: "Consider an appropriate feedback message for a user who is giving a presentation in a remote meeting, where the audience is interested, but the user is feeling a little nervous."

[0154] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0155] Step 1:

[0156] The device uses its built-in camera to acquire video data of conversation participants in real time. The input here is the physical image of the conversation participants, and the output is video data in digital format. The acquired video data is temporarily stored inside the device and used for subsequent processing. Specifically, the device selects the optimal frame rate and resolution for smooth video capture.

[0157] Step 2:

[0158] The terminal converts the acquired video data into a digital data format optimized for network transmission. The input is the raw digital video data acquired in step 1, and the output is compressed digital data. In this process, a specific compression algorithm is used to reduce the data size and improve transfer efficiency. Specifically, the terminal saves the compressed video to temporary storage and prepares it for transmission.

[0159] Step 3:

[0160] Subsequently, the terminal transmits the compressed video data to the server via the communication network. The input for this step is the converted compressed video data, and the output is the completion status of the data transmission to the server. During transmission, data is encrypted using a secure protocol to ensure privacy protection. Specifically, once transmission is complete, the terminal clears its information and prepares for the next data acquisition.

[0161] Step 4:

[0162] The server receives compressed video data from the communication network and inputs it into the analysis engine. The input for this step is the transmitted compressed digital video data, and the output is in an analyzable data format. The server uses the analysis engine to process the video frame by frame and recognize facial expressions and body movements. Specifically, a deep learning-based facial recognition model is used in this step.

[0163] Step 5:

[0164] Based on the analyzed results, the server evaluates the emotional state. The input for this step is the analyzed facial expression and body movement data, and the output is a digital representation of the evaluated emotional state. The emotion evaluation engine analyzes facial changes and voice tone to identify the subject's emotional state. Specifically, the server uses pattern matching technology to classify the emotional state.

[0165] Step 6:

[0166] The server utilizes a generative AI model to generate feedback based on sentiment assessment. The input for this step is sentiment state data, and the output is a text-based feedback message. The generated feedback is then processed using natural language processing techniques to make it easily understandable to the user. Specifically, the AI ​​generates appropriate output based on pre-trained prompt sentences.

[0167] Step 7:

[0168] The server converts the generated feedback into audio data and sends it to the terminal via the communication network. The input is a text-based feedback message, and the output is digital data in audio format. The message encoded as audio data is transmitted in real time using streaming technology. Specifically, the server utilizes a speech synthesis engine to achieve natural-sounding audio output.

[0169] Step 8:

[0170] The device plays the received audio data and provides feedback to the user. The input for this step is audio feedback data, and the output is audio feedback to the user. The device plays the audio in high quality and notifies the user in an easily understandable format. Specifically, the device utilizes its speaker function to play the audio clearly.

[0171] (Application Example 2)

[0172] 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."

[0173] Robots used in modern homes require natural communication with users, but existing technologies struggle to accurately assess a user's emotional state and provide dynamic responses based on that assessment. Therefore, new technologies are needed to enable users to build more natural and reassuring relationships with robots.

[0174] 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.

[0175] In this invention, the server includes means for acquiring video data of a person from a video sensor, means for analyzing the video data to recognize facial features and body movements, means for converting the response generated based on the recognition results into audio information and outputting it, an emotion evaluation engine for evaluating the user's emotional state, and means for dynamically adjusting an appropriate response based on the evaluated emotional state. This makes it possible for the robot to understand the user's emotions in the home and provide appropriate feedback in real time.

[0176] A "video sensor" is a device used to acquire video data of a target object.

[0177] "Video data of a person" refers to image information that focuses on a specific person, including facial expressions and movements.

[0178] "Analysis" refers to the process of extracting useful information from video data, and is an analytical process for recognizing facial features and body movements.

[0179] "Facial features" refer to information related to an individual's facial expressions and movements.

[0180] "Body movements" refers to information about a person's physical gestures and poses.

[0181] "Recognition results" refer to data showing a person's facial expressions and movements obtained through analysis.

[0182] A "response" is an audio feedback or message generated based on the recognition result.

[0183] "Audio information" refers to messages and notifications expressed through sound.

[0184] An "emotion evaluation engine" is a software module for accurately evaluating a user's emotional state.

[0185] "Means of dynamic adjustment" refers to methods and processes for changing response content in real time in response to the user's emotional evaluation.

[0186] To implement this invention, a system centered around a household robot is constructed as the main component. The details are described below.

[0187] The robot's hardware includes a video sensor (camera), an audio output device (speaker), and a processing unit, which serve as a platform for monitoring and interacting with the user's activities within their home. The video sensor is responsible for acquiring the user's video data in real time, and the acquired video data is sent to the local processing unit.

[0188] In the processing unit, video analysis software runs to recognize facial features and body movements from video data. An example of software used here is the Emotion Analysis SDK. This software analyzes facial features and generates recognition results. These recognition results are further analyzed by an emotion evaluation engine to assess the user's emotional state. This engine detects subtle changes in the user's facial expressions and determines the type and intensity of emotions with high accuracy.

[0189] Next, the server dynamically generates an appropriate voice response based on the output of the emotion evaluation engine. This process applies natural language processing techniques to generate a response based on the recognition results and emotional state. Speech synthesis technologies such as the Text-to-Speech API are used to generate the voice response. The generated voice is then provided to the user through a voice output device.

[0190] This process allows users to receive emotionally-driven, interactive responses in real time through their home robot. For example, when a user is visible on the video sensor, the robot can detect their fatigue and provide voice feedback such as, "You look tired. Shall we take a break?" The user can then feel more comfortable with the robot's response.

[0191] A concrete example of a prompt for using a generative AI model is, "Please tell me how to analyze what emotions a user felt during their workday based on camera footage and generate a response appropriate to those emotions." By utilizing this example prompt, it is possible to design interactions that are more suitable for the user.

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

[0193] Step 1:

[0194] The terminal acquires user video data in real time via a video sensor. This video data serves as input for processing the user's facial expressions and body movements as road and environmental information.

[0195] Step 2:

[0196] The terminal sends the acquired video data to the processing unit, which uses the Emotion Analysis SDK to analyze facial features and body movements. As a result of the analysis, facial feature data and gesture data are output.

[0197] Step 3:

[0198] The server receives the recognition results from the video analysis and uses an emotion evaluation engine to accurately assess the user's emotional state. During this process, it calculates emotion labels and their intensity based on changes in facial expressions and movements. The output includes the type and intensity of the emotion.

[0199] Step 4:

[0200] The server generates an appropriate voice response based on the emotion evaluation results obtained by the emotion evaluation engine. Natural language processing techniques are used in this process, and a generative AI model determines the content of the response. The response content is then generated and output.

[0201] Step 5:

[0202] The server uses the Text-to-Speech API to convert the generated response into speech data. The converted speech data is then output.

[0203] Step 6:

[0204] The terminal presents the user with converted audio information via an audio output device. This audio feedback allows the user to communicate interactively with the robot.

[0205] 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.

[0206] 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.

[0207] 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.

[0208] [Second Embodiment]

[0209] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0210] 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.

[0211] 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).

[0212] 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.

[0213] 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.

[0214] 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).

[0215] 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.

[0216] 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.

[0217] 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.

[0218] 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.

[0219] 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.

[0220] 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".

[0221] This invention provides a system that enables individuals to understand the emotions and intentions of other conversation participants by analyzing nonverbal conversational information acquired through video and providing audio feedback. The system consists of a camera and a communication module mounted on a specific terminal, and operates by transmitting the data to a server via the internet. The server analyzes the received video data and recognizes facial expressions and body gestures.

[0222] First, the device acquires video of the conversation participants via its camera. The device's software processes this video and then sends the data to the server. The server processes this data using a dedicated analysis model to detect facial features and body movements. The analysis provides information related to the conversation participants' facial expressions and gestures, and the server generates appropriate feedback based on this. This feedback is converted into speech using natural language processing technology and sent back to the device.

[0223] One example is when a user participates in a remote meeting via a device. The device's camera captures video of other participants in the meeting and sends this data to a server in real time. The server analyzes nonverbal signals from the video, such as smiles and nods, and generates audio feedback indicating that "the participant is interested in the presentation." This feedback is provided to the user from the device, making it easier for the user to understand the flow and atmosphere of the conversation.

[0224] Thus, the present invention provides a means to improve the quality of communication in remote meetings even when there are visual constraints, and enables support for more effective dialogue.

[0225] The following describes the processing flow.

[0226] Step 1:

[0227] The device activates its camera for the meeting and acquires video data of the person in front of it in real time. The device encodes the video stream into a digital format and processes it into a format suitable for analysis.

[0228] Step 2:

[0229] The device transmits the acquired video data to the server via the internet using a secure protocol. The data is encrypted during transmission, ensuring privacy and security.

[0230] Step 3:

[0231] The server sends the received video data to an analysis process, which is then input into a specially trained AI model. This model automatically recognizes facial expressions and body gestures and generates analysis results.

[0232] Step 4:

[0233] The server extracts information indicating emotions and intentions from the analysis results and generates feedback based on that information. For example, it might convert it into a message such as, "The person you're talking to is understanding."

[0234] Step 5:

[0235] The server generates feedback, which is then converted into an audio format using natural language processing technology, creating audio data in a way that is easy for the user to understand.

[0236] Step 6:

[0237] The server sends the audio data back to the terminal. The terminal receives the audio data and provides audio feedback to the user through speakers or earphones.

[0238] Step 7:

[0239] The user listens to voice feedback, which helps them understand the other person's emotions and intentions while continuing to communicate.

[0240] (Example 1)

[0241] 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."

[0242] Conventional communication systems have struggled to adequately grasp nonverbal expressions (facial expressions and gestures) through visual information during remote meetings and online interactions. This has made it difficult for participants in remote locations to accurately understand each other's emotions and intentions, sometimes hindering smooth communication. There is a need for a solution that addresses these challenges and enables smooth communication by efficiently analyzing visual information and providing feedback.

[0243] 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.

[0244] In this invention, the server includes means for acquiring visual information, means for analyzing the visual information and recognizing features, and means for converting the response generated based on the recognition results into audio information and outputting it. This makes it possible to understand the nonverbal expressions of remote participants in real time and provide feedback generated based on them.

[0245] "Visual information" refers to images and video data acquired through cameras and other imaging devices.

[0246] "Features" refer to identifiable information extracted from visual information, such as facial expressions and body gestures.

[0247] "Response" refers to information or feedback generated based on recognized features.

[0248] "Audio information" refers to data obtained by converting text or other information into speech.

[0249] "Nonverbal expression" refers to physical expressions such as facial expressions and gestures that accompany verbal communication.

[0250] "Feedback" refers to a response or reaction provided based on perceived information.

[0251] This invention is a system for recognizing a person's nonverbal expressions remotely and providing voice feedback based on these expressions. Specifically, it involves the coordinated operation of three elements: a terminal, a server, and a user.

[0252] First, the device uses a camera to acquire the user's visual information in real time. The device is equipped with image processing capabilities and is responsible for compressing the acquired video data and sending it to the server. As a specific example of use, the device is equipped with a high-resolution camera and a secure communication chipset, and compresses and transmits data using the H.264 video codec.

[0253] Next, the server uses an artificial intelligence model to analyze the received visual information. The analysis engine installed on the server applies deep learning technology to recognize facial features and gestures. Specifically, machine learning frameworks such as TensorFlow and PyTorch are used. Based on the analysis results, the server generates a response using natural language processing technology and converts it into audio information. At this stage, the server uses a speech synthesis API to convert the text into audio data.

[0254] Finally, the user receives audio feedback transmitted from the server through their device. This feedback helps the user understand the emotions and intentions of other remote participants. The device plays the audio data through a speaker or headphones.

[0255] As a concrete example, this system can be applied to remote meetings. For instance, the system could recognize when a participant smiles during a presentation and generate audio feedback such as, "The participant is interested in the presentation."

[0256] An example of a prompt message would be: "Explain how to analyze emotions from camera footage in real time and generate the results as audio feedback."

[0257] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0258] Step 1:

[0259] The device uses its camera to acquire visual information. The input is real-time video data, and the output is this data converted into an appropriate format. Specifically, the device's camera captures video at 60 frames per second and compresses the data using the H.264 video codec.

[0260] Step 2:

[0261] The terminal sends the acquired video data to the server. The input is compressed video data from the terminal, and the output is the video stream that reaches the server via the internet. Specifically, the terminal sends data using the HTTP or WebSocket protocol and performs efficient packet management to minimize latency.

[0262] Step 3:

[0263] The server analyzes the visual information it receives. The input is a video stream sent from the terminal, and the output is analysis data showing facial expressions and body gestures. Specifically, the server runs a deep learning model using TensorFlow or PyTorch and uses the OpenCV library to detect facial landmarks.

[0264] Step 4:

[0265] The server generates a response based on the analysis results. The input is feature data obtained from the analysis, and the output is a text message indicating feedback. Specifically, the server uses natural language processing techniques to generate text that expresses emotion, such as "I'm interested."

[0266] Step 5:

[0267] The server generates a text message, converts it into audio information, and sends it to the terminal. The input is the generated text message, and the output is audio data. Specifically, a speech synthesis API is used to convert the text into audio, and the data is sent back to the terminal.

[0268] Step 6:

[0269] The device plays and provides audio feedback to the user. The input is audio data sent from the server, and the output is audio that the user can hear. Specifically, the device outputs audio through speakers or earphones, delivering feedback to the user in real time.

[0270] (Application Example 1)

[0271] 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."

[0272] The problem that this invention aims to solve is to construct a means of effectively acquiring information in real time to understand the emotions and intentions of those receiving care and to provide appropriate care to caregivers in the field of elderly care, and to provide this information to caregivers.

[0273] 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.

[0274] In this invention, the server includes a device for acquiring video data of a person, a function for analyzing the video data to recognize facial expressions and body gestures, and a function for converting feedback generated based on the recognition results into audio data and outputting it. This makes it possible for caregivers to acquire information in real time to understand the emotions and intentions of the person being cared for and to provide appropriate care.

[0275] A "device for acquiring video data of a person" is a device that uses cameras, sensors, etc., to acquire video information of a person receiving care in a care environment in real time.

[0276] The "function to recognize facial expressions and body gestures" is a function that analyzes and recognizes facial and body movements that indicate the emotions and intentions of the person being cared for, based on acquired video data.

[0277] The "function to convert feedback into audio data and output it" is a function that converts the information generated based on the recognition results into an audio format that can be easily understood by caregivers and outputs it.

[0278] The "function that provides caregivers with real-time information to understand the emotions and intentions of those receiving care" is a function that enables appropriate responses on-site by quickly providing caregivers with audio information indicating the condition of those receiving care.

[0279] The system for realizing this invention mainly consists of an image acquisition device, an analysis server, and a feedback provision means.

[0280] First, the terminal uses a camera as a video acquisition device to acquire video data of the person being cared for in real time. This data is transmitted to a server via the internet. The server uses a dedicated model to analyze the video data. This model recognizes facial expressions and body movements and extracts the emotions and intentions of the person being cared for. For this analysis, computer vision libraries such as OpenCV can be used.

[0281] Based on the analysis results obtained by the server, feedback is generated. This feedback is provided in voice format so that users can understand it immediately. A TextToSpeech library is used to generate the voice data. The generated feedback is provided to the caregiver in real time through the terminal, which helps the caregiver immediately grasp the condition of the care recipient. As a result, the caregiver can respond more appropriately.

[0282] As a specific example, consider the case where a caregiver is wearing smart glasses in a care facility. The camera of the smart glasses captures the expression of the care recipient, and the data is analyzed. Then, voice feedback such as "The care recipient is relaxed" can be received in real time, allowing the caregiver to proceed with other tasks with peace of mind.

[0283] An example of a prompt sentence for the generation AI model is "Analyze the emotion from the expression of the care recipient and provide feedback in real time." By using this prompt sentence, it is possible to clearly instruct how the library and software should react.

[0284] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0285] Step 1:

[0286] The terminal acquires the video data of the care recipient in real time through the camera. The input is the video data from the camera, and the output is a digital video file for analysis. The camera records the movements and gestures of the care recipient's face and sends the data to the next processing step.

[0287] Step 2:

[0288] The terminal transmits the acquired video data to the server via the internet. The input is the digital video file obtained in step 1, and the output is a data stream on the server. The video data is then prepared for secure and rapid processing on the server.

[0289] Step 3:

[0290] The server analyzes the received video data using a dedicated model to recognize facial expressions and body gestures. The input is a data stream on the server, and the output is a dataset containing the recognition results. The server performs image processing using computer vision libraries such as OpenCV to extract information about the person's facial expressions and movements.

[0291] Step 4:

[0292] The server generates feedback that reflects the emotions and intentions of the person being cared for, based on the recognition results. The input is a dataset containing the recognition results obtained in step 3, and the output is a text-based feedback message. The server generates the feedback text using a generative AI model based on the analysis results.

[0293] Step 5:

[0294] The server converts the generated feedback into audio data and sends it to the terminal. The input is the text-formatted feedback message obtained in step 4, and the output is an audio file. The server uses the TextToSpeech library to create the audio feedback.

[0295] Step 6:

[0296] The terminal provides the caregiver with audio feedback transmitted from the server. The input is the audio file obtained in step 5, and the output is real-time audio notification. The terminal plays the audio feedback to the caregiver through its speaker, informing them of the care recipient's condition.

[0297] 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.

[0298] This invention relates to a system that recognizes a person's facial expressions and gestures through video analysis during remote meetings and conversations, and further understands the user's emotional state using an emotion engine. This system is implemented using multiple components: a terminal, a server, and an emotion engine.

[0299] First, the device uses its camera to acquire video data of the conversation participants in real time. This video data serves as primary data for analyzing nonverbal signals during the conversation. The device converts the video data into an optimized digital data format and transmits it to a server via the internet.

[0300] Next, the server acquires this video data and uses a dedicated analysis engine to recognize facial expressions and body gestures. The analysis results obtained in this process become important elements necessary for generating feedback. Furthermore, the server uses an emotion engine to analyze and evaluate the user's emotional state from the video data. This emotion engine detects subtle changes in the user's facial expressions and voice tone, and evaluates the emotional state with high accuracy.

[0301] Based on the analyzed information, the server generates voice feedback that is easy for the user to understand. The feedback is dynamically adjusted to take into account not only the emotions and intentions of the person being spoken to, but also the user's own emotional state.

[0302] As a specific example, a scenario where a user participates in an online meeting and gives a presentation can be considered. The terminal acquires the videos of both the conversation partner and the user and transmits them to the server. The server analyzes the information, and the emotion engine detects that the conversation partner is showing interest and that the user is nervous. Based on this, adjusted feedback such as "The participants are interested in your presentation. It would be better to talk more relaxedly." is provided to the user in voice. In this way, the present invention provides support for improving the quality of communication.

[0303] The processing flow will be described below.

[0304] Step 1:

[0305] The terminal activates the camera and acquires in real time the video data of the person participating in the remote meeting. The video data is encoded in digital format and prepared for immediate processing.

[0306] Step 2:

[0307] The terminal transmits the encoded video data to the server through a secure communication channel. The data is encrypted before transmission and protected from unauthorized access.

[0308] Step 3:

[0309] The server inputs the received video data into an analysis engine to identify facial expressions and body gestures. The analysis engine uses a machine learning model to accurately recognize characteristic expression patterns and movements.

[0310] Step 4:

[0311] The server activates an emotion engine to evaluate the emotional state of the person. The emotion engine analyzes the expression data and voice tone to detect the emotional state (e.g., nervousness, joy) in real time.

[0312] Step 5:

[0313] The server generates feedback for the user based on the analysis results and sentiment evaluation. For example, it adjusts the feedback considering the level of interest of the person being spoken to and the user's level of relaxation.

[0314] Step 6:

[0315] The server generates feedback, which is then converted into audio data using natural language processing. The audio feedback is presented in a format that is easy for the user to understand.

[0316] Step 7:

[0317] The server sends audio data to the terminal. The terminal decodes the received audio feedback and presents it to the user through a speaker or earphones.

[0318] Step 8:

[0319] By receiving voice feedback, users can more effectively understand the emotions and intentions of their conversation partners, as well as their own emotional state, which helps them to communicate appropriately.

[0320] (Example 2)

[0321] 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".

[0322] In remote meetings and conversations, accurately recognizing participants' emotional states and providing feedback based on them is difficult. Traditional systems often fail to fully utilize information obtained from facial expressions and voice, leading to a decline in communication quality. Therefore, there is a need to improve communication quality by understanding the emotional states of both the conversation partner and the user in real time and providing dynamically adjusted feedback.

[0323] 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.

[0324] In this invention, the server includes means for acquiring video data of a person in real time using a video acquisition device, means for converting the video data into an optimized digital data format and transmitting it through a communication network, means for analyzing the received video data and recognizing facial expressions and body movements, means for evaluating the emotional state based on the recognized expressions and movements, and means for generating feedback using generative AI technology based on the emotional evaluation, converting it into audio data, and outputting it. This makes it possible to accurately grasp the emotional states of both parties in a conversation and improve the quality of communication.

[0325] A "video acquisition device" is a device used to acquire video data of participants in remote meetings or discussions in real time.

[0326] A "digital data format" is a data format optimized for transmitting acquired video data over a network.

[0327] A "communication network" is a network infrastructure used to send and receive data between a terminal and a server.

[0328] An "analysis engine" is a program that processes video data to recognize facial expressions and body movements.

[0329] "Emotional assessment" is a process that accurately determines a participant's emotional state based on recognized facial expressions and actions.

[0330] "Generative AI technology" is an artificial intelligence technology that automatically generates feedback for users based on acquired data.

[0331] "Audio data" refers to audio data used to present the generated feedback message to the user.

[0332] This invention is designed as a system that understands the emotional state of participants in remote meetings and conversations and provides dynamic feedback based on that understanding. This system is implemented using multiple components, primarily terminals, servers, and generative AI models.

[0333] First, the device uses its built-in video acquisition device to acquire video data of the conversation participants in real time. This device can primarily utilize existing hardware technologies such as webcams and smartphone cameras.

[0334] The acquired video data is converted into a digital data format optimized by the terminal. This conversion uses data compression technology and format conversion algorithms. The terminal then transmits this converted digital data to the server via the communication network.

[0335] The server inputs the received video data into an analysis engine to recognize facial expressions and body movements. This analysis engine can be built using open-source libraries or commercial image recognition software. Next, the server applies a dedicated emotion evaluation algorithm to assess emotions based on the recognition results. This algorithm analyzes the user's facial expressions and vocal characteristics to estimate their emotional state with high accuracy.

[0336] Based on the analyzed data, the server uses generative AI technology to generate specific and appropriate feedback. This employs an advanced text generation model using natural language processing. This feedback is converted into audio data and transmitted to the terminal using streaming technology.

[0337] As a concrete example, consider a scenario where a user is giving a presentation in an online meeting. In this situation, the device acquires video data of the person it is talking to and sends it to the server. The server analyzes the data and identifies participants who are listening attentively and users who are nervous. Based on this, the generative AI model generates feedback such as, "Participants are interested in your presentation. Please relax and continue." This audio feedback is then conveyed to the user through the device.

[0338] An example of a prompt is: "Consider an appropriate feedback message for a user who is giving a presentation in a remote meeting, where the audience is interested, but the user is feeling a little nervous."

[0339] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0340] Step 1:

[0341] The device uses its built-in camera to acquire video data of conversation participants in real time. The input here is the physical image of the conversation participants, and the output is video data in digital format. The acquired video data is temporarily stored inside the device and used for subsequent processing. Specifically, the device selects the optimal frame rate and resolution for smooth video capture.

[0342] Step 2:

[0343] The terminal converts the acquired video data into a digital data format optimized for network transmission. The input is the raw digital video data acquired in step 1, and the output is compressed digital data. In this process, a specific compression algorithm is used to reduce the data size and improve transfer efficiency. Specifically, the terminal saves the compressed video to temporary storage and prepares it for transmission.

[0344] Step 3:

[0345] Subsequently, the terminal transmits the compressed video data to the server via the communication network. The input for this step is the converted compressed video data, and the output is the completion status of the data transmission to the server. During transmission, data is encrypted using a secure protocol to ensure privacy protection. Specifically, once transmission is complete, the terminal clears its information and prepares for the next data acquisition.

[0346] Step 4:

[0347] The server receives compressed video data from the communication network and inputs it into the analysis engine. The input for this step is the transmitted compressed digital video data, and the output is in an analyzable data format. The server uses the analysis engine to process the video frame by frame and recognize facial expressions and body movements. Specifically, a deep learning-based facial recognition model is used in this step.

[0348] Step 5:

[0349] Based on the analyzed results, the server evaluates the emotional state. The input for this step is the analyzed facial expression and body movement data, and the output is a digital representation of the evaluated emotional state. The emotion evaluation engine analyzes facial changes and voice tone to identify the subject's emotional state. Specifically, the server uses pattern matching technology to classify the emotional state.

[0350] Step 6:

[0351] The server utilizes a generative AI model to generate feedback based on sentiment assessment. The input for this step is sentiment state data, and the output is a text-based feedback message. The generated feedback is then processed using natural language processing techniques to make it easily understandable to the user. Specifically, the AI ​​generates appropriate output based on pre-trained prompt sentences.

[0352] Step 7:

[0353] The server converts the generated feedback into audio data and sends it to the terminal via the communication network. The input is a text-based feedback message, and the output is digital data in audio format. The message encoded as audio data is transmitted in real time using streaming technology. Specifically, the server utilizes a speech synthesis engine to achieve natural-sounding audio output.

[0354] Step 8:

[0355] The device plays the received audio data and provides feedback to the user. The input for this step is audio feedback data, and the output is audio feedback to the user. The device plays the audio in high quality and notifies the user in an easily understandable format. Specifically, the device utilizes its speaker function to play the audio clearly.

[0356] (Application Example 2)

[0357] 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".

[0358] Robots used in modern homes require natural communication with users, but existing technologies struggle to accurately assess a user's emotional state and provide dynamic responses based on that assessment. Therefore, new technologies are needed to enable users to build more natural and reassuring relationships with robots.

[0359] 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.

[0360] In this invention, the server includes means for acquiring video data of a person from a video sensor, means for analyzing the video data to recognize facial features and body movements, means for converting the response generated based on the recognition results into audio information and outputting it, an emotion evaluation engine for evaluating the user's emotional state, and means for dynamically adjusting an appropriate response based on the evaluated emotional state. This makes it possible for the robot to understand the user's emotions in the home and provide appropriate feedback in real time.

[0361] A "video sensor" is a device used to acquire video data of a target object.

[0362] "Video data of a person" refers to image information that focuses on a specific person, including facial expressions and movements.

[0363] "Analysis" refers to the process of extracting useful information from video data, and is an analytical process for recognizing facial features and body movements.

[0364] "Facial features" refer to information related to an individual's facial expressions and movements.

[0365] "Body movements" refers to information about a person's physical gestures and poses.

[0366] "Recognition results" refer to data showing a person's facial expressions and movements obtained through analysis.

[0367] A "response" is an audio feedback or message generated based on the recognition result.

[0368] "Audio information" refers to messages and notifications expressed through sound.

[0369] An "emotion evaluation engine" is a software module for accurately evaluating a user's emotional state.

[0370] "Means of dynamic adjustment" refers to methods and processes for changing response content in real time in response to the user's emotional evaluation.

[0371] To implement this invention, a system centered around a household robot is constructed as the main component. The details are described below.

[0372] The robot's hardware includes a video sensor (camera), an audio output device (speaker), and a processing unit, which serve as a platform for monitoring and interacting with the user's activities within their home. The video sensor is responsible for acquiring the user's video data in real time, and the acquired video data is sent to the local processing unit.

[0373] In the processing unit, video analysis software runs to recognize facial features and body movements from video data. An example of software used here is the Emotion Analysis SDK. This software analyzes facial features and generates recognition results. These recognition results are further analyzed by an emotion evaluation engine to assess the user's emotional state. This engine detects subtle changes in the user's facial expressions and determines the type and intensity of emotions with high accuracy.

[0374] Next, the server dynamically generates an appropriate voice response based on the output of the emotion evaluation engine. This process applies natural language processing techniques to generate a response based on the recognition results and emotional state. Speech synthesis technologies such as the Text-to-Speech API are used to generate the voice response. The generated voice is then provided to the user through a voice output device.

[0375] This process allows users to receive emotionally-driven, interactive responses in real time through their home robot. For example, when a user is visible on the video sensor, the robot can detect their fatigue and provide voice feedback such as, "You look tired. Shall we take a break?" The user can then feel more comfortable with the robot's response.

[0376] A concrete example of a prompt for using a generative AI model is, "Please tell me how to analyze what emotions a user felt during their workday based on camera footage and generate a response appropriate to those emotions." By utilizing this example prompt, it is possible to design interactions that are more suitable for the user.

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

[0378] Step 1:

[0379] The terminal acquires user video data in real time via a video sensor. This video data serves as input for processing the user's facial expressions and body movements as road and environmental information.

[0380] Step 2:

[0381] The terminal sends the acquired video data to the processing unit, which uses the Emotion Analysis SDK to analyze facial features and body movements. As a result of the analysis, facial feature data and gesture data are output.

[0382] Step 3:

[0383] The server receives the recognition results from the video analysis and uses an emotion evaluation engine to accurately assess the user's emotional state. During this process, it calculates emotion labels and their intensity based on changes in facial expressions and movements. The output includes the type and intensity of the emotion.

[0384] Step 4:

[0385] The server generates an appropriate voice response based on the emotion evaluation results obtained by the emotion evaluation engine. Natural language processing techniques are used in this process, and a generative AI model determines the content of the response. The response content is then generated and output.

[0386] Step 5:

[0387] The server uses the Text-to-Speech API to convert the generated response into speech data. The converted speech data is then output.

[0388] Step 6:

[0389] The terminal presents the user with converted audio information via an audio output device. This audio feedback allows the user to communicate interactively with the robot.

[0390] 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.

[0391] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One 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.

[0392] 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.

[0393] [Third Embodiment]

[0394] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0395] 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.

[0396] 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).

[0397] 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.

[0398] 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.

[0399] 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).

[0400] 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.

[0401] 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.

[0402] 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.

[0403] 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.

[0404] 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.

[0405] 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".

[0406] This invention provides a system that enables individuals to understand the emotions and intentions of other conversation participants by analyzing nonverbal conversational information acquired through video and providing audio feedback. The system consists of a camera and a communication module mounted on a specific terminal, and operates by transmitting the data to a server via the internet. The server analyzes the received video data and recognizes facial expressions and body gestures.

[0407] First, the device acquires video of the conversation participants via its camera. The device's software processes this video and then sends the data to the server. The server processes this data using a dedicated analysis model to detect facial features and body movements. The analysis provides information related to the conversation participants' facial expressions and gestures, and the server generates appropriate feedback based on this. This feedback is converted into speech using natural language processing technology and sent back to the device.

[0408] One example is when a user participates in a remote meeting via a device. The device's camera captures video of other participants in the meeting and sends this data to a server in real time. The server analyzes nonverbal signals from the video, such as smiles and nods, and generates audio feedback indicating that "the participant is interested in the presentation." This feedback is provided to the user from the device, making it easier for the user to understand the flow and atmosphere of the conversation.

[0409] Thus, the present invention provides a means to improve the quality of communication in remote meetings even when there are visual constraints, and enables support for more effective dialogue.

[0410] The following describes the processing flow.

[0411] Step 1:

[0412] The device activates its camera for the meeting and acquires video data of the person in front of it in real time. The device encodes the video stream into a digital format and processes it into a format suitable for analysis.

[0413] Step 2:

[0414] The device transmits the acquired video data to the server via the internet using a secure protocol. The data is encrypted during transmission, ensuring privacy and security.

[0415] Step 3:

[0416] The server sends the received video data to an analysis process, which is then input into a specially trained AI model. This model automatically recognizes facial expressions and body gestures and generates analysis results.

[0417] Step 4:

[0418] The server extracts information indicating emotions and intentions from the analysis results and generates feedback based on that information. For example, it might convert it into a message such as, "The person you're talking to is understanding."

[0419] Step 5:

[0420] The server generates feedback, which is then converted into an audio format using natural language processing technology, creating audio data in a way that is easy for the user to understand.

[0421] Step 6:

[0422] The server sends the audio data back to the terminal. The terminal receives the audio data and provides audio feedback to the user through speakers or earphones.

[0423] Step 7:

[0424] The user listens to voice feedback, which helps them understand the other person's emotions and intentions while continuing to communicate.

[0425] (Example 1)

[0426] 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."

[0427] Conventional communication systems have struggled to adequately grasp nonverbal expressions (facial expressions and gestures) through visual information during remote meetings and online interactions. This has made it difficult for participants in remote locations to accurately understand each other's emotions and intentions, sometimes hindering smooth communication. There is a need for a solution that addresses these challenges and enables smooth communication by efficiently analyzing visual information and providing feedback.

[0428] 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.

[0429] In this invention, the server includes means for acquiring visual information, means for analyzing the visual information and recognizing features, and means for converting the response generated based on the recognition results into audio information and outputting it. This makes it possible to understand the nonverbal expressions of remote participants in real time and provide feedback generated based on them.

[0430] "Visual information" refers to images and video data acquired through cameras and other imaging devices.

[0431] "Features" refer to identifiable information extracted from visual information, such as facial expressions and body gestures.

[0432] "Response" refers to information or feedback generated based on recognized features.

[0433] "Audio information" refers to data obtained by converting text or other information into speech.

[0434] "Nonverbal expression" refers to physical expressions such as facial expressions and gestures that accompany verbal communication.

[0435] "Feedback" refers to a response or reaction provided based on perceived information.

[0436] This invention is a system for recognizing a person's nonverbal expressions remotely and providing voice feedback based on these expressions. Specifically, it involves the coordinated operation of three elements: a terminal, a server, and a user.

[0437] First, the device uses a camera to acquire the user's visual information in real time. The device is equipped with image processing capabilities and is responsible for compressing the acquired video data and sending it to the server. As a specific example of use, the device is equipped with a high-resolution camera and a secure communication chipset, and compresses and transmits data using the H.264 video codec.

[0438] Next, the server uses an artificial intelligence model to analyze the received visual information. The analysis engine installed on the server applies deep learning technology to recognize facial features and gestures. Specifically, machine learning frameworks such as TensorFlow and PyTorch are used. Based on the analysis results, the server generates a response using natural language processing technology and converts it into audio information. At this stage, the server uses a speech synthesis API to convert the text into audio data.

[0439] Finally, the user receives audio feedback transmitted from the server through their device. This feedback helps the user understand the emotions and intentions of other remote participants. The device plays the audio data through a speaker or headphones.

[0440] As a concrete example, this system can be applied to remote meetings. For instance, the system could recognize when a participant smiles during a presentation and generate audio feedback such as, "The participant is interested in the presentation."

[0441] An example of a prompt message would be: "Explain how to analyze emotions from camera footage in real time and generate the results as audio feedback."

[0442] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0443] Step 1:

[0444] The device uses its camera to acquire visual information. The input is real-time video data, and the output is this data converted into an appropriate format. Specifically, the device's camera captures video at 60 frames per second and compresses the data using the H.264 video codec.

[0445] Step 2:

[0446] The terminal sends the acquired video data to the server. The input is compressed video data from the terminal, and the output is the video stream that reaches the server via the internet. Specifically, the terminal sends data using the HTTP or WebSocket protocol and performs efficient packet management to minimize latency.

[0447] Step 3:

[0448] The server analyzes the visual information it receives. The input is a video stream sent from the terminal, and the output is analysis data showing facial expressions and body gestures. Specifically, the server runs a deep learning model using TensorFlow or PyTorch and uses the OpenCV library to detect facial landmarks.

[0449] Step 4:

[0450] The server generates a response based on the analysis results. The input is feature data obtained from the analysis, and the output is a text message indicating feedback. Specifically, the server uses natural language processing techniques to generate text that expresses emotion, such as "I'm interested."

[0451] Step 5:

[0452] The server generates a text message, converts it into audio information, and sends it to the terminal. The input is the generated text message, and the output is audio data. Specifically, a speech synthesis API is used to convert the text into audio, and the data is sent back to the terminal.

[0453] Step 6:

[0454] The device plays and provides audio feedback to the user. The input is audio data sent from the server, and the output is audio that the user can hear. Specifically, the device outputs audio through speakers or earphones, delivering feedback to the user in real time.

[0455] (Application Example 1)

[0456] 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."

[0457] The problem that this invention aims to solve is to construct a means of effectively acquiring information in real time to understand the emotions and intentions of those receiving care and to provide appropriate care to caregivers in the field of elderly care, and to provide this information to caregivers.

[0458] 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.

[0459] In this invention, the server includes a device for acquiring video data of a person, a function for analyzing the video data to recognize facial expressions and body gestures, and a function for converting feedback generated based on the recognition results into audio data and outputting it. This makes it possible for caregivers to acquire information in real time to understand the emotions and intentions of the person being cared for and to provide appropriate care.

[0460] A "device for acquiring video data of a person" is a device that uses cameras, sensors, etc., to acquire video information of a person receiving care in a care environment in real time.

[0461] The "function to recognize facial expressions and body gestures" is a function that analyzes and recognizes facial and body movements that indicate the emotions and intentions of the person being cared for, based on acquired video data.

[0462] The "function to convert feedback into audio data and output it" is a function that converts the information generated based on the recognition results into an audio format that can be easily understood by caregivers and outputs it.

[0463] The "function that provides caregivers with real-time information to understand the emotions and intentions of those receiving care" is a function that enables appropriate responses on-site by quickly providing caregivers with audio information indicating the condition of those receiving care.

[0464] The system for realizing this invention mainly consists of an image acquisition device, an analysis server, and a feedback provision means.

[0465] First, the terminal uses a camera as a video acquisition device to acquire video data of the person being cared for in real time. This data is transmitted to a server via the internet. The server uses a dedicated model to analyze the video data. This model recognizes facial expressions and body movements and extracts the emotions and intentions of the person being cared for. For this analysis, computer vision libraries such as OpenCV can be used.

[0466] Based on the analysis results obtained from the server, feedback is generated. This feedback is provided in audio format so that the user can understand it immediately. The TextToSpeech library is used to generate the audio data. The generated feedback is provided to caregivers in real time via the terminal, helping them to immediately grasp the condition of the person being cared for. This enables caregivers to respond more appropriately.

[0467] As a concrete example, consider a scenario where a caregiver wears smart glasses in a care facility. The smart glasses' camera captures the care recipient's facial expressions, and this data is analyzed. The caregiver can then receive real-time voice feedback such as, "The care recipient is relaxed," which allows them to proceed with other tasks with peace of mind.

[0468] An example of a prompt for a generative AI model is, "Analyze the emotions from the care recipient's facial expressions and provide real-time feedback." Using this prompt clearly instructs the library or software on how it should respond.

[0469] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0470] Step 1:

[0471] The device acquires real-time video data of the person being cared for through its camera. The input is video data from the camera, and the output is a digital video file for analysis. The camera records the facial movements and gestures of the person being cared for and transmits this data to the next processing step.

[0472] Step 2:

[0473] The terminal transmits the acquired video data to the server via the internet. The input is the digital video file obtained in step 1, and the output is a data stream on the server. The video data is then prepared for secure and rapid processing on the server.

[0474] Step 3:

[0475] The server analyzes the received video data using a dedicated model to recognize facial expressions and body gestures. The input is a data stream on the server, and the output is a dataset containing the recognition results. The server performs image processing using computer vision libraries such as OpenCV to extract information about the person's facial expressions and movements.

[0476] Step 4:

[0477] The server generates feedback that reflects the emotions and intentions of the person being cared for, based on the recognition results. The input is a dataset containing the recognition results obtained in step 3, and the output is a text-based feedback message. The server generates the feedback text using a generative AI model based on the analysis results.

[0478] Step 5:

[0479] The server converts the generated feedback into audio data and sends it to the terminal. The input is the text-formatted feedback message obtained in step 4, and the output is an audio file. The server uses the TextToSpeech library to create the audio feedback.

[0480] Step 6:

[0481] The terminal provides the caregiver with audio feedback transmitted from the server. The input is the audio file obtained in step 5, and the output is real-time audio notification. The terminal plays the audio feedback to the caregiver through its speaker, informing them of the care recipient's condition.

[0482] 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.

[0483] This invention relates to a system that recognizes a person's facial expressions and gestures through video analysis during remote meetings and conversations, and further understands the user's emotional state using an emotion engine. This system is implemented using multiple components: a terminal, a server, and an emotion engine.

[0484] First, the device uses its camera to acquire video data of the conversation participants in real time. This video data serves as primary data for analyzing nonverbal signals during the conversation. The device converts the video data into an optimized digital data format and transmits it to a server via the internet.

[0485] Next, the server acquires this video data and uses a dedicated analysis engine to recognize facial expressions and body gestures. The analysis results obtained in this process become important elements necessary for generating feedback. Furthermore, the server uses an emotion engine to analyze and evaluate the user's emotional state from the video data. This emotion engine detects subtle changes in the user's facial expressions and voice tone, and evaluates the emotional state with high accuracy.

[0486] Based on the analyzed information, the server generates voice feedback that is easy for the user to understand. The feedback is dynamically adjusted to take into account not only the emotions and intentions of the person being spoken to, but also the user's own emotional state.

[0487] As a concrete example, consider a scenario where a user participates in an online meeting and gives a presentation. The device acquires video from both the user and the person they are talking to and sends it to the server. The server analyzes this information, and the emotion engine detects whether the person they are talking to is interested and whether the user is nervous. Based on this, the user receives personalized feedback in audio form, such as, "The participants are interested in your presentation; you should try to relax and speak more." In this way, the present invention provides support to improve the quality of communication.

[0488] The following describes the processing flow.

[0489] Step 1:

[0490] The device activates its camera and acquires video data of the participants in the remote meeting in real time. The video data is encoded in a digital format and prepared for immediate processing.

[0491] Step 2:

[0492] The device transmits encoded video data to the server via a secure communication channel. The data is encrypted before transmission to protect it from unauthorized access.

[0493] Step 3:

[0494] The server inputs the received video data into the analysis engine, which identifies facial expressions and body gestures. The analysis engine uses a machine learning model to recognize characteristic facial patterns and movements with high accuracy.

[0495] Step 4:

[0496] The server activates the emotion engine and evaluates the person's emotional state. The emotion engine analyzes facial expression data and voice tone to detect emotional states (e.g., tension, joy) in real time.

[0497] Step 5:

[0498] The server generates feedback for the user based on the analysis results and sentiment evaluation. For example, it adjusts the feedback considering the level of interest of the person being spoken to and the user's level of relaxation.

[0499] Step 6:

[0500] The server generates feedback, which is then converted into audio data using natural language processing. The audio feedback is presented in a format that is easy for the user to understand.

[0501] Step 7:

[0502] The server sends audio data to the terminal. The terminal decodes the received audio feedback and presents it to the user through a speaker or earphones.

[0503] Step 8:

[0504] By receiving voice feedback, users can more effectively understand the emotions and intentions of their conversation partners, as well as their own emotional state, which helps them to communicate appropriately.

[0505] (Example 2)

[0506] 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."

[0507] In remote meetings and conversations, accurately recognizing participants' emotional states and providing feedback based on them is difficult. Traditional systems often fail to fully utilize information obtained from facial expressions and voice, leading to a decline in communication quality. Therefore, there is a need to improve communication quality by understanding the emotional states of both the conversation partner and the user in real time and providing dynamically adjusted feedback.

[0508] 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.

[0509] In this invention, the server includes means for acquiring video data of a person in real time using a video acquisition device, means for converting the video data into an optimized digital data format and transmitting it through a communication network, means for analyzing the received video data and recognizing facial expressions and body movements, means for evaluating the emotional state based on the recognized expressions and movements, and means for generating feedback using generative AI technology based on the emotional evaluation, converting it into audio data, and outputting it. This makes it possible to accurately grasp the emotional states of both parties in a conversation and improve the quality of communication.

[0510] A "video acquisition device" is a device used to acquire video data of participants in remote meetings or discussions in real time.

[0511] A "digital data format" is a data format optimized for transmitting acquired video data over a network.

[0512] A "communication network" is a network infrastructure used to send and receive data between a terminal and a server.

[0513] An "analysis engine" is a program that processes video data to recognize facial expressions and body movements.

[0514] "Emotional assessment" is a process that accurately determines a participant's emotional state based on recognized facial expressions and actions.

[0515] "Generative AI technology" is an artificial intelligence technology that automatically generates feedback for users based on acquired data.

[0516] "Audio data" refers to audio data used to present the generated feedback message to the user.

[0517] This invention is designed as a system that understands the emotional state of participants in remote meetings and conversations and provides dynamic feedback based on that understanding. This system is implemented using multiple components, primarily terminals, servers, and generative AI models.

[0518] First, the device uses its built-in video acquisition device to acquire video data of the conversation participants in real time. This device can primarily utilize existing hardware technologies such as webcams and smartphone cameras.

[0519] The acquired video data is converted into a digital data format optimized by the terminal. This conversion uses data compression technology and format conversion algorithms. The terminal then transmits this converted digital data to the server via the communication network.

[0520] The server inputs the received video data into an analysis engine to recognize facial expressions and body movements. This analysis engine can be built using open-source libraries or commercial image recognition software. Next, the server applies a dedicated emotion evaluation algorithm to assess emotions based on the recognition results. This algorithm analyzes the user's facial expressions and vocal characteristics to estimate their emotional state with high accuracy.

[0521] Based on the analyzed data, the server uses generative AI technology to generate specific and appropriate feedback. This employs an advanced text generation model using natural language processing. This feedback is converted into audio data and transmitted to the terminal using streaming technology.

[0522] As a concrete example, consider a scenario where a user is giving a presentation in an online meeting. In this situation, the device acquires video data of the person it is talking to and sends it to the server. The server analyzes the data and identifies participants who are listening attentively and users who are nervous. Based on this, the generative AI model generates feedback such as, "Participants are interested in your presentation. Please relax and continue." This audio feedback is then conveyed to the user through the device.

[0523] An example of a prompt is: "Consider an appropriate feedback message for a user who is giving a presentation in a remote meeting, where the audience is interested, but the user is feeling a little nervous."

[0524] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0525] Step 1:

[0526] The device uses its built-in camera to acquire video data of conversation participants in real time. The input here is the physical image of the conversation participants, and the output is video data in digital format. The acquired video data is temporarily stored inside the device and used for subsequent processing. Specifically, the device selects the optimal frame rate and resolution for smooth video capture.

[0527] Step 2:

[0528] The terminal converts the acquired video data into a digital data format optimized for network transmission. The input is the raw digital video data acquired in step 1, and the output is compressed digital data. In this process, a specific compression algorithm is used to reduce the data size and improve transfer efficiency. Specifically, the terminal saves the compressed video to temporary storage and prepares it for transmission.

[0529] Step 3:

[0530] Subsequently, the terminal transmits the compressed video data to the server via the communication network. The input for this step is the converted compressed video data, and the output is the completion status of the data transmission to the server. During transmission, data is encrypted using a secure protocol to ensure privacy protection. Specifically, once transmission is complete, the terminal clears its information and prepares for the next data acquisition.

[0531] Step 4:

[0532] The server receives compressed video data from the communication network and inputs it into the analysis engine. The input for this step is the transmitted compressed digital video data, and the output is in an analyzable data format. The server uses the analysis engine to process the video frame by frame and recognize facial expressions and body movements. Specifically, a deep learning-based facial recognition model is used in this step.

[0533] Step 5:

[0534] Based on the analyzed results, the server evaluates the emotional state. The input for this step is the analyzed facial expression and body movement data, and the output is a digital representation of the evaluated emotional state. The emotion evaluation engine analyzes facial changes and voice tone to identify the subject's emotional state. Specifically, the server uses pattern matching technology to classify the emotional state.

[0535] Step 6:

[0536] The server utilizes a generative AI model to generate feedback based on sentiment assessment. The input for this step is sentiment state data, and the output is a text-based feedback message. The generated feedback is then processed using natural language processing techniques to make it easily understandable to the user. Specifically, the AI ​​generates appropriate output based on pre-trained prompt sentences.

[0537] Step 7:

[0538] The server converts the generated feedback into audio data and sends it to the terminal via the communication network. The input is a text-based feedback message, and the output is digital data in audio format. The message encoded as audio data is transmitted in real time using streaming technology. Specifically, the server utilizes a speech synthesis engine to achieve natural-sounding audio output.

[0539] Step 8:

[0540] The device plays the received audio data and provides feedback to the user. The input for this step is audio feedback data, and the output is audio feedback to the user. The device plays the audio in high quality and notifies the user in an easily understandable format. Specifically, the device utilizes its speaker function to play the audio clearly.

[0541] (Application Example 2)

[0542] 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."

[0543] Robots used in modern homes require natural communication with users, but existing technologies struggle to accurately assess a user's emotional state and provide dynamic responses based on that assessment. Therefore, new technologies are needed to enable users to build more natural and reassuring relationships with robots.

[0544] 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.

[0545] In this invention, the server includes means for acquiring video data of a person from a video sensor, means for analyzing the video data to recognize facial features and body movements, means for converting the response generated based on the recognition results into audio information and outputting it, an emotion evaluation engine for evaluating the user's emotional state, and means for dynamically adjusting an appropriate response based on the evaluated emotional state. This makes it possible for the robot to understand the user's emotions in the home and provide appropriate feedback in real time.

[0546] A "video sensor" is a device used to acquire video data of a target object.

[0547] "Video data of a person" refers to image information that focuses on a specific person, including facial expressions and movements.

[0548] "Analysis" refers to the process of extracting useful information from video data, and is an analytical process for recognizing facial features and body movements.

[0549] "Facial features" refer to information related to an individual's facial expressions and movements.

[0550] "Body movements" refers to information about a person's physical gestures and poses.

[0551] "Recognition results" refer to data showing a person's facial expressions and movements obtained through analysis.

[0552] A "response" is an audio feedback or message generated based on the recognition result.

[0553] "Audio information" refers to messages and notifications expressed through sound.

[0554] An "emotion evaluation engine" is a software module for accurately evaluating a user's emotional state.

[0555] "Means of dynamic adjustment" refers to methods and processes for changing response content in real time in response to the user's emotional evaluation.

[0556] To implement this invention, a system centered around a household robot is constructed as the main component. The details are described below.

[0557] The robot's hardware includes a video sensor (camera), an audio output device (speaker), and a processing unit, which serve as a platform for monitoring and interacting with the user's activities within their home. The video sensor is responsible for acquiring the user's video data in real time, and the acquired video data is sent to the local processing unit.

[0558] In the processing unit, video analysis software runs to recognize facial features and body movements from video data. An example of software used here is the Emotion Analysis SDK. This software analyzes facial features and generates recognition results. These recognition results are further analyzed by an emotion evaluation engine to assess the user's emotional state. This engine detects subtle changes in the user's facial expressions and determines the type and intensity of emotions with high accuracy.

[0559] Next, the server dynamically generates an appropriate voice response based on the output of the emotion evaluation engine. This process applies natural language processing techniques to generate a response based on the recognition results and emotional state. Speech synthesis technologies such as the Text-to-Speech API are used to generate the voice response. The generated voice is then provided to the user through a voice output device.

[0560] This process allows users to receive emotionally-driven, interactive responses in real time through their home robot. For example, when a user is visible on the video sensor, the robot can detect their fatigue and provide voice feedback such as, "You look tired. Shall we take a break?" The user can then feel more comfortable with the robot's response.

[0561] A concrete example of a prompt for using a generative AI model is, "Please tell me how to analyze what emotions a user felt during their workday based on camera footage and generate a response appropriate to those emotions." By utilizing this example prompt, it is possible to design interactions that are more suitable for the user.

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

[0563] Step 1:

[0564] The terminal acquires user video data in real time via a video sensor. This video data serves as input for processing the user's facial expressions and body movements as road and environmental information.

[0565] Step 2:

[0566] The terminal sends the acquired video data to the processing unit, which uses the Emotion Analysis SDK to analyze facial features and body movements. As a result of the analysis, facial feature data and gesture data are output.

[0567] Step 3:

[0568] The server receives the recognition results from the video analysis and uses an emotion evaluation engine to accurately assess the user's emotional state. During this process, it calculates emotion labels and their intensity based on changes in facial expressions and movements. The output includes the type and intensity of the emotion.

[0569] Step 4:

[0570] The server generates an appropriate voice response based on the emotion evaluation results obtained by the emotion evaluation engine. Natural language processing techniques are used in this process, and a generative AI model determines the content of the response. The response content is then generated and output.

[0571] Step 5:

[0572] The server uses the Text-to-Speech API to convert the generated response into speech data. The converted speech data is then output.

[0573] Step 6:

[0574] The terminal presents the user with converted audio information via an audio output device. This audio feedback allows the user to communicate interactively with the robot.

[0575] 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.

[0576] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One 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.

[0577] 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.

[0578] [Fourth Embodiment]

[0579] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0580] 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.

[0581] 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).

[0582] 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.

[0583] 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.

[0584] 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).

[0585] 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.

[0586] 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.

[0587] 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.

[0588] 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.

[0589] 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.

[0590] 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.

[0591] 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".

[0592] This invention provides a system that enables individuals to understand the emotions and intentions of other conversation participants by analyzing nonverbal conversational information acquired through video and providing audio feedback. The system consists of a camera and a communication module mounted on a specific terminal, and operates by transmitting the data to a server via the internet. The server analyzes the received video data and recognizes facial expressions and body gestures.

[0593] First, the device acquires video of the conversation participants via its camera. The device's software processes this video and then sends the data to the server. The server processes this data using a dedicated analysis model to detect facial features and body movements. The analysis provides information related to the conversation participants' facial expressions and gestures, and the server generates appropriate feedback based on this. This feedback is converted into speech using natural language processing technology and sent back to the device.

[0594] One example is when a user participates in a remote meeting via a device. The device's camera captures video of other participants in the meeting and sends this data to a server in real time. The server analyzes nonverbal signals from the video, such as smiles and nods, and generates audio feedback indicating that "the participant is interested in the presentation." This feedback is provided to the user from the device, making it easier for the user to understand the flow and atmosphere of the conversation.

[0595] Thus, the present invention provides a means to improve the quality of communication in remote meetings even when there are visual constraints, and enables support for more effective dialogue.

[0596] The following describes the processing flow.

[0597] Step 1:

[0598] The device activates its camera for the meeting and acquires video data of the person in front of it in real time. The device encodes the video stream into a digital format and processes it into a format suitable for analysis.

[0599] Step 2:

[0600] The device transmits the acquired video data to the server via the internet using a secure protocol. The data is encrypted during transmission, ensuring privacy and security.

[0601] Step 3:

[0602] The server sends the received video data to an analysis process, which is then input into a specially trained AI model. This model automatically recognizes facial expressions and body gestures and generates analysis results.

[0603] Step 4:

[0604] The server extracts information indicating emotions and intentions from the analysis results and generates feedback based on that information. For example, it might convert it into a message such as, "The person you're talking to is understanding."

[0605] Step 5:

[0606] The server generates feedback, which is then converted into an audio format using natural language processing technology, creating audio data in a way that is easy for the user to understand.

[0607] Step 6:

[0608] The server sends the audio data back to the terminal. The terminal receives the audio data and provides audio feedback to the user through speakers or earphones.

[0609] Step 7:

[0610] The user listens to voice feedback, which helps them understand the other person's emotions and intentions while continuing to communicate.

[0611] (Example 1)

[0612] 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".

[0613] Conventional communication systems have struggled to adequately grasp nonverbal expressions (facial expressions and gestures) through visual information during remote meetings and online interactions. This has made it difficult for participants in remote locations to accurately understand each other's emotions and intentions, sometimes hindering smooth communication. There is a need for a solution that addresses these challenges and enables smooth communication by efficiently analyzing visual information and providing feedback.

[0614] 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.

[0615] In this invention, the server includes means for acquiring visual information, means for analyzing the visual information and recognizing features, and means for converting the response generated based on the recognition results into audio information and outputting it. This makes it possible to understand the nonverbal expressions of remote participants in real time and provide feedback generated based on them.

[0616] "Visual information" refers to images and video data acquired through cameras and other imaging devices.

[0617] "Features" refer to identifiable information extracted from visual information, such as facial expressions and body gestures.

[0618] "Response" refers to information or feedback generated based on recognized features.

[0619] "Audio information" refers to data obtained by converting text or other information into speech.

[0620] "Nonverbal expression" refers to physical expressions such as facial expressions and gestures that accompany verbal communication.

[0621] "Feedback" refers to a response or reaction provided based on perceived information.

[0622] This invention is a system for recognizing a person's nonverbal expressions remotely and providing voice feedback based on these expressions. Specifically, it involves the coordinated operation of three elements: a terminal, a server, and a user.

[0623] First, the device uses a camera to acquire the user's visual information in real time. The device is equipped with image processing capabilities and is responsible for compressing the acquired video data and sending it to the server. As a specific example of use, the device is equipped with a high-resolution camera and a secure communication chipset, and compresses and transmits data using the H.264 video codec.

[0624] Next, the server uses an artificial intelligence model to analyze the received visual information. The analysis engine installed on the server applies deep learning technology to recognize facial features and gestures. Specifically, machine learning frameworks such as TensorFlow and PyTorch are used. Based on the analysis results, the server generates a response using natural language processing technology and converts it into audio information. At this stage, the server uses a speech synthesis API to convert the text into audio data.

[0625] Finally, the user receives audio feedback transmitted from the server through their device. This feedback helps the user understand the emotions and intentions of other remote participants. The device plays the audio data through a speaker or headphones.

[0626] As a concrete example, this system can be applied to remote meetings. For instance, the system could recognize when a participant smiles during a presentation and generate audio feedback such as, "The participant is interested in the presentation."

[0627] An example of a prompt message would be: "Explain how to analyze emotions from camera footage in real time and generate the results as audio feedback."

[0628] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0629] Step 1:

[0630] The device uses its camera to acquire visual information. The input is real-time video data, and the output is this data converted into an appropriate format. Specifically, the device's camera captures video at 60 frames per second and compresses the data using the H.264 video codec.

[0631] Step 2:

[0632] The terminal sends the acquired video data to the server. The input is compressed video data from the terminal, and the output is the video stream that reaches the server via the internet. Specifically, the terminal sends data using the HTTP or WebSocket protocol and performs efficient packet management to minimize latency.

[0633] Step 3:

[0634] The server analyzes the visual information it receives. The input is a video stream sent from the terminal, and the output is analysis data showing facial expressions and body gestures. Specifically, the server runs a deep learning model using TensorFlow or PyTorch and uses the OpenCV library to detect facial landmarks.

[0635] Step 4:

[0636] The server generates a response based on the analysis results. The input is feature data obtained from the analysis, and the output is a text message indicating feedback. Specifically, the server uses natural language processing techniques to generate text that expresses emotion, such as "I'm interested."

[0637] Step 5:

[0638] The server generates a text message, converts it into audio information, and sends it to the terminal. The input is the generated text message, and the output is audio data. Specifically, a speech synthesis API is used to convert the text into audio, and the data is sent back to the terminal.

[0639] Step 6:

[0640] The device plays and provides audio feedback to the user. The input is audio data sent from the server, and the output is audio that the user can hear. Specifically, the device outputs audio through speakers or earphones, delivering feedback to the user in real time.

[0641] (Application Example 1)

[0642] 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".

[0643] The problem that this invention aims to solve is to construct a means of effectively acquiring information in real time to understand the emotions and intentions of those receiving care and to provide appropriate care to caregivers in the field of elderly care, and to provide this information to caregivers.

[0644] 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.

[0645] In this invention, the server includes a device for acquiring video data of a person, a function for analyzing the video data to recognize facial expressions and body gestures, and a function for converting feedback generated based on the recognition results into audio data and outputting it. This makes it possible for caregivers to acquire information in real time to understand the emotions and intentions of the person being cared for and to provide appropriate care.

[0646] A "device for acquiring video data of a person" is a device that uses cameras, sensors, etc., to acquire video information of a person receiving care in a care environment in real time.

[0647] The "function to recognize facial expressions and body gestures" is a function that analyzes and recognizes facial and body movements that indicate the emotions and intentions of the person being cared for, based on acquired video data.

[0648] The "function to convert feedback into audio data and output it" is a function that converts the information generated based on the recognition results into an audio format that can be easily understood by caregivers and outputs it.

[0649] The "function that provides caregivers with real-time information to understand the emotions and intentions of those receiving care" is a function that enables appropriate responses on-site by quickly providing caregivers with audio information indicating the condition of those receiving care.

[0650] The system for realizing this invention mainly consists of an image acquisition device, an analysis server, and a feedback provision means.

[0651] First, the terminal uses a camera as a video acquisition device to acquire video data of the person being cared for in real time. This data is transmitted to a server via the internet. The server uses a dedicated model to analyze the video data. This model recognizes facial expressions and body movements and extracts the emotions and intentions of the person being cared for. For this analysis, computer vision libraries such as OpenCV can be used.

[0652] Based on the analysis results obtained from the server, feedback is generated. This feedback is provided in audio format so that the user can understand it immediately. The TextToSpeech library is used to generate the audio data. The generated feedback is provided to caregivers in real time via the terminal, helping them to immediately grasp the condition of the person being cared for. This enables caregivers to respond more appropriately.

[0653] As a concrete example, consider a scenario where a caregiver wears smart glasses in a care facility. The smart glasses' camera captures the care recipient's facial expressions, and this data is analyzed. The caregiver can then receive real-time voice feedback such as, "The care recipient is relaxed," which allows them to proceed with other tasks with peace of mind.

[0654] An example of a prompt for a generative AI model is, "Analyze the emotions from the care recipient's facial expressions and provide real-time feedback." Using this prompt clearly instructs the library or software on how it should respond.

[0655] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0656] Step 1:

[0657] The device acquires real-time video data of the person being cared for through its camera. The input is video data from the camera, and the output is a digital video file for analysis. The camera records the facial movements and gestures of the person being cared for and transmits this data to the next processing step.

[0658] Step 2:

[0659] The terminal transmits the acquired video data to the server via the internet. The input is the digital video file obtained in step 1, and the output is a data stream on the server. The video data is then prepared for secure and rapid processing on the server.

[0660] Step 3:

[0661] The server analyzes the received video data using a dedicated model to recognize facial expressions and body gestures. The input is a data stream on the server, and the output is a dataset containing the recognition results. The server performs image processing using computer vision libraries such as OpenCV to extract information about the person's facial expressions and movements.

[0662] Step 4:

[0663] The server generates feedback that reflects the emotions and intentions of the person being cared for, based on the recognition results. The input is a dataset containing the recognition results obtained in step 3, and the output is a text-based feedback message. The server generates the feedback text using a generative AI model based on the analysis results.

[0664] Step 5:

[0665] The server converts the generated feedback into audio data and sends it to the terminal. The input is the text-formatted feedback message obtained in step 4, and the output is an audio file. The server uses the TextToSpeech library to create the audio feedback.

[0666] Step 6:

[0667] The terminal provides the caregiver with audio feedback transmitted from the server. The input is the audio file obtained in step 5, and the output is real-time audio notification. The terminal plays the audio feedback to the caregiver through its speaker, informing them of the care recipient's condition.

[0668] 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.

[0669] This invention relates to a system that recognizes a person's facial expressions and gestures through video analysis during remote meetings and conversations, and further understands the user's emotional state using an emotion engine. This system is implemented using multiple components: a terminal, a server, and an emotion engine.

[0670] First, the device uses its camera to acquire video data of the conversation participants in real time. This video data serves as primary data for analyzing nonverbal signals during the conversation. The device converts the video data into an optimized digital data format and transmits it to a server via the internet.

[0671] Next, the server acquires this video data and uses a dedicated analysis engine to recognize facial expressions and body gestures. The analysis results obtained in this process become important elements necessary for generating feedback. Furthermore, the server uses an emotion engine to analyze and evaluate the user's emotional state from the video data. This emotion engine detects subtle changes in the user's facial expressions and voice tone, and evaluates the emotional state with high accuracy.

[0672] Based on the analyzed information, the server generates voice feedback that is easy for the user to understand. The feedback is dynamically adjusted to take into account not only the emotions and intentions of the person being spoken to, but also the user's own emotional state.

[0673] As a concrete example, consider a scenario where a user participates in an online meeting and gives a presentation. The device acquires video from both the user and the person they are talking to and sends it to the server. The server analyzes this information, and the emotion engine detects whether the person they are talking to is interested and whether the user is nervous. Based on this, the user receives personalized feedback in audio form, such as, "The participants are interested in your presentation; you should try to relax and speak more." In this way, the present invention provides support to improve the quality of communication.

[0674] The following describes the processing flow.

[0675] Step 1:

[0676] The device activates its camera and acquires video data of the participants in the remote meeting in real time. The video data is encoded in a digital format and prepared for immediate processing.

[0677] Step 2:

[0678] The device transmits encoded video data to the server via a secure communication channel. The data is encrypted before transmission to protect it from unauthorized access.

[0679] Step 3:

[0680] The server inputs the received video data into the analysis engine, which identifies facial expressions and body gestures. The analysis engine uses a machine learning model to recognize characteristic facial patterns and movements with high accuracy.

[0681] Step 4:

[0682] The server activates the emotion engine and evaluates the person's emotional state. The emotion engine analyzes facial expression data and voice tone to detect emotional states (e.g., tension, joy) in real time.

[0683] Step 5:

[0684] The server generates feedback for the user based on the analysis results and sentiment evaluation. For example, it adjusts the feedback considering the level of interest of the person being spoken to and the user's level of relaxation.

[0685] Step 6:

[0686] The server generates feedback, which is then converted into audio data using natural language processing. The audio feedback is presented in a format that is easy for the user to understand.

[0687] Step 7:

[0688] The server sends audio data to the terminal. The terminal decodes the received audio feedback and presents it to the user through a speaker or earphones.

[0689] Step 8:

[0690] By receiving voice feedback, users can more effectively understand the emotions and intentions of their conversation partners, as well as their own emotional state, which helps them to communicate appropriately.

[0691] (Example 2)

[0692] 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".

[0693] In remote meetings and conversations, accurately recognizing participants' emotional states and providing feedback based on them is difficult. Traditional systems often fail to fully utilize information obtained from facial expressions and voice, leading to a decline in communication quality. Therefore, there is a need to improve communication quality by understanding the emotional states of both the conversation partner and the user in real time and providing dynamically adjusted feedback.

[0694] 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.

[0695] In this invention, the server includes means for acquiring video data of a person in real time using a video acquisition device, means for converting the video data into an optimized digital data format and transmitting it through a communication network, means for analyzing the received video data and recognizing facial expressions and body movements, means for evaluating the emotional state based on the recognized expressions and movements, and means for generating feedback using generative AI technology based on the emotional evaluation, converting it into audio data, and outputting it. This makes it possible to accurately grasp the emotional states of both parties in a conversation and improve the quality of communication.

[0696] A "video acquisition device" is a device used to acquire video data of participants in remote meetings or discussions in real time.

[0697] A "digital data format" is a data format optimized for transmitting acquired video data over a network.

[0698] A "communication network" is a network infrastructure used to send and receive data between a terminal and a server.

[0699] An "analysis engine" is a program that processes video data to recognize facial expressions and body movements.

[0700] "Emotional assessment" is a process that accurately determines a participant's emotional state based on recognized facial expressions and actions.

[0701] "Generative AI technology" is an artificial intelligence technology that automatically generates feedback for users based on acquired data.

[0702] "Audio data" refers to audio data used to present the generated feedback message to the user.

[0703] This invention is designed as a system that understands the emotional state of participants in remote meetings and conversations and provides dynamic feedback based on that understanding. This system is implemented using multiple components, primarily terminals, servers, and generative AI models.

[0704] First, the device uses its built-in video acquisition device to acquire video data of the conversation participants in real time. This device can primarily utilize existing hardware technologies such as webcams and smartphone cameras.

[0705] The acquired video data is converted into a digital data format optimized by the terminal. This conversion uses data compression technology and format conversion algorithms. The terminal then transmits this converted digital data to the server via the communication network.

[0706] The server inputs the received video data into an analysis engine to recognize facial expressions and body movements. This analysis engine can be built using open-source libraries or commercial image recognition software. Next, the server applies a dedicated emotion evaluation algorithm to assess emotions based on the recognition results. This algorithm analyzes the user's facial expressions and vocal characteristics to estimate their emotional state with high accuracy.

[0707] Based on the analyzed data, the server uses generative AI technology to generate specific and appropriate feedback. This employs an advanced text generation model using natural language processing. This feedback is converted into audio data and transmitted to the terminal using streaming technology.

[0708] As a concrete example, consider a scenario where a user is giving a presentation in an online meeting. In this situation, the device acquires video data of the person it is talking to and sends it to the server. The server analyzes the data and identifies participants who are listening attentively and users who are nervous. Based on this, the generative AI model generates feedback such as, "Participants are interested in your presentation. Please relax and continue." This audio feedback is then conveyed to the user through the device.

[0709] An example of a prompt is: "Consider an appropriate feedback message for a user who is giving a presentation in a remote meeting, where the audience is interested, but the user is feeling a little nervous."

[0710] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0711] Step 1:

[0712] The device uses its built-in camera to acquire video data of conversation participants in real time. The input here is the physical image of the conversation participants, and the output is video data in digital format. The acquired video data is temporarily stored inside the device and used for subsequent processing. Specifically, the device selects the optimal frame rate and resolution for smooth video capture.

[0713] Step 2:

[0714] The terminal converts the acquired video data into a digital data format optimized for network transmission. The input is the raw digital video data acquired in step 1, and the output is compressed digital data. In this process, a specific compression algorithm is used to reduce the data size and improve transfer efficiency. Specifically, the terminal saves the compressed video to temporary storage and prepares it for transmission.

[0715] Step 3:

[0716] Subsequently, the terminal transmits the compressed video data to the server via the communication network. The input for this step is the converted compressed video data, and the output is the completion status of the data transmission to the server. During transmission, data is encrypted using a secure protocol to ensure privacy protection. Specifically, once transmission is complete, the terminal clears its information and prepares for the next data acquisition.

[0717] Step 4:

[0718] The server receives compressed video data from the communication network and inputs it into the analysis engine. The input for this step is the transmitted compressed digital video data, and the output is in an analyzable data format. The server uses the analysis engine to process the video frame by frame and recognize facial expressions and body movements. Specifically, a deep learning-based facial recognition model is used in this step.

[0719] Step 5:

[0720] Based on the analyzed results, the server evaluates the emotional state. The input for this step is the analyzed facial expression and body movement data, and the output is a digital representation of the evaluated emotional state. The emotion evaluation engine analyzes facial changes and voice tone to identify the subject's emotional state. Specifically, the server uses pattern matching technology to classify the emotional state.

[0721] Step 6:

[0722] The server utilizes a generative AI model to generate feedback based on sentiment assessment. The input for this step is sentiment state data, and the output is a text-based feedback message. The generated feedback is then processed using natural language processing techniques to make it easily understandable to the user. Specifically, the AI ​​generates appropriate output based on pre-trained prompt sentences.

[0723] Step 7:

[0724] The server converts the generated feedback into audio data and sends it to the terminal via the communication network. The input is a text-based feedback message, and the output is digital data in audio format. The message encoded as audio data is transmitted in real time using streaming technology. Specifically, the server utilizes a speech synthesis engine to achieve natural-sounding audio output.

[0725] Step 8:

[0726] The device plays the received audio data and provides feedback to the user. The input for this step is audio feedback data, and the output is audio feedback to the user. The device plays the audio in high quality and notifies the user in an easily understandable format. Specifically, the device utilizes its speaker function to play the audio clearly.

[0727] (Application Example 2)

[0728] 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".

[0729] Robots used in modern homes require natural communication with users, but existing technologies struggle to accurately assess a user's emotional state and provide dynamic responses based on that assessment. Therefore, new technologies are needed to enable users to build more natural and reassuring relationships with robots.

[0730] 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.

[0731] In this invention, the server includes means for acquiring video data of a person from a video sensor, means for analyzing the video data to recognize facial features and body movements, means for converting the response generated based on the recognition results into audio information and outputting it, an emotion evaluation engine for evaluating the user's emotional state, and means for dynamically adjusting an appropriate response based on the evaluated emotional state. This makes it possible for the robot to understand the user's emotions in the home and provide appropriate feedback in real time.

[0732] A "video sensor" is a device used to acquire video data of a target object.

[0733] "Video data of a person" refers to image information that focuses on a specific person, including facial expressions and movements.

[0734] "Analysis" refers to the process of extracting useful information from video data, and is an analytical process for recognizing facial features and body movements.

[0735] "Facial features" refer to information related to an individual's facial expressions and movements.

[0736] "Body movements" refers to information about a person's physical gestures and poses.

[0737] "Recognition results" refer to data showing a person's facial expressions and movements obtained through analysis.

[0738] A "response" is an audio feedback or message generated based on the recognition result.

[0739] "Audio information" refers to messages and notifications expressed through sound.

[0740] An "emotion evaluation engine" is a software module for accurately evaluating a user's emotional state.

[0741] "Means of dynamic adjustment" refers to methods and processes for changing response content in real time in response to the user's emotional evaluation.

[0742] To implement this invention, a system centered around a household robot is constructed as the main component. The details are described below.

[0743] The robot's hardware includes a video sensor (camera), an audio output device (speaker), and a processing unit, which serve as a platform for monitoring and interacting with the user's activities within their home. The video sensor is responsible for acquiring the user's video data in real time, and the acquired video data is sent to the local processing unit.

[0744] In the processing unit, video analysis software runs to recognize facial features and body movements from video data. An example of software used here is the Emotion Analysis SDK. This software analyzes facial features and generates recognition results. These recognition results are further analyzed by an emotion evaluation engine to assess the user's emotional state. This engine detects subtle changes in the user's facial expressions and determines the type and intensity of emotions with high accuracy.

[0745] Next, the server dynamically generates an appropriate voice response based on the output of the emotion evaluation engine. This process applies natural language processing techniques to generate a response based on the recognition results and emotional state. Speech synthesis technologies such as the Text-to-Speech API are used to generate the voice response. The generated voice is then provided to the user through a voice output device.

[0746] This process allows users to receive emotionally-driven, interactive responses in real time through their home robot. For example, when a user is visible on the video sensor, the robot can detect their fatigue and provide voice feedback such as, "You look tired. Shall we take a break?" The user can then feel more comfortable with the robot's response.

[0747] A concrete example of a prompt for using a generative AI model is, "Please tell me how to analyze what emotions a user felt during their workday based on camera footage and generate a response appropriate to those emotions." By utilizing this example prompt, it is possible to design interactions that are more suitable for the user.

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

[0749] Step 1:

[0750] The terminal acquires user video data in real time via a video sensor. This video data serves as input for processing the user's facial expressions and body movements as road and environmental information.

[0751] Step 2:

[0752] The terminal sends the acquired video data to the processing unit, which uses the Emotion Analysis SDK to analyze facial features and body movements. As a result of the analysis, facial feature data and gesture data are output.

[0753] Step 3:

[0754] The server receives the recognition results from the video analysis and uses an emotion evaluation engine to accurately assess the user's emotional state. During this process, it calculates emotion labels and their intensity based on changes in facial expressions and movements. The output includes the type and intensity of the emotion.

[0755] Step 4:

[0756] The server generates an appropriate voice response based on the emotion evaluation results obtained by the emotion evaluation engine. Natural language processing techniques are used in this process, and a generative AI model determines the content of the response. The response content is then generated and output.

[0757] Step 5:

[0758] The server uses the Text-to-Speech API to convert the generated response into speech data. The converted speech data is then output.

[0759] Step 6:

[0760] The terminal presents the user with converted audio information via an audio output device. This audio feedback allows the user to communicate interactively with the robot.

[0761] 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.

[0762] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One 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.

[0763] 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.

[0764] 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.

[0765] 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.

[0766] 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.

[0767] 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.

[0768] 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.

[0769] 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."

[0770] 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.

[0771] 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.

[0772] 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.

[0773] 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.

[0774] 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.

[0775] 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.

[0776] 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.

[0777] 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.

[0778] 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.

[0779] 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.

[0780] 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.

[0781] 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.

[0782] The following is further disclosed regarding the embodiments described above.

[0783] (Claim 1)

[0784] A means of acquiring video data of a person,

[0785] A means for analyzing the aforementioned video data to recognize facial expressions and body gestures,

[0786] A means for converting the feedback generated based on the aforementioned recognition result into audio data and outputting it,

[0787] A system that includes this.

[0788] (Claim 2)

[0789] The system according to claim 1, wherein the generated feedback is an audio message indicating the emotional state of the conversation partner.

[0790] (Claim 3)

[0791] The system according to claim 1, wherein natural language processing technology is applied to the generation of the aforementioned feedback.

[0792] "Example 1"

[0793] (Claim 1)

[0794] Means for acquiring visual information of a person,

[0795] A means for analyzing the aforementioned visual information and recognizing features,

[0796] A means for converting the response generated based on the aforementioned recognition result into audio information and outputting it,

[0797] Means for applying natural language processing techniques to generate the aforementioned response,

[0798] A communication means for efficiently transmitting and receiving the aforementioned visual information,

[0799] A system that includes this.

[0800] (Claim 2)

[0801] The system according to claim 1, wherein the generated response is an audio message indicating the emotional state of the person being communicated with.

[0802] (Claim 3)

[0803] The system according to claim 1, which uses an artificial intelligence model to generate the response.

[0804] "Application Example 1"

[0805] (Claim 1)

[0806] A device for acquiring video data of people,

[0807] The aforementioned video data is analyzed to recognize facial expressions and body gestures,

[0808] A function that converts the feedback generated based on the aforementioned recognition results into audio data and outputs it,

[0809] The aforementioned feedback is provided to caregivers, and a function is presented in real time to understand the emotions and intentions of the person being cared for,

[0810] A system that includes this.

[0811] (Claim 2)

[0812] The system according to claim 1, wherein the generated feedback is an audio message indicating the emotional state of the person being spoken to, and is intended for use in a caregiving environment.

[0813] (Claim 3)

[0814] The system according to claim 1, wherein natural language processing technology is applied to generate the aforementioned feedback and the feedback is notified to the caregiver as voice feedback.

[0815] "Example 2 of combining an emotion engine"

[0816] (Claim 1)

[0817] A means of acquiring video data of a person in real time using a video acquisition device,

[0818] A means for converting the aforementioned video data into an optimized digital data format and transmitting it through a communication network,

[0819] A means for analyzing received video data and recognizing facial expressions and body movements,

[0820] A means for evaluating emotional states based on recognized facial expressions and movements,

[0821] A means for generating feedback using generative AI technology based on the aforementioned emotion evaluation, converting it into audio data, and outputting it,

[0822] A system that includes this.

[0823] (Claim 2)

[0824] The system according to claim 1, wherein the generated feedback is a dynamically adjusted voice message indicating the emotional state of the conversation partner and the user themselves.

[0825] (Claim 3)

[0826] The system according to claim 1, wherein natural language processing technology and computer vision technology are applied to the generation of the aforementioned feedback.

[0827] "Application example 2 when combining with an emotional engine"

[0828] (Claim 1)

[0829] A means of acquiring video data of a person from a video sensor,

[0830] A means for analyzing the aforementioned video data to recognize facial features and body movements,

[0831] A means for converting the response generated based on the aforementioned recognition result into audio information and outputting it,

[0832] A sentiment evaluation engine that evaluates the user's emotional state,

[0833] A means of dynamically adjusting the appropriate response based on the assessed emotional state,

[0834] A system that includes this.

[0835] (Claim 2)

[0836] The system according to claim 1, wherein the generated response is an audio message indicating the emotional state of the conversation partner.

[0837] (Claim 3)

[0838] The system according to claim 1, wherein natural language processing techniques are applied to the generation of the response. [Explanation of symbols]

[0839] 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 video data of a person, A means for analyzing the aforementioned video data to recognize facial expressions and body gestures, A means for converting the feedback generated based on the aforementioned recognition result into audio data and outputting it, A system that includes this.

2. The system according to claim 1, wherein the generated feedback is an audio message indicating the emotional state of the person being spoken to.

3. The system according to claim 1, wherein natural language processing technology is applied to the generation of the aforementioned feedback.

Images