System

Abstract

A system is provided.SOLUTION: A system comprising: means for capturing facial expressions of a user; means for collecting voice of the user; means for transmitting the captured facial expression data and collected voice data to a server; means for recognizing emotions in the server from the facial expression data and voice data; means for generating an appropriate response based on the recognized emotions; and means for providing the generated response to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In modern society, it is important to effectively recognize human emotions and provide appropriate responses in areas such as communication, nursing care, and psychotherapy. However, conventional systems have been unable to fully utilize facial expression and voice data, resulting in issues such as low accuracy in real-time emotion recognition and an inability to provide appropriate responses immediately. Therefore, the present invention aims to solve these issues and provide a system that effectively recognizes emotions and provides support to users in real time. [Means for solving the problem]

[0005] The present invention is a system including means for capturing a facial expression of a user, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to a server, means for recognizing emotions from the facial expression data and voice data in the server, means for generating an appropriate response based on the recognized emotion, and means for providing the generated response to the user.

[0006] Furthermore, the server includes a means for storing the recognized emotion information in a database and improving the accuracy of emotion recognition from next time onward based on the information. The response provided to the user includes a voice message generated using voice synthesis technology. This makes it possible to recognize the user's emotion with high accuracy and provide an optimal response at the right time, improving the user experience.

[0007] "User" refers to the person or end user of the System.

[0008] "Facial expression" refers to the outward features that indicate emotions and reactions based on the movement and position of facial muscles.

[0009] "Voice data" refers to recorded information of a user's speaking voice and tone of voice acquired by a sound collection means such as a microphone.

[0010] A "server" refers to a computer system or networked device that analyzes received data and performs emotion recognition and response generation.

[0011] "Emotion recognition" refers to the process of analyzing facial expressions and voice data to identify the emotions a user is feeling.

[0012] "Response generation" refers to the process of generating appropriate messages or courses of action based on the results of emotion recognition.

[0013] "Database" refers to a system or storage for storing and managing information for emotion recognition and response generation.

[0014] "Speech synthesis technology" refers to technology that converts text data into speech and provides the user with a natural voice message. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

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

[0019] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

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

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0023] [First embodiment]

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

[0025] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0032] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0033] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

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

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

[0036] The system of this invention recognizes emotions from the user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses.

[0037] Overall system configuration

[0038] 1. Collecting facial expressions and voice

[0039] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[0040] 2. Data transmission

[0041] The facial expression and voice data collected by the device is sent to a server via a network, where it is encrypted for security purposes.

[0042] 3. Data Analysis

[0043] The server analyzes the received data, extracting facial landmarks using image processing technology and analyzing voice features using voice processing technology, thereby identifying the user's emotions from facial expressions and tone of voice.

[0044] 4. Emotion recognition

[0045] The server uses an AI model to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[0046] 5. Response Generation

[0047] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[0048] 6. Providing a Response

[0049] The terminal receives the response from the server and provides it to the user, possibly via a screen display or audio playback.

[0050] 7. Gathering Feedback

[0051] After responding, the terminal collects the user's reaction again and sends it to the server to help generate subsequent responses.

[0052] Specific use cases

[0053] Example 1: Caring for the elderly in nursing care settings

[0054] 1. Usage

[0055] An elderly person is relaxing in a care home with a device set up on a table, its camera and microphone capturing the elderly person's face and voice.

[0056] 2. Processing Flow

[0057] The device collects the elderly person's facial expressions (e.g., tired expression) and voice data (low tone of voice).

[0058] The terminal transmits this data to the server.

[0059] The server analyzes the data and extracts facial landmarks and audio features.

[0060] The server's AI model recognizes emotions and gives results such as "tired" or "needs to relax."

[0061] The server generates the response "Play some relaxing music."

[0062] The terminal receives the response from the server and plays relaxing music.

[0063] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0064] Example 2: Managing stress while working remotely

[0065] 1. Usage

[0066] An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[0067] 2. Processing Flow

[0068] The device collects the employee's facial expressions (e.g., stern expression) and voice data (fast speaking tone).

[0069] The terminal transmits this data to the server.

[0070] The server analyzes the data and extracts facial landmarks and audio features.

[0071] The server's AI model recognizes emotions and gives results such as "stress" or "anxiety."

[0072] The server generates the response "Do some deep breathing exercises."

[0073] The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[0074] The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[0075] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and work efficiency. This system is particularly useful in nursing homes and remote work environments.

[0076] The processing flow will be explained below.

[0077] Step 1:

[0078] The device uses a camera to capture the user's facial expressions.

[0079] The device uses a microphone to collect the user's voice.

[0080] Step 2:

[0081] The facial expression data captured by the device and the collected voice data are converted into a certain format.

[0082] The device sends the converted data to the server, where it is encrypted for added security.

[0083] Step 3:

[0084] The server parses the data it receives.

[0085] The server uses image processing techniques to extract facial landmarks.

[0086] The server uses voice processing technology to extract features (pitch, volume, speaking speed, etc.) from the voice data.

[0087] Step 4:

[0088] The server uses an AI model to recognize the user's emotions from pre-processed data.

[0089] The server stores the recognized emotion data in a database.

[0090] Step 5:

[0091] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion.

[0092] The server generates a response by referencing pre-configured rules and case studies.

[0093] Step 6:

[0094] The server creates a text version of the generated response.

[0095] The server converts the text message into a voice message using voice synthesis technology as needed.

[0096] Step 7:

[0097] The terminal receives the response message from the server.

[0098] The terminal provides the response message to the user by displaying it on the screen or playing it as an audio message.

[0099] Step 8:

[0100] The terminal captures the user's reaction again after the response and collects feedback data based on this.

[0101] The terminal transmits the collected feedback data to the server.

[0102] Step 9:

[0103] The server analyzes the collected feedback data and updates the database to improve the accuracy of the AI ​​model.

[0104] The server uses this information to improve emotion recognition and response generation performance in future sessions.

[0105] Example 1

[0106] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0107] Conventional emotion recognition systems have had difficulty accurately recognizing users' emotions and providing appropriate real-time responses. They also lacked the feedback functionality to appropriately adjust responses in response to changes in the user's emotions. Furthermore, they lacked a mechanism for utilizing collected data to improve recognition accuracy in future sessions. This resulted in a decline in the quality of support provided to users, making it difficult to improve customer satisfaction.

[0108] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0109] In this invention, the server includes means for capturing images of a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to a central processing unit, means for recognizing emotions from the facial expression data and voice data in the central processing unit, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, and means for recollecting the user's reactions and using them to generate subsequent responses. This makes it possible to accurately recognize the user's emotions and provide appropriate real-time responses. Furthermore, by utilizing the collected data to improve recognition accuracy in subsequent responses, the quality of support provided to the user can be improved.

[0110] The "means for capturing the user's facial expression" refers to a device or software that uses a camera or image processing device to capture the user's facial expression.

[0111] The "means for collecting the user's voice" refers to a device or software that uses a microphone or voice collection device to record the user's speech or voice.

[0112] "Means for transmitting captured facial expression data and collected voice data to a central processing unit" refers to a device or software that uses a network communication function to transfer collected facial expression data and voice data to a central processing unit (server).

[0113] "Means for recognizing emotions from facial expression data and voice data in a central processing unit" refers to a device or software that executes an algorithm or artificial intelligence model to analyze the collected facial expression data and voice data and identify the user's emotional state.

[0114] The "means for generating an appropriate response based on the recognized emotion" is a program or algorithm for generating a response that is consistent with the identified emotional state.

[0115] The "means for providing the generated response to the user" is a device or software for displaying or audibly outputting the generated response to the user.

[0116] "Means for recollecting user reactions and using them to generate subsequent responses" refers to a device or software that recollects user reactions to the generated response and uses that data to generate the next response or to improve the accuracy of emotion recognition.

[0117] A "storage device" is hardware or software capable of storing data for long or short periods of time.

[0118] The system of this invention recognizes emotions using a user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses. An embodiment of the present invention will be described in detail below.

[0119] Facial expression and voice collection

[0120] The device uses a camera and microphone to collect the user's facial expressions and voice. The device is equipped with a high-resolution camera and a highly sensitive microphone, and is installed with OpenCV as facial recognition software and Google (registered trademark) Speech API as voice collection software, making it possible to acquire data in real time.

[0121] Sending data

[0122] The device encrypts the collected facial expression and voice data and transmits it to a server via a network, using the Advanced Encryption Standard (AES) to ensure data security.

[0123] Data analysis

[0124] The server analyzes the received data, first using image processing technology (e.g., Dlib) to extract facial landmarks (eyes, nose, and mouth positions), and then using audio processing technology (e.g., Librosa) to analyze audio features (pitch, tone, speed, etc.).

[0125] emotion recognition

[0126] The server uses an AI model (e.g., a convolutional neural network using TENSORFLOW®) to recognize the user's emotions from facial expressions and voice data. This AI model works on a pre-trained dataset.

[0127] Response Generation

[0128] The server generates an appropriate response based on the recognized emotion, which is generated in text format and optionally converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[0129] Providing a response

[0130] The device receives the response from the server and provides it to the user. In this case, the method of providing the response may be a screen display or audio playback (e.g., through the device's built-in speaker).

[0131] Collecting feedback

[0132] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected facial expression and voice data is also encrypted and sent.

[0133] Specific use cases

[0134] Example 1: Elderly care in nursing care settings

[0135] 1. An elderly person is relaxing in a care home. A device is set up on the table, with a camera and microphone capturing the elderly person's face and voice.

[0136] 2. The device collects the elderly person's tired facial expression and low-pitched voice.

[0137] 3. The device encrypts this data and sends it to the server.

[0138] 4. The server analyzes the data and extracts facial landmarks and audio features.

[0139] 5. The server's AI model recognizes emotions such as "tired" and "need to relax."

[0140] 6. The server generates the response "Play some relaxing music."

[0141] 7. The device receives the response from the server and plays relaxing music.

[0142] 8. The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0143] Example 2: Managing stress while working remotely

[0144] 1. An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[0145] 2. The device collects audio of the employee's stern facial expression and fast-paced speech.

[0146] 3. The device encrypts this data and sends it to the server.

[0147] 4. The server analyzes the data and extracts facial landmarks and audio features.

[0148] 5. The server's AI model recognizes emotions such as "stress" and "anxiety."

[0149] 6. The server generates the response "Do some deep breathing exercises."

[0150] 7. The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[0151] 8. The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[0152] Prompt Sentence Examples

[0153] "Design a system that recognizes facial expressions (grim faces) and vocalizations (fast speech) that indicate stress in remote workers and suggests deep breathing exercises."

[0154] "Design a system that generates a response, such as playing relaxing music when an elderly person in a care home looks tired."

[0155] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0156] Step 1:

[0157] The device activates the camera and microphone to collect the user's facial expressions and voice in real time. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and has facial recognition software (e.g., OpenCV) and voice collection software (e.g., Google Speech API) installed. For example, when a user smiles and says "hello," their facial expressions and voice are collected.

[0158] Input: User's facial expressions and voice

[0159] Output: Collected facial expression data and voice data

[0160] Step 2:

[0161] The device encrypts the collected facial expression and voice data and sends it to the server via the network. At this time, the Advanced Encryption Standard (AES) is used to ensure data security. For example, a JSON file containing image data of the user's smile and the voice data of "hello" is generated and encrypted.

[0162] Input: Collected facial expression data and voice data

[0163] Output: Encrypted data

[0164] Step 3:

[0165] The server decrypts the received encrypted data, extracts facial landmarks (the positions of the eyes, nose, and mouth) using image processing technology (e.g., Dlib), and analyzes voice features (pitch, tone, speed, etc.) using voice processing technology (e.g., Librosa). For example, it can identify the degree of upturning of the corners of the eyes and the angle of the mouth from an image of the user's smile and analyze the tone of the voice saying "hello."

[0166] Input: Encrypted data

[0167] Output: Analyzed landmark data and audio feature data

[0168] Step 4:

[0169] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from the analyzed landmark data and voice feature data. This AI model operates based on a pre-trained dataset. For example, it identifies the emotion "joy" from the user's smiling face and bright tone of voice.

[0170] Input: Analyzed landmark data and audio feature data

[0171] Output: Recognized emotion data

[0172] Step 5:

[0173] The server generates an appropriate response based on the recognized emotion. This response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. For example, a response such as "Hello, you look well! How's your day?" can be generated and converted into a voice message.

[0174] Input: Recognized emotion data

[0175] Output: The text and audio data of the generated response

[0176] Step 6:

[0177] The device receives the response data from the server and provides it to the user. This response is output by a method such as displaying it on the screen or playing a voice message. For example, a voice message such as "Hello, you look well! How is your day?" is played from the device's speaker.

[0178] Input: The text and audio data of the generated response

[0179] Output: The response provided to the user

[0180] Step 7:

[0181] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected data is also encrypted and sent. For example, if the user smiles again or says, "Actually, I'm a little tired today," their facial expression and voice will be collected again, encrypted, and sent.

[0182] Input: Facial expression data and voice data based on the user's reaction

[0183] Output: Encrypted re-collected data

[0184] (Application example 1)

[0185] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0186] In brick-and-mortar stores, it is necessary to recognize customers' emotions in real time and provide appropriate responses and assistance. However, conventional systems have difficulty implementing specific responses or promotions based on customers' emotions, posing challenges in improving the quality of customer service and customer satisfaction. In addition, responses based on emotion recognition are one-way, and feedback is not used to improve the accuracy of subsequent responses.

[0187] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0188] In this invention, the server includes means for capturing a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to the server, means for recognizing emotions from the facial expression data and voice data, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, means for providing customer service assistance based on emotion recognition in a physical store, and means for playing a promotional video or offering a special discount in accordance with the generated response. This not only enables specific and appropriate responses based on the emotions of customers in a physical store, but also improves customer satisfaction. Furthermore, by storing the recognized emotion information in a database and improving the accuracy of emotion recognition from the next time onwards, the quality of customer service can be continuously improved.

[0189] A "user" is a person who uses the system.

[0190] "Facial expressions" are expressions that appear on the user's face and indicate their emotions and physical state.

[0191] "Voice" refers to voice data including the voice or speech of the user.

[0192] "Capture" refers to obtaining a user's facial expression and voice using devices such as a camera and microphone.

[0193] A "server" is a computing technology device that analyzes collected facial expression and voice data and generates appropriate responses.

[0194] "Emotion recognition" is the technical process of identifying a user's emotional state from collected facial expression and voice data.

[0195] A "response" is a reply or action to the user that is generated based on the recognized emotion.

[0196] "Brick and mortar store" means a physical store that directly offers goods or services.

[0197] The "customer service assistant" is an interactive system that provides appropriate service to users based on emotion recognition results.

[0198] A "promotional video" is video content used to promote products or services for commercial purposes.

[0199] "Special discounts" are discounted services that are lower than the usual price and are offered under certain conditions or circumstances.

[0200] A "database" is an information technology device for systematically storing and managing collected emotional information.

[0201] "Speech synthesis technology" is a technical method for converting text data into voice data.

[0202] The system for implementing this invention mainly consists of a terminal that collects facial expression and voice data, a server that analyzes this data, and a means for providing responses. The terminal that serves as the user interface is equipped with a camera and microphone, and can capture and collect the user's facial expression and voice in real time.

[0203] The device transmits facial expression data captured by the camera and audio data collected by the microphone to a server over the network, where the data is encrypted for security purposes before being transmitted.

[0204] The server then analyzes the received data. Specifically, it uses image processing technology to extract facial landmarks and audio processing technology to analyze audio features. This analysis process uses open source image processing and audio processing libraries (e.g., OpenCV and librosa).

[0205] On the server, a pre-trained generative AI model is used to recognize user emotions from facial expressions and voice data. The AI ​​model is trained on a wide range of emotion datasets and can identify emotions with high accuracy.

[0206] Based on the recognized emotion, the server generates an appropriate response, which may be generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[0207] The generated response is then sent back over the network to the terminal, which then provides the received response to the user. Typically, the response is displayed on the terminal's display or played as an audio message through a speaker.

[0208] Furthermore, in physical stores, customer service assistants can be provided based on emotion recognition. For example, if a user is recognized as interested, a promotional video for that product can be played. If a user is determined to be stressed, a special discount can be offered to encourage a purchase.

[0209] As a concrete example, consider a brick-and-mortar store promoting a new product. When a customer enters the store, a camera captures their facial expressions and collects audio data. If the system determines that the customer is interested (e.g., a smiling face, a lively voice), it plays a promotional video related to the product.

[0210] Use the following as an example prompt:

[0211] "What emotion is this customer experiencing? (Choices: 'happy', 'sad', 'neutral') Please determine this based on facial expression and voice data."

[0212] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0213] Step 1:

[0214] Collecting user facial expressions and voice

[0215] The device uses a camera and a microphone to collect the user's facial expression and voice data. As input, it captures the user's real-time video and audio. The camera captures the user's facial expression, and the microphone collects the voice. As output, it generates the captured facial expression data and collected voice data.

[0216] Step 2:

[0217] Sending data

[0218] The device transmits the collected facial expression data and voice data to a server via a network. The data is encrypted before transmission to ensure security. The input is facial expression data and voice data, and the output is a data packet received by the server.

[0219] Step 3:

[0220] Data analysis

[0221] The server analyzes the received facial expression data and audio data. It extracts facial landmarks using an image processing library (e.g., OpenCV) and analyzes audio features using an audio processing library (e.g., librosa). The inputs are facial expression data and audio data, and the analyzed facial landmarks and audio features are generated as outputs.

[0222] Step 4:

[0223] emotion recognition

[0224] The server uses a pre-trained generative AI model to recognize the user's emotions from the analyzed data. The inputs are facial landmarks and audio features, and the output is recognized emotion data. Specifically, the AI ​​model identifies emotions such as "happy," "sad," and "neutral."

[0225] Step 5:

[0226] Response Generation

[0227] The server generates an appropriate response based on the recognized emotion. The response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. The input is the recognized emotion data, and the output is a text response and a voice message.

[0228] Step 6:

[0229] Providing a response

[0230] The terminal provides the user with the response received from the server. The response is communicated to the user by a screen display or loudspeaker playback. The inputs are text responses and voice messages, and the output generates visual and auditory information that is presented to the user.

[0231] Step 7:

[0232] Collecting feedback

[0233] The device again collects the user's reaction after the response and sends the data to the server to help generate subsequent responses. The user's facial expression and voice data are again collected as input, and feedback data is generated as output.

[0234] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0235] The system of this invention combines an emotion engine to recognize a user's emotions at a high level and provide appropriate responses and support. A system using an emotion engine is primarily composed of a server and a terminal, and has comprehensive functions for improving the accuracy of emotion recognition.

[0236] Overall system configuration

[0237] 1. Collecting facial expressions and voice

[0238] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[0239] 2. Data transmission

[0240] The facial expression and voice data collected by the device is converted into a specific format and sent to a server, where it is encrypted to ensure security.

[0241] 3. Data Analysis and Emotion Recognition

[0242] The server analyzes the received data and recognizes the user's emotions using an emotion engine.

[0243] The emotion engine uses machine learning models to provide a fusion model for emotion recognition that integrates both facial expression data and voice data, enabling it to identify a user's emotional state with high accuracy.

[0244] 4. Response Generation

[0245] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[0246] 5. Providing a Response

[0247] The terminal receives the response from the server and provides it to the user, either by displaying it on the screen or by playing back audio.

[0248] 6. Gathering Feedback

[0249] After the response, the terminal collects the user's reaction again and sends the feedback data to the server to help generate subsequent responses.

[0250] Specific use cases

[0251] Example 1: Managing emotions of elderly people receiving home care

[0252] 1. Usage

[0253] An elderly person is receiving care at home. A device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[0254] 2. Processing Flow

[0255] The device collects the elderly person's facial expressions (e.g., depressed expressions) and voice data (slow speech).

[0256] The terminal transmits this data to the server.

[0257] The server analyzes the data and extracts facial landmarks and audio features.

[0258] The server's emotion engine recognizes emotions and gives results such as "depression" or "fatigue."

[0259] The server generates the response "Play some relaxing music."

[0260] The terminal receives the response from the server and plays relaxing music.

[0261] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0262] Example 2: Managing student focus during remote learning

[0263] 1. Usage

[0264] A student is studying remotely, with a device set up on their desk, and a camera and microphone capturing their face and voice.

[0265] 2. Processing Flow

[0266] The device collects the student's facial expressions (e.g., expression of concentration) and voice data (clear pronunciation).

[0267] The terminal transmits this data to the server.

[0268] The server analyzes the data and extracts facial landmarks and audio features.

[0269] The server's emotion engine recognizes emotions and gives results such as "focused" or "interesting."

[0270] The server generates a response saying "Tell me when to take a break."

[0271] The device receives a response from the server and notifies the user by voice when it is time to take a break.

[0272] The student takes a break according to their response, and the device collects their response again and sends it to the server.

[0273] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and learning efficiency. This system is particularly useful for elderly people receiving care at home and students undergoing remote learning.

[0274] The processing flow will be explained below.

[0275] Step 1:

[0276] The device uses a camera to capture the user's facial expressions, and the facial recognition software installed on the device detects the contours and features of the face and stores them as image data.

[0277] Step 2:

[0278] The device uses a microphone to collect the user's voice, and the voice collection software installed on the device records the user's speech and saves it as audio data.

[0279] Step 3:

[0280] The device converts the captured facial expression data and collected audio data into a certain format (JPEG image, WAV audio file, etc.), which is then encrypted and prepared for transmission.

[0281] Step 4:

[0282] The device sends the encrypted facial expression and voice data to a server over a network, where data integrity and security are ensured.

[0283] Step 5:

[0284] The server analyzes the facial expression data it receives, and the image processing technology installed on the server extracts facial landmarks (the positions of the eyebrows, eyes, mouth, etc.) and detects their movements.

[0285] Step 6:

[0286] The server analyzes the received voice data and uses voice processing technology to extract voice features such as pitch, volume, and speaking rate.

[0287] Step 7:

[0288] The server uses an emotion engine to recognize the user's emotions from the pre-processed facial expression data and voice data. The emotion engine uses a machine learning model to fuse the data and achieve highly accurate emotion recognition.

[0289] Step 8:

[0290] The server stores the recognized emotion data in a database, which is then used as training data to improve the accuracy of future emotion recognition and response generation.

[0291] Step 9:

[0292] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion, which is generated by referencing pre-defined rules and case studies.

[0293] Step 10:

[0294] The server creates the generated response in text format and, if necessary, converts the text message into a voice message using speech synthesis technology.

[0295] Step 11:

[0296] The terminal receives the response message from the server, which is then decrypted and prepared for presentation to the user.

[0297] Step 12:

[0298] The device provides a response message to the user. Specifically, the response message is conveyed to the user through a screen display or audio playback. For example, a voice message saying, "Would you like to play some relaxing music?" is played.

[0299] Step 13:

[0300] The user acts on the response, for example, by following a suggestion from the device and starting to listen to relaxing music.

[0301] Step 14:

[0302] The device then recaptures the user's reaction after the response, collecting changes in facial expressions and voice and saving them as new data.

[0303] Step 15:

[0304] The device sends the newly collected feedback data to the server, where it is re-encrypted and sent over the network to the server.

[0305] Step 16:

[0306] The server analyzes the collected feedback data, which updates the learning database of the AI ​​model, improving the accuracy of emotion recognition and response generation from the next time onwards.

[0307] Example 2

[0308] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0309] Conventional technologies lack systems that consistently recognize a user's emotions and generate appropriate responses based on those emotions. Furthermore, even if a user's facial expressions and voice are collected, the data is transmitted and analyzed separately, resulting in an inefficient process. Furthermore, there are insufficient mechanisms for improving the accuracy of user emotion recognition and for incorporating the results of previous responses into feedback. This results in an inability to provide appropriate support to users, hindering improvements in their quality of life and learning efficiency.

[0310] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting the user's voice and facial expression, a means for transmitting the collected data, a means for performing data analysis and emotion recognition, a means for generating an appropriate response, a means for providing the generated response, and a means for collecting feedback data. This makes it possible to recognize the user's emotion with high accuracy and provide appropriate support in real time based on the recognition result.

[0311] "User" refers to any individual or entity that uses this system.

[0312] "Means for capturing facial expressions" refers to a device or software for capturing a user's facial expressions as images or videos.

[0313] "Audio collection means" refers to any device or software that records a user's speech or voice.

[0314] "Means for converting into a data format" refers to software for converting the collected facial expression data and voice data into a format that can be sent to a server (e.g., JSON format).

[0315] "Means for encryption and transmission" refers to a method for transmitting data to a server using TLS encryption to ensure data security.

[0316] "Server" refers to a centralized computing device that analyzes collected data, recognizes emotions, and generates responses.

[0317] "Means for recognizing emotions using machine learning models" refers to algorithms that use machine learning frameworks such as TensorFlow and PyTorch to analyze a user's facial expressions and voice data to identify emotions.

[0318] "Means for generating an appropriate response" refers to software for generating feedback to the user in the form of a text or voice message based on the recognized emotion.

[0319] "Means for generating a voice message using speech synthesis technology" refers to technology (e.g., speech synthesis software) for converting the generated text response into speech format.

[0320] "Means for providing a response" refers to an interface (such as a display or speaker) for providing the generated text or audio response to the user.

[0321] "Means for collecting feedback data" refers to a method for re-collecting user reactions to the generated responses and using that data to generate future responses and improve the accuracy of emotion recognition.

[0322] This invention relates to a system for recognizing a user's emotions and providing an appropriate response. This system, which is composed of a terminal and a server, improves the accuracy of emotion recognition and provides appropriate support to the user in real time.

[0323] Hardware and software used

[0324] 1. Terminal

[0325] Camera: Used to capture the user's facial expressions. Specifically, a common webcam (e.g., Logitech C920) is used.

[0326] Microphone: Used to collect the user's voice. Specifically, a general USB microphone (e.g., Blue Yeti) is used.

[0327] Software: Utilizes OpenCV and Google Cloud Speech-to-Text APIs for facial recognition and voice data collection.

[0328] 2. Server

[0329] Data analysis: The server uses machine learning frameworks such as TensorFlow and PyTorch to process facial expression and voice data.

[0330] Emotion Recognition: Based on data analysis, a fusion model is used to recognize user emotions with high accuracy.

[0331] Response generation: Generates appropriate responses based on the recognized emotions and generates voice messages using text-to-speech technology (e.g., Amazon Polly) if necessary.

[0332] System operation procedures and examples

[0333] The system works as follows: the device collects the user's facial expressions and voice and sends the data to the server. The server analyzes the data and recognizes the user's emotions. Based on the recognized emotions, the server generates an appropriate response and provides it to the user again via the device. The device then collects the user's feedback and sends it to the server to help improve accuracy next time.

[0334] Specific use cases:

[0335] Example 1: Managing emotions of elderly people receiving home care

[0336] Usage scenario: An elderly person is receiving care at home. The device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[0337] Operational flow: The device collects the elderly person's facial expressions (e.g., a depressed expression) and voice data (speaking slowly) and sends them to the server. The server analyzes the data and recognizes emotions such as "depression" and "fatigue." The server generates a response such as "play relaxing music," and the device plays the music. The elderly person relaxes while listening to the music, and the device collects their responses again and sends them to the server.

[0338] Example 2: Managing student focus during remote learning

[0339] Usage scenario: A student is learning remotely, with the device set up at a desk and the camera and microphone capturing the student's face and voice.

[0340] Operational flow: The device collects the student's facial expressions (e.g., a concentrated expression) and voice data (clear pronunciation) and sends them to the server. The server analyzes the data and recognizes emotions such as "concentration" and "interested." The server generates a response saying "Notify me when it's time to take a break," and the device notifies the student by voice. The student takes an appropriate break, and their response is collected again and sent to the server.

[0341] Prompt Sentence Examples

[0342] Below are some example prompts to be input to the generative AI model:

[0343] "How can we provide appropriate support to elderly people who are depressed while receiving care at home?"

[0344] "What approaches are effective in helping students stay focused while learning remotely?"

[0345] In this way, it is possible to provide appropriate support in real time, especially for elderly people receiving care at home or students undergoing remote learning. This system is extremely effective in improving users' quality of life and learning efficiency.

[0346] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0347] Step 1:

[0348] Facial expression and voice data collection

[0349] The device uses a camera and microphone to collect the user's facial expressions and voice in real time.

[0350] Input: Image data from the camera and audio data from the microphone.

[0351] Processing: Use OpenCV for facial recognition and Google Cloud Speech-to-Text to convert audio data to text.

[0352] Output: Facial expression data (e.g. facial landmark points) and speech text data.

[0353] Specific operation: The camera captures the user's face, and the microphone records what the user says. For example, it collects audio of an elderly person saying, "I'm not feeling well today."

[0354] Step 2:

[0355] Data preprocessing and transmission

[0356] The facial expression data and voice data collected by the device are converted into a fixed data format (JSON format) and sent to the server using TLS encryption.

[0357] Input: Facial expression data, speech text data.

[0358] Processing: Data conversion and TLS encryption.

[0359] Output: A JSON file containing encrypted facial expression and audio data.

[0360] Specific operation: The device converts facial expression data and voice text into JSON format, encrypts it with TLS, and sends it to the server. For example, it contains data such as "Facial landmarks show that the corners of the mouth are down."

[0361] Step 3:

[0362] Data analysis and emotion recognition

[0363] The server analyzes the received data and uses machine learning models to recognize emotions. Specifically, it uses TensorFlow to process facial expression data and voice data.

[0364] Input: Encrypted facial expression and voice data in JSON format.

[0365] Processing: Decryption, data analysis, and emotion recognition using machine learning models.

[0366] Output: Perceived emotion (e.g., "depressed," "tired").

[0367] Specific operation: The server receives the data, extracts facial landmarks and audio features, and inputs them into the fusion model to identify emotions. For example, it extracts features such as "facial landmarks with downturned corners of the mouth" and "negative audio content" and recognizes the emotion as "depressed."

[0368] Step 4:

[0369] Response Generation

[0370] The server generates an appropriate response based on the recognized emotion and creates a voice message using speech synthesis technology.

[0371] Input: Perceived emotion (e.g., "depressed," "tired").

[0372] Processing: Response text generation and speech synthesis.

[0373] Output: Response text and voice message.

[0374] Specific operation: The server generates the text "Playing relaxing music" and synthesizes it to create a voice message.

[0375] Step 5:

[0376] Providing a response

[0377] The device receives the response from the server and provides it to the user, for example by displaying text on the display or playing an audio message over the speaker.

[0378] Input: Response text and voice message.

[0379] Action: Display text or play audio.

[0380] Output: Feedback provided to the user.

[0381] Specific operation: The device will play a voice message saying "Playing relaxing music" and then play relaxing music.

[0382] Step 6:

[0383] Collecting feedback

[0384] The device collects the user's responses and sends them back to the server to help improve accuracy next time.

[0385] Input: User reaction data (facial expressions and voice).

[0386] Processing: Collect and send data again.

[0387] Output: Feedback data.

[0388] Specific operation: The system collects data on whether the elderly person is relaxed while listening to music or if they have positive reactions (e.g., saying "thank you"), and sends the data to a server. For example, this data includes facial landmarks indicating a smile and audio data saying "thank you."

[0389] (Application example 2)

[0390] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0391] In traditional brick-and-mortar stores, it was difficult to immediately grasp customer emotions and reactions and respond accordingly. As a result, it was not possible to provide services that satisfied customers, and the effect of increasing purchasing motivation was limited. It was also difficult to collect feedback based on customer reactions, and this could not be used to improve services.

[0392] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for capturing the user's facial expression, means for collecting the user's voice, and means for transmitting the captured facial expression data and collected voice data to the server. This makes it possible to recognize customer emotions in real time in a physical store and provide appropriate product recommendations or customer service. Furthermore, the customer's reactions can be collected again and transmitted to the server as feedback data, thereby improving the accuracy of emotion recognition from the next time onwards.

[0393] The "means for capturing a user's facial expression" is a device or system that uses a camera or sensor to acquire image data of a user's facial expression.

[0394] The "means for collecting user's voice" refers to a device or system that uses a microphone or a voice recognition device to acquire the user's speech as voice data.

[0395] The "means for transmitting the captured facial expression data and collected voice data to a server" is a communication device or protocol for sending the facial expression data and voice data acquired over a network to a remote server.

[0396] The "means for recognizing emotions from facial expression data and voice data on a server" refers to a technology that uses software and machine learning models installed on the server to analyze received facial expression data and voice data and identify the user's emotions.

[0397] The "means for generating an appropriate response based on the recognized emotion" refers to an algorithm or software that allows the server to create an optimal text or voice message based on the user's emotional information.

[0398] The "means for providing the generated response to the user" refers to a device or system that transmits the generated response message to the user via a display or speaker.

[0399] "A means for recognizing customer emotions in real time in physical stores and making appropriate product suggestions or customer service responses" refers to software and hardware that utilizes emotion recognition technology to dynamically make product suggestions and customer service responses that correspond to the customer's emotional state.

[0400] The "means for recollecting customer reactions and sending them to the server as feedback data" refers to a device or protocol for recollecting the customer's facial expressions and reactions after the interaction and sending that data to an analysis server.

[0401] The system of the present invention recognizes customer emotions at a high level and provides appropriate product suggestions and customer service in a physical store. A specific implementation method thereof will be described in detail below.

[0402] Hardware Configuration

[0403] This system mainly consists of the following hardware:

[0404] 1. Smart glasses or in-store robots: Used to collect customers' facial expressions and voice.

[0405] 2. Camera: Captures the user's facial expressions.

[0406] 3. Microphone: Collects the user's voice.

[0407] 4. Server: Performs data analysis and emotion recognition.

[0408] Software Configuration

[0409] The system includes the following software components:

[0410] 1. OpenCV: A library for collecting and processing facial expressions.

[0411] 2. requests: HTTP request library for sending data.

[0412] 3. Emotion Engine: A machine learning model to analyze the received data and recognize emotions.

[0413] Data processing and calculation

[0414] The system processing flow is as follows:

[0415] 1. Data collection: Using cameras and microphones on smart glasses or robots, customers' facial expressions and voice data are collected in real time, allowing instantaneous acquisition of customer emotional information.

[0416] 2. Data transmission: The collected facial expression and voice data is encrypted and transmitted from the device to the server, ensuring data security.

[0417] 3. Data Analysis: The server analyzes the received data. Specifically, it uses OpenCV to process facial expression data and uses an emotion engine to recognize emotions. The voice data is processed in the same way, and the data is integrated to identify emotions.

[0418] 4. Response Generation: The server generates appropriate product suggestions and customer service responses based on the recognized emotion data. Responses are created as text or voice messages.

[0419] 5. Response provision: The generated response data is provided to the customer through smart glasses or a robot, thereby enhancing customer interaction.

[0420] 6. Feedback collection: After responding, the system collects the customer's reaction again and sends the data to the server, providing feedback to improve the accuracy of emotion recognition in the future.

[0421] Specific examples

[0422] As a specific example, the following scenario can be considered.

[0423] 1. A customer enters the store: A staff member wearing smart glasses approaches the customer and collects facial expressions and voice recordings.

[0424] 2. Data analysis: The collected data is sent to the server, which analyzes it using an emotion engine. If the result indicates that the customer is interested, the server generates appropriate product suggestions.

[0425] 3. Product proposals and responses: The generated product proposals are transmitted to the staff via smart glasses, who then make the proposals to the customer. The customer's responses are collected again and sent to the server as feedback.

[0426] Prompt Sentence Examples

[0427] Examples of prompts for generative AI models include:

[0428] "Please generate a program for a system that analyzes customer facial expressions and voice data, recognizes emotions, and provides appropriate product suggestions or customer service responses to the customer."

[0429] As described above, this invention utilizes emotion recognition technology to improve the quality of service in brick-and-mortar stores and increase customer satisfaction.

[0430] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0431] Step 1:

[0432] The device uses smart glasses or in-store robots to collect customers' facial expressions and voice in real time.

[0433] Input: Customer's facial expression (video data), voice (audio data)

[0434] How it works: The device's camera captures facial expressions, and the microphone collects audio. The facial expressions are captured as image data using the OpenCV library, and the audio is captured as audio data using a voice recognition device.

[0435] Output: Acquired facial expression data and voice data

[0436] Step 2:

[0437] The facial expression data and voice data collected by the terminal are transmitted to a server.

[0438] Input: Facial expression data, voice data

[0439] How it works: The device encrypts facial expression and voice data and sends it securely to the server using HTTP requests, with the data sent in JSON format.

[0440] Output: Facial expression data and voice data sent to the server

[0441] Step 3:

[0442] The server analyzes the received facial expression data and voice data and recognizes emotions using an emotion engine.

[0443] Input: Facial expression data and voice data sent to the server

[0444] How it works: The server uses OpenCV to analyze facial expression data and extract facial landmarks. It also extracts voice features from voice data. The emotion engine integrates these data and uses machine learning models to recognize emotions.

[0445] Output: Recognized emotion data (e.g., joy, interest, anxiety, etc.)

[0446] Step 4:

[0447] The server generates an appropriate response based on the recognized emotion data.

[0448] Input: Recognized emotion data

[0449] How it works: The server uses a response generation algorithm to generate the optimal response (e.g., product suggestions, customer service messages, etc.) based on the emotion. Responses are constructed as text or voice messages.

[0450] Output: Generated response data (text or audio)

[0451] Step 5:

[0452] The terminal receives the response data from the server and provides it to the client.

[0453] Input: Generated response data

[0454] Operation: The device processes the received response data and displays it on the smart glasses display or plays it aloud through the robot's speaker.

[0455] Output: The response (visual or audio message) provided to the customer

[0456] Step 6:

[0457] The terminal again collects the customer's response and sends it to the server as feedback data.

[0458] Input: New customer facial expression data, voice data

[0459] How it works: The device again collects the customer's facial expressions and voice and sends them to the server. Based on this data, the server performs feedback analysis to improve the accuracy of the emotion recognition model.

[0460] Output: Feedback data (facial expression and voice data) sent to the server

[0461] Through these steps, the system can recognize customer emotions in real time and significantly improve the quality of service in physical stores.

[0462] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0463] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0464] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0465] [Second embodiment]

[0466] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0467] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0468] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0469] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0470] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0471] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0472] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0473] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0474] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0475] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0476] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0477] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0478] The system of this invention recognizes emotions from the user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses.

[0479] Overall system configuration

[0480] 1. Collecting facial expressions and voice

[0481] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[0482] 2. Data transmission

[0483] The facial expression and voice data collected by the device is sent to a server via a network, where it is encrypted for security purposes.

[0484] 3. Data Analysis

[0485] The server analyzes the received data, extracting facial landmarks using image processing technology and analyzing voice features using voice processing technology, thereby identifying the user's emotions from facial expressions and tone of voice.

[0486] 4. Emotion recognition

[0487] The server uses an AI model to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[0488] 5. Response Generation

[0489] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[0490] 6. Providing a Response

[0491] The terminal receives the response from the server and provides it to the user, possibly via a screen display or audio playback.

[0492] 7. Gathering Feedback

[0493] After responding, the terminal collects the user's reaction again and sends it to the server to help generate subsequent responses.

[0494] Specific use cases

[0495] Example 1: Caring for the elderly in nursing care settings

[0496] 1. Usage

[0497] An elderly person is relaxing in a care home with a device set up on a table, its camera and microphone capturing the elderly person's face and voice.

[0498] 2. Processing Flow

[0499] The device collects the elderly person's facial expressions (e.g., tired expression) and voice data (low tone of voice).

[0500] The terminal transmits this data to the server.

[0501] The server analyzes the data and extracts facial landmarks and audio features.

[0502] The server's AI model recognizes emotions and gives results such as "tired" or "needs to relax."

[0503] The server generates the response "Play some relaxing music."

[0504] The terminal receives the response from the server and plays relaxing music.

[0505] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0506] Example 2: Managing stress while working remotely

[0507] 1. Usage

[0508] An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[0509] 2. Processing Flow

[0510] The device collects the employee's facial expressions (e.g., stern expression) and voice data (fast speaking tone).

[0511] The terminal transmits this data to the server.

[0512] The server analyzes the data and extracts facial landmarks and audio features.

[0513] The server's AI model recognizes emotions and gives results such as "stress" or "anxiety."

[0514] The server generates the response "Do some deep breathing exercises."

[0515] The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[0516] The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[0517] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and work efficiency. This system is particularly useful in nursing homes and remote work environments.

[0518] The processing flow will be explained below.

[0519] Step 1:

[0520] The device uses a camera to capture the user's facial expressions.

[0521] The device uses a microphone to collect the user's voice.

[0522] Step 2:

[0523] The facial expression data captured by the device and the collected voice data are converted into a certain format.

[0524] The device sends the converted data to the server, where it is encrypted for added security.

[0525] Step 3:

[0526] The server parses the data it receives.

[0527] The server uses image processing techniques to extract facial landmarks.

[0528] The server uses voice processing technology to extract features (pitch, volume, speaking speed, etc.) from the voice data.

[0529] Step 4:

[0530] The server uses an AI model to recognize the user's emotions from pre-processed data.

[0531] The server stores the recognized emotion data in a database.

[0532] Step 5:

[0533] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion.

[0534] The server generates a response by referencing pre-configured rules and case studies.

[0535] Step 6:

[0536] The server creates a text version of the generated response.

[0537] The server converts the text message into a voice message using voice synthesis technology as needed.

[0538] Step 7:

[0539] The terminal receives the response message from the server.

[0540] The terminal provides the response message to the user by displaying it on the screen or playing it as an audio message.

[0541] Step 8:

[0542] The terminal captures the user's reaction again after the response and collects feedback data based on this.

[0543] The terminal transmits the collected feedback data to the server.

[0544] Step 9:

[0545] The server analyzes the collected feedback data and updates the database to improve the accuracy of the AI ​​model.

[0546] The server uses this information to improve emotion recognition and response generation performance in future sessions.

[0547] Example 1

[0548] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0549] Conventional emotion recognition systems have had difficulty accurately recognizing users' emotions and providing appropriate real-time responses. They also lacked the feedback functionality to appropriately adjust responses in response to changes in the user's emotions. Furthermore, they lacked a mechanism for utilizing collected data to improve recognition accuracy in future sessions. This resulted in a decline in the quality of support provided to users, making it difficult to improve customer satisfaction.

[0550] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0551] In this invention, the server includes means for capturing images of a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to a central processing unit, means for recognizing emotions from the facial expression data and voice data in the central processing unit, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, and means for recollecting the user's reactions and using them to generate subsequent responses. This makes it possible to accurately recognize the user's emotions and provide appropriate real-time responses. Furthermore, by utilizing the collected data to improve recognition accuracy in subsequent responses, the quality of support provided to the user can be improved.

[0552] The "means for capturing the user's facial expression" refers to a device or software that uses a camera or image processing device to capture the user's facial expression.

[0553] The "means for collecting the user's voice" refers to a device or software that uses a microphone or voice collection device to record the user's speech or voice.

[0554] "Means for transmitting captured facial expression data and collected voice data to a central processing unit" refers to a device or software that uses a network communication function to transfer collected facial expression data and voice data to a central processing unit (server).

[0555] "Means for recognizing emotions from facial expression data and voice data in a central processing unit" refers to a device or software that executes an algorithm or artificial intelligence model to analyze the collected facial expression data and voice data and identify the user's emotional state.

[0556] The "means for generating an appropriate response based on the recognized emotion" is a program or algorithm for generating a response that is consistent with the identified emotional state.

[0557] The "means for providing the generated response to the user" is a device or software for displaying or audibly outputting the generated response to the user.

[0558] "Means for recollecting user reactions and using them to generate subsequent responses" refers to a device or software that recollects user reactions to the generated response and uses that data to generate the next response or to improve the accuracy of emotion recognition.

[0559] A "storage device" is hardware or software capable of storing data for long or short periods of time.

[0560] The system of this invention recognizes emotions using a user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses. An embodiment of the present invention will be described in detail below.

[0561] Facial expression and voice collection

[0562] The device uses a camera and microphone to collect the user's facial expressions and voice. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and is installed with OpenCV as facial recognition software and Google Speech API as voice collection software, making it possible to acquire data in real time.

[0563] Sending data

[0564] The device encrypts the collected facial expression and voice data and transmits it to a server via a network, using the Advanced Encryption Standard (AES) to ensure data security.

[0565] Data analysis

[0566] The server analyzes the received data, first using image processing technology (e.g., Dlib) to extract facial landmarks (eyes, nose, and mouth positions), and then using audio processing technology (e.g., Librosa) to analyze audio features (pitch, tone, speed, etc.).

[0567] emotion recognition

[0568] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[0569] Response Generation

[0570] The server generates an appropriate response based on the recognized emotion, which is generated in text format and optionally converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[0571] Providing a response

[0572] The device receives the response from the server and provides it to the user. In this case, the method of providing the response may be a screen display or audio playback (e.g., through the device's built-in speaker).

[0573] Collecting feedback

[0574] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected facial expression and voice data is also encrypted and sent.

[0575] Specific use cases

[0576] Example 1: Elderly care in nursing care settings

[0577] 1. An elderly person is relaxing in a care home. A device is set up on the table, with a camera and microphone capturing the elderly person's face and voice.

[0578] 2. The device collects the elderly person's tired facial expression and low-pitched voice.

[0579] 3. The device encrypts this data and sends it to the server.

[0580] 4. The server analyzes the data and extracts facial landmarks and audio features.

[0581] 5. The server's AI model recognizes emotions such as "tired" and "need to relax."

[0582] 6. The server generates the response "Play some relaxing music."

[0583] 7. The device receives the response from the server and plays relaxing music.

[0584] 8. The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0585] Example 2: Managing stress while working remotely

[0586] 1. An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[0587] 2. The device collects audio of the employee's stern facial expression and fast-paced speech.

[0588] 3. The device encrypts this data and sends it to the server.

[0589] 4. The server analyzes the data and extracts facial landmarks and audio features.

[0590] 5. The server's AI model recognizes emotions such as "stress" and "anxiety."

[0591] 6. The server generates the response "Do some deep breathing exercises."

[0592] 7. The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[0593] 8. The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[0594] Prompt Sentence Examples

[0595] "Design a system that recognizes facial expressions (grim faces) and vocalizations (fast speech) that indicate stress in remote workers and suggests deep breathing exercises."

[0596] "Design a system that generates a response, such as playing relaxing music when an elderly person in a care home looks tired."

[0597] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0598] Step 1:

[0599] The device activates the camera and microphone to collect the user's facial expressions and voice in real time. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and has facial recognition software (e.g., OpenCV) and voice collection software (e.g., Google Speech API) installed. For example, when a user smiles and says "hello," their facial expressions and voice are collected.

[0600] Input: User's facial expressions and voice

[0601] Output: Collected facial expression data and voice data

[0602] Step 2:

[0603] The device encrypts the collected facial expression and voice data and sends it to the server via the network. At this time, the Advanced Encryption Standard (AES) is used to ensure data security. For example, a JSON file containing image data of the user's smile and the voice data of "hello" is generated and encrypted.

[0604] Input: Collected facial expression data and voice data

[0605] Output: Encrypted data

[0606] Step 3:

[0607] The server decrypts the received encrypted data, extracts facial landmarks (the positions of the eyes, nose, and mouth) using image processing technology (e.g., Dlib), and analyzes voice features (pitch, tone, speed, etc.) using voice processing technology (e.g., Librosa). For example, it can identify the degree of upturning of the corners of the eyes and the angle of the mouth from an image of the user's smile and analyze the tone of the voice saying "hello."

[0608] Input: Encrypted data

[0609] Output: Analyzed landmark data and audio feature data

[0610] Step 4:

[0611] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from the analyzed landmark data and voice feature data. This AI model operates based on a pre-trained dataset. For example, it identifies the emotion "joy" from the user's smiling face and bright tone of voice.

[0612] Input: Analyzed landmark data and audio feature data

[0613] Output: Recognized emotion data

[0614] Step 5:

[0615] The server generates an appropriate response based on the recognized emotion. This response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. For example, a response such as "Hello, you look well! How's your day?" can be generated and converted into a voice message.

[0616] Input: Recognized emotion data

[0617] Output: The text and audio data of the generated response

[0618] Step 6:

[0619] The device receives the response data from the server and provides it to the user. This response is output by a method such as displaying it on the screen or playing a voice message. For example, a voice message such as "Hello, you look well! How is your day?" is played from the device's speaker.

[0620] Input: The text and audio data of the generated response

[0621] Output: The response provided to the user

[0622] Step 7:

[0623] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected data is also encrypted and sent. For example, if the user smiles again or says, "Actually, I'm a little tired today," their facial expression and voice will be collected again, encrypted, and sent.

[0624] Input: Facial expression data and voice data based on the user's reaction

[0625] Output: Encrypted re-collected data

[0626] (Application example 1)

[0627] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0628] In brick-and-mortar stores, it is necessary to recognize customers' emotions in real time and provide appropriate responses and assistance. However, conventional systems have difficulty implementing specific responses or promotions based on customers' emotions, posing challenges in improving the quality of customer service and customer satisfaction. In addition, responses based on emotion recognition are one-way, and feedback is not used to improve the accuracy of subsequent responses.

[0629] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0630] In this invention, the server includes means for capturing a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to the server, means for recognizing emotions from the facial expression data and voice data, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, means for providing customer service assistance based on emotion recognition in a physical store, and means for playing a promotional video or offering a special discount in accordance with the generated response. This not only enables specific and appropriate responses based on the emotions of customers in a physical store, but also improves customer satisfaction. Furthermore, by storing the recognized emotion information in a database and improving the accuracy of emotion recognition from the next time onwards, the quality of customer service can be continuously improved.

[0631] A "user" is a person who uses the system.

[0632] "Facial expressions" are expressions that appear on the user's face and indicate their emotions and physical state.

[0633] "Voice" refers to voice data including the voice or speech of the user.

[0634] "Capture" refers to obtaining a user's facial expression and voice using devices such as a camera and microphone.

[0635] A "server" is a computing technology device that analyzes collected facial expression and voice data and generates appropriate responses.

[0636] "Emotion recognition" is the technical process of identifying a user's emotional state from collected facial expression and voice data.

[0637] A "response" is a reply or action to the user that is generated based on the recognized emotion.

[0638] "Brick and mortar store" means a physical store that directly offers goods or services.

[0639] The "customer service assistant" is an interactive system that provides appropriate service to users based on emotion recognition results.

[0640] A "promotional video" is video content used to promote products or services for commercial purposes.

[0641] "Special discounts" are discounted services that are lower than the usual price and are offered under certain conditions or circumstances.

[0642] A "database" is an information technology device for systematically storing and managing collected emotional information.

[0643] "Speech synthesis technology" is a technical method for converting text data into voice data.

[0644] The system for implementing this invention mainly consists of a terminal that collects facial expression and voice data, a server that analyzes this data, and a means for providing responses. The terminal that serves as the user interface is equipped with a camera and microphone, and can capture and collect the user's facial expression and voice in real time.

[0645] The device transmits facial expression data captured by the camera and audio data collected by the microphone to a server over the network, where the data is encrypted for security purposes before being transmitted.

[0646] The server then analyzes the received data. Specifically, it uses image processing technology to extract facial landmarks and audio processing technology to analyze audio features. This analysis process uses open source image processing and audio processing libraries (e.g., OpenCV and librosa).

[0647] On the server, a pre-trained generative AI model is used to recognize user emotions from facial expressions and voice data. The AI ​​model is trained on a wide range of emotion datasets and can identify emotions with high accuracy.

[0648] Based on the recognized emotion, the server generates an appropriate response, which may be generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[0649] The generated response is then sent back over the network to the terminal, which then provides the received response to the user. Typically, the response is displayed on the terminal's display or played as an audio message through a speaker.

[0650] Furthermore, in physical stores, customer service assistants can be provided based on emotion recognition. For example, if a user is recognized as interested, a promotional video for that product can be played. If a user is determined to be stressed, a special discount can be offered to encourage a purchase.

[0651] As a concrete example, consider a brick-and-mortar store promoting a new product. When a customer enters the store, a camera captures their facial expressions and collects audio data. If the system determines that the customer is interested (e.g., a smiling face, a lively voice), it plays a promotional video related to the product.

[0652] Use the following as an example prompt:

[0653] "What emotion is this customer experiencing? (Choices: 'happy', 'sad', 'neutral') Please determine this based on facial expression and voice data."

[0654] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0655] Step 1:

[0656] Collecting user facial expressions and voice

[0657] The device uses a camera and a microphone to collect the user's facial expression and voice data. As input, it captures the user's real-time video and audio. The camera captures the user's facial expression, and the microphone collects the voice. As output, it generates the captured facial expression data and collected voice data.

[0658] Step 2:

[0659] Sending data

[0660] The device transmits the collected facial expression data and voice data to a server via a network. The data is encrypted before transmission to ensure security. The input is facial expression data and voice data, and the output is a data packet received by the server.

[0661] Step 3:

[0662] Data analysis

[0663] The server analyzes the received facial expression data and audio data. It extracts facial landmarks using an image processing library (e.g., OpenCV) and analyzes audio features using an audio processing library (e.g., librosa). The inputs are facial expression data and audio data, and the analyzed facial landmarks and audio features are generated as outputs.

[0664] Step 4:

[0665] emotion recognition

[0666] The server uses a pre-trained generative AI model to recognize the user's emotions from the analyzed data. The inputs are facial landmarks and audio features, and the output is recognized emotion data. Specifically, the AI ​​model identifies emotions such as "happy," "sad," and "neutral."

[0667] Step 5:

[0668] Response Generation

[0669] The server generates an appropriate response based on the recognized emotion. The response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. The input is the recognized emotion data, and the output is a text response and a voice message.

[0670] Step 6:

[0671] Providing a response

[0672] The terminal provides the user with the response received from the server. The response is communicated to the user by a screen display or loudspeaker playback. The inputs are text responses and voice messages, and the output generates visual and auditory information that is presented to the user.

[0673] Step 7:

[0674] Collecting feedback

[0675] The device again collects the user's reaction after the response and sends the data to the server to help generate subsequent responses. The user's facial expression and voice data are again collected as input, and feedback data is generated as output.

[0676] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0677] The system of this invention combines an emotion engine to recognize a user's emotions at a high level and provide appropriate responses and support. A system using an emotion engine is primarily composed of a server and a terminal, and has comprehensive functions for improving the accuracy of emotion recognition.

[0678] Overall system configuration

[0679] 1. Collecting facial expressions and voice

[0680] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[0681] 2. Data transmission

[0682] The facial expression and voice data collected by the device is converted into a specific format and sent to a server, where it is encrypted to ensure security.

[0683] 3. Data Analysis and Emotion Recognition

[0684] The server analyzes the received data and recognizes the user's emotions using an emotion engine.

[0685] The emotion engine uses machine learning models to provide a fusion model for emotion recognition that integrates both facial expression data and voice data, enabling it to identify a user's emotional state with high accuracy.

[0686] 4. Response Generation

[0687] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[0688] 5. Providing a Response

[0689] The terminal receives the response from the server and provides it to the user, either by displaying it on the screen or by playing back audio.

[0690] 6. Gathering Feedback

[0691] After the response, the terminal collects the user's reaction again and sends the feedback data to the server to help generate subsequent responses.

[0692] Specific use cases

[0693] Example 1: Managing emotions of elderly people receiving home care

[0694] 1. Usage

[0695] An elderly person is receiving care at home. A device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[0696] 2. Processing Flow

[0697] The device collects the elderly person's facial expressions (e.g., depressed expressions) and voice data (slow speech).

[0698] The terminal transmits this data to the server.

[0699] The server analyzes the data and extracts facial landmarks and audio features.

[0700] The server's emotion engine recognizes emotions and gives results such as "depression" or "fatigue."

[0701] The server generates the response "Play some relaxing music."

[0702] The terminal receives the response from the server and plays relaxing music.

[0703] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0704] Example 2: Managing student focus during remote learning

[0705] 1. Usage

[0706] A student is studying remotely, with a device set up on their desk, and a camera and microphone capturing their face and voice.

[0707] 2. Processing Flow

[0708] The device collects the student's facial expressions (e.g., expression of concentration) and voice data (clear pronunciation).

[0709] The terminal transmits this data to the server.

[0710] The server analyzes the data and extracts facial landmarks and audio features.

[0711] The server's emotion engine recognizes emotions and gives results such as "focused" or "interesting."

[0712] The server generates a response saying "Tell me when to take a break."

[0713] The device receives a response from the server and notifies the user by voice when it is time to take a break.

[0714] The student takes a break according to their response, and the device collects their response again and sends it to the server.

[0715] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and learning efficiency. This system is particularly useful for elderly people receiving care at home and students undergoing remote learning.

[0716] The processing flow will be explained below.

[0717] Step 1:

[0718] The device uses a camera to capture the user's facial expressions, and the facial recognition software installed on the device detects the contours and features of the face and stores them as image data.

[0719] Step 2:

[0720] The device uses a microphone to collect the user's voice, and the voice collection software installed on the device records the user's speech and saves it as audio data.

[0721] Step 3:

[0722] The device converts the captured facial expression data and collected audio data into a certain format (JPEG image, WAV audio file, etc.), which is then encrypted and prepared for transmission.

[0723] Step 4:

[0724] The device sends the encrypted facial expression and voice data to a server over a network, where data integrity and security are ensured.

[0725] Step 5:

[0726] The server analyzes the facial expression data it receives, and the image processing technology installed on the server extracts facial landmarks (the positions of the eyebrows, eyes, mouth, etc.) and detects their movements.

[0727] Step 6:

[0728] The server analyzes the received voice data and uses voice processing technology to extract voice features such as pitch, volume, and speaking rate.

[0729] Step 7:

[0730] The server uses an emotion engine to recognize the user's emotions from the pre-processed facial expression data and voice data. The emotion engine uses a machine learning model to fuse the data and achieve highly accurate emotion recognition.

[0731] Step 8:

[0732] The server stores the recognized emotion data in a database, which is then used as training data to improve the accuracy of future emotion recognition and response generation.

[0733] Step 9:

[0734] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion, which is generated by referencing pre-defined rules and case studies.

[0735] Step 10:

[0736] The server creates the generated response in text format and, if necessary, converts the text message into a voice message using speech synthesis technology.

[0737] Step 11:

[0738] The terminal receives the response message from the server, which is then decrypted and prepared for presentation to the user.

[0739] Step 12:

[0740] The device provides a response message to the user. Specifically, the response message is conveyed to the user through a screen display or audio playback. For example, a voice message saying, "Would you like to play some relaxing music?" is played.

[0741] Step 13:

[0742] The user acts on the response, for example, by following a suggestion from the device and starting to listen to relaxing music.

[0743] Step 14:

[0744] The device then recaptures the user's reaction after the response, collecting changes in facial expressions and voice and saving them as new data.

[0745] Step 15:

[0746] The device sends the newly collected feedback data to the server, where it is re-encrypted and sent over the network to the server.

[0747] Step 16:

[0748] The server analyzes the collected feedback data, which updates the learning database of the AI ​​model, improving the accuracy of emotion recognition and response generation from the next time onwards.

[0749] Example 2

[0750] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0751] Conventional technologies lack systems that consistently recognize a user's emotions and generate appropriate responses based on those emotions. Furthermore, even if a user's facial expressions and voice are collected, the data is transmitted and analyzed separately, resulting in an inefficient process. Furthermore, there are insufficient mechanisms for improving the accuracy of user emotion recognition and for incorporating the results of previous responses into feedback. This results in an inability to provide appropriate support to users, hindering improvements in their quality of life and learning efficiency.

[0752] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting the user's voice and facial expression, a means for transmitting the collected data, a means for performing data analysis and emotion recognition, a means for generating an appropriate response, a means for providing the generated response, and a means for collecting feedback data. This makes it possible to recognize the user's emotion with high accuracy and provide appropriate support in real time based on the recognition result.

[0753] "User" refers to any individual or entity that uses this system.

[0754] "Means for capturing facial expressions" refers to a device or software for capturing a user's facial expressions as images or videos.

[0755] "Audio collection means" refers to any device or software that records a user's speech or voice.

[0756] "Means for converting into a data format" refers to software for converting the collected facial expression data and voice data into a format that can be sent to a server (e.g., JSON format).

[0757] "Means for encryption and transmission" refers to a method for transmitting data to a server using TLS encryption to ensure data security.

[0758] "Server" refers to a centralized computing device that analyzes collected data, recognizes emotions, and generates responses.

[0759] "Means for recognizing emotions using machine learning models" refers to algorithms that use machine learning frameworks such as TensorFlow and PyTorch to analyze a user's facial expressions and voice data to identify emotions.

[0760] "Means for generating an appropriate response" refers to software for generating feedback to the user in the form of a text or voice message based on the recognized emotion.

[0761] "Means for generating a voice message using speech synthesis technology" refers to technology (e.g., speech synthesis software) for converting the generated text response into speech format.

[0762] "Means for providing a response" refers to an interface (such as a display or speaker) for providing the generated text or audio response to the user.

[0763] "Means for collecting feedback data" refers to a method for re-collecting user reactions to the generated responses and using that data to generate future responses and improve the accuracy of emotion recognition.

[0764] This invention relates to a system for recognizing a user's emotions and providing an appropriate response. This system, which is composed of a terminal and a server, improves the accuracy of emotion recognition and provides appropriate support to the user in real time.

[0765] Hardware and software used

[0766] 1. Terminal

[0767] Camera: Used to capture the user's facial expressions. Specifically, a common webcam (e.g., Logitech C920) is used.

[0768] Microphone: Used to collect the user's voice. Specifically, a general USB microphone (e.g., Blue Yeti) is used.

[0769] Software: Utilizes OpenCV and Google Cloud Speech-to-Text APIs for facial recognition and voice data collection.

[0770] 2. Server

[0771] Data analysis: The server uses machine learning frameworks such as TensorFlow and PyTorch to process facial expression and voice data.

[0772] Emotion Recognition: Based on data analysis, a fusion model is used to recognize user emotions with high accuracy.

[0773] Response generation: Generates appropriate responses based on the recognized emotions and generates voice messages using text-to-speech technology (e.g., Amazon Polly) if necessary.

[0774] System operation procedures and examples

[0775] The system works as follows: the device collects the user's facial expressions and voice and sends the data to the server. The server analyzes the data and recognizes the user's emotions. Based on the recognized emotions, the server generates an appropriate response and provides it to the user again via the device. The device then collects the user's feedback and sends it to the server to help improve accuracy next time.

[0776] Specific use cases:

[0777] Example 1: Managing emotions of elderly people receiving home care

[0778] Usage scenario: An elderly person is receiving care at home. The device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[0779] Operational flow: The device collects the elderly person's facial expressions (e.g., a depressed expression) and voice data (speaking slowly) and sends them to the server. The server analyzes the data and recognizes emotions such as "depression" and "fatigue." The server generates a response such as "play relaxing music," and the device plays the music. The elderly person relaxes while listening to the music, and the device collects their responses again and sends them to the server.

[0780] Example 2: Managing student focus during remote learning

[0781] Usage scenario: A student is learning remotely, with the device set up at a desk and the camera and microphone capturing the student's face and voice.

[0782] Operational flow: The device collects the student's facial expressions (e.g., a concentrated expression) and voice data (clear pronunciation) and sends them to the server. The server analyzes the data and recognizes emotions such as "concentration" and "interested." The server generates a response saying "Notify me when it's time to take a break," and the device notifies the student by voice. The student takes an appropriate break, and their response is collected again and sent to the server.

[0783] Prompt Sentence Examples

[0784] Below are some example prompts to be input to the generative AI model:

[0785] "How can we provide appropriate support to elderly people who are depressed while receiving care at home?"

[0786] "What approaches are effective in helping students stay focused while learning remotely?"

[0787] In this way, it is possible to provide appropriate support in real time, especially for elderly people receiving care at home or students undergoing remote learning. This system is extremely effective in improving users' quality of life and learning efficiency.

[0788] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0789] Step 1:

[0790] Facial expression and voice data collection

[0791] The device uses a camera and microphone to collect the user's facial expressions and voice in real time.

[0792] Input: Image data from the camera and audio data from the microphone.

[0793] Processing: Use OpenCV for facial recognition and Google Cloud Speech-to-Text to convert audio data to text.

[0794] Output: Facial expression data (e.g. facial landmark points) and speech text data.

[0795] Specific operation: The camera captures the user's face, and the microphone records what the user says. For example, it collects audio of an elderly person saying, "I'm not feeling well today."

[0796] Step 2:

[0797] Data preprocessing and transmission

[0798] The facial expression data and voice data collected by the device are converted into a fixed data format (JSON format) and sent to the server using TLS encryption.

[0799] Input: Facial expression data, speech text data.

[0800] Processing: Data conversion and TLS encryption.

[0801] Output: A JSON file containing encrypted facial expression and audio data.

[0802] Specific operation: The device converts facial expression data and voice text into JSON format, encrypts it with TLS, and sends it to the server. For example, it contains data such as "Facial landmarks show that the corners of the mouth are down."

[0803] Step 3:

[0804] Data analysis and emotion recognition

[0805] The server analyzes the received data and uses machine learning models to recognize emotions. Specifically, it uses TensorFlow to process facial expression data and voice data.

[0806] Input: Encrypted facial expression and voice data in JSON format.

[0807] Processing: Decryption, data analysis, and emotion recognition using machine learning models.

[0808] Output: Perceived emotion (e.g., "depressed," "tired").

[0809] Specific operation: The server receives the data, extracts facial landmarks and audio features, and inputs them into the fusion model to identify emotions. For example, it extracts features such as "facial landmarks with downturned corners of the mouth" and "negative audio content" and recognizes the emotion as "depressed."

[0810] Step 4:

[0811] Response Generation

[0812] The server generates an appropriate response based on the recognized emotion and creates a voice message using speech synthesis technology.

[0813] Input: Perceived emotion (e.g., "depressed," "tired").

[0814] Processing: Response text generation and speech synthesis.

[0815] Output: Response text and voice message.

[0816] Specific operation: The server generates the text "Playing relaxing music" and synthesizes it to create a voice message.

[0817] Step 5:

[0818] Providing a response

[0819] The device receives the response from the server and provides it to the user, for example by displaying text on the display or playing an audio message over the speaker.

[0820] Input: Response text and voice message.

[0821] Action: Display text or play audio.

[0822] Output: Feedback provided to the user.

[0823] Specific operation: The device will play a voice message saying "Playing relaxing music" and then play relaxing music.

[0824] Step 6:

[0825] Collecting feedback

[0826] The device collects the user's responses and sends them back to the server to help improve accuracy next time.

[0827] Input: User reaction data (facial expressions and voice).

[0828] Processing: Collect and send data again.

[0829] Output: Feedback data.

[0830] Specific operation: The system collects data on whether the elderly person is relaxed while listening to music or if they have positive reactions (e.g., saying "thank you"), and sends the data to a server. For example, this data includes facial landmarks indicating a smile and audio data saying "thank you."

[0831] (Application example 2)

[0832] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0833] In traditional brick-and-mortar stores, it was difficult to immediately grasp customer emotions and reactions and respond accordingly. As a result, it was not possible to provide services that satisfied customers, and the effect of increasing purchasing motivation was limited. It was also difficult to collect feedback based on customer reactions, and this could not be used to improve services.

[0834] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for capturing the user's facial expression, means for collecting the user's voice, and means for transmitting the captured facial expression data and collected voice data to the server. This makes it possible to recognize customer emotions in real time in a physical store and provide appropriate product recommendations or customer service. Furthermore, the customer's reactions can be collected again and transmitted to the server as feedback data, thereby improving the accuracy of emotion recognition from the next time onwards.

[0835] The "means for capturing a user's facial expression" is a device or system that uses a camera or sensor to acquire image data of a user's facial expression.

[0836] The "means for collecting user's voice" refers to a device or system that uses a microphone or a voice recognition device to acquire the user's speech as voice data.

[0837] The "means for transmitting the captured facial expression data and collected voice data to a server" is a communication device or protocol for sending the facial expression data and voice data acquired over a network to a remote server.

[0838] The "means for recognizing emotions from facial expression data and voice data on a server" refers to a technology that uses software and machine learning models installed on the server to analyze received facial expression data and voice data and identify the user's emotions.

[0839] The "means for generating an appropriate response based on the recognized emotion" refers to an algorithm or software that allows the server to create an optimal text or voice message based on the user's emotional information.

[0840] The "means for providing the generated response to the user" refers to a device or system that transmits the generated response message to the user via a display or speaker.

[0841] "A means for recognizing customer emotions in real time in physical stores and making appropriate product suggestions or customer service responses" refers to software and hardware that utilizes emotion recognition technology to dynamically make product suggestions and customer service responses that correspond to the customer's emotional state.

[0842] The "means for recollecting customer reactions and sending them to the server as feedback data" refers to a device or protocol for recollecting the customer's facial expressions and reactions after the interaction and sending that data to an analysis server.

[0843] The system of the present invention recognizes customer emotions at a high level and provides appropriate product suggestions and customer service in a physical store. A specific implementation method thereof will be described in detail below.

[0844] Hardware Configuration

[0845] This system mainly consists of the following hardware:

[0846] 1. Smart glasses or in-store robots: Used to collect customers' facial expressions and voice.

[0847] 2. Camera: Captures the user's facial expressions.

[0848] 3. Microphone: Collects the user's voice.

[0849] 4. Server: Performs data analysis and emotion recognition.

[0850] Software Configuration

[0851] The system includes the following software components:

[0852] 1. OpenCV: A library for collecting and processing facial expressions.

[0853] 2. requests: HTTP request library for sending data.

[0854] 3. Emotion Engine: A machine learning model to analyze the received data and recognize emotions.

[0855] Data processing and calculation

[0856] The system processing flow is as follows:

[0857] 1. Data collection: Using cameras and microphones on smart glasses or robots, customers' facial expressions and voice data are collected in real time, allowing instantaneous acquisition of customer emotional information.

[0858] 2. Data transmission: The collected facial expression and voice data is encrypted and transmitted from the device to the server, ensuring data security.

[0859] 3. Data Analysis: The server analyzes the received data. Specifically, it uses OpenCV to process facial expression data and uses an emotion engine to recognize emotions. The voice data is processed in the same way, and the data is integrated to identify emotions.

[0860] 4. Response Generation: The server generates appropriate product suggestions and customer service responses based on the recognized emotion data. Responses are created as text or voice messages.

[0861] 5. Response provision: The generated response data is provided to the customer through smart glasses or a robot, thereby enhancing customer interaction.

[0862] 6. Feedback collection: After responding, the system collects the customer's reaction again and sends the data to the server, providing feedback to improve the accuracy of emotion recognition in the future.

[0863] Specific examples

[0864] As a specific example, the following scenario can be considered.

[0865] 1. A customer enters the store: A staff member wearing smart glasses approaches the customer and collects facial expressions and voice recordings.

[0866] 2. Data analysis: The collected data is sent to the server, which analyzes it using an emotion engine. If the result indicates that the customer is interested, the server generates appropriate product suggestions.

[0867] 3. Product proposals and responses: The generated product proposals are transmitted to the staff via smart glasses, who then make the proposals to the customer. The customer's responses are collected again and sent to the server as feedback.

[0868] Prompt Sentence Examples

[0869] Examples of prompts for generative AI models include:

[0870] "Please generate a program for a system that analyzes customer facial expressions and voice data, recognizes emotions, and provides appropriate product suggestions or customer service responses to the customer."

[0871] As described above, this invention utilizes emotion recognition technology to improve the quality of service in brick-and-mortar stores and increase customer satisfaction.

[0872] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0873] Step 1:

[0874] The device uses smart glasses or in-store robots to collect customers' facial expressions and voice in real time.

[0875] Input: Customer's facial expression (video data), voice (audio data)

[0876] How it works: The device's camera captures facial expressions, and the microphone collects audio. The facial expressions are captured as image data using the OpenCV library, and the audio is captured as audio data using a voice recognition device.

[0877] Output: Acquired facial expression data and voice data

[0878] Step 2:

[0879] The facial expression data and voice data collected by the terminal are transmitted to a server.

[0880] Input: Facial expression data, voice data

[0881] How it works: The device encrypts facial expression and voice data and sends it securely to the server using HTTP requests, with the data sent in JSON format.

[0882] Output: Facial expression data and voice data sent to the server

[0883] Step 3:

[0884] The server analyzes the received facial expression data and voice data and recognizes emotions using an emotion engine.

[0885] Input: Facial expression data and voice data sent to the server

[0886] How it works: The server uses OpenCV to analyze facial expression data and extract facial landmarks. It also extracts voice features from voice data. The emotion engine integrates these data and uses machine learning models to recognize emotions.

[0887] Output: Recognized emotion data (e.g., joy, interest, anxiety, etc.)

[0888] Step 4:

[0889] The server generates an appropriate response based on the recognized emotion data.

[0890] Input: Recognized emotion data

[0891] How it works: The server uses a response generation algorithm to generate the optimal response (e.g., product suggestions, customer service messages, etc.) based on the emotion. Responses are constructed as text or voice messages.

[0892] Output: Generated response data (text or audio)

[0893] Step 5:

[0894] The terminal receives the response data from the server and provides it to the client.

[0895] Input: Generated response data

[0896] Operation: The device processes the received response data and displays it on the smart glasses display or plays it aloud through the robot's speaker.

[0897] Output: The response (visual or audio message) provided to the customer

[0898] Step 6:

[0899] The terminal again collects the customer's response and sends it to the server as feedback data.

[0900] Input: New customer facial expression data, voice data

[0901] How it works: The device again collects the customer's facial expressions and voice and sends them to the server. Based on this data, the server performs feedback analysis to improve the accuracy of the emotion recognition model.

[0902] Output: Feedback data (facial expression and voice data) sent to the server

[0903] Through these steps, the system can recognize customer emotions in real time and significantly improve the quality of service in physical stores.

[0904] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0905] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0906] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0907] [Third embodiment]

[0908] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0909] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0910] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0911] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0912] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0913] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0914] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0915] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0916] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0917] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0918] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0919] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0920] The system of this invention recognizes emotions from the user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses.

[0921] Overall system configuration

[0922] 1. Collecting facial expressions and voice

[0923] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[0924] 2. Data transmission

[0925] The facial expression and voice data collected by the device is sent to a server via a network, where it is encrypted for security purposes.

[0926] 3. Data Analysis

[0927] The server analyzes the received data, extracting facial landmarks using image processing technology and analyzing voice features using voice processing technology, thereby identifying the user's emotions from facial expressions and tone of voice.

[0928] 4. Emotion recognition

[0929] The server uses an AI model to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[0930] 5. Response Generation

[0931] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[0932] 6. Providing a Response

[0933] The terminal receives the response from the server and provides it to the user, possibly via a screen display or audio playback.

[0934] 7. Gathering Feedback

[0935] After responding, the terminal collects the user's reaction again and sends it to the server to help generate subsequent responses.

[0936] Specific use cases

[0937] Example 1: Caring for the elderly in nursing care settings

[0938] 1. Usage

[0939] An elderly person is relaxing in a care home with a device set up on a table, its camera and microphone capturing the elderly person's face and voice.

[0940] 2. Processing Flow

[0941] The device collects the elderly person's facial expressions (e.g., tired expression) and voice data (low tone of voice).

[0942] The terminal transmits this data to the server.

[0943] The server analyzes the data and extracts facial landmarks and audio features.

[0944] The server's AI model recognizes emotions and gives results such as "tired" or "needs to relax."

[0945] The server generates the response "Play some relaxing music."

[0946] The terminal receives the response from the server and plays relaxing music.

[0947] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[0948] Example 2: Managing stress while working remotely

[0949] 1. Usage

[0950] An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[0951] 2. Processing Flow

[0952] The device collects the employee's facial expressions (e.g., stern expression) and voice data (fast speaking tone).

[0953] The terminal transmits this data to the server.

[0954] The server analyzes the data and extracts facial landmarks and audio features.

[0955] The server's AI model recognizes emotions and gives results such as "stress" or "anxiety."

[0956] The server generates the response "Do some deep breathing exercises."

[0957] The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[0958] The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[0959] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and work efficiency. This system is particularly useful in nursing homes and remote work environments.

[0960] The processing flow will be explained below.

[0961] Step 1:

[0962] The device uses a camera to capture the user's facial expressions.

[0963] The device uses a microphone to collect the user's voice.

[0964] Step 2:

[0965] The facial expression data captured by the device and the collected voice data are converted into a certain format.

[0966] The device sends the converted data to the server, where it is encrypted for added security.

[0967] Step 3:

[0968] The server parses the data it receives.

[0969] The server uses image processing techniques to extract facial landmarks.

[0970] The server uses voice processing technology to extract features (pitch, volume, speaking speed, etc.) from the voice data.

[0971] Step 4:

[0972] The server uses an AI model to recognize the user's emotions from pre-processed data.

[0973] The server stores the recognized emotion data in a database.

[0974] Step 5:

[0975] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion.

[0976] The server generates a response by referencing pre-configured rules and case studies.

[0977] Step 6:

[0978] The server creates a text version of the generated response.

[0979] The server converts the text message into a voice message using voice synthesis technology as needed.

[0980] Step 7:

[0981] The terminal receives the response message from the server.

[0982] The terminal provides the response message to the user by displaying it on the screen or playing it as an audio message.

[0983] Step 8:

[0984] The terminal captures the user's reaction again after the response and collects feedback data based on this.

[0985] The terminal transmits the collected feedback data to the server.

[0986] Step 9:

[0987] The server analyzes the collected feedback data and updates the database to improve the accuracy of the AI ​​model.

[0988] The server uses this information to improve emotion recognition and response generation performance in future sessions.

[0989] Example 1

[0990] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0991] Conventional emotion recognition systems have had difficulty accurately recognizing users' emotions and providing appropriate real-time responses. They also lacked the feedback functionality to appropriately adjust responses in response to changes in the user's emotions. Furthermore, they lacked a mechanism for utilizing collected data to improve recognition accuracy in future sessions. This resulted in a decline in the quality of support provided to users, making it difficult to improve customer satisfaction.

[0992] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0993] In this invention, the server includes means for capturing images of a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to a central processing unit, means for recognizing emotions from the facial expression data and voice data in the central processing unit, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, and means for recollecting the user's reactions and using them to generate subsequent responses. This makes it possible to accurately recognize the user's emotions and provide appropriate real-time responses. Furthermore, by utilizing the collected data to improve recognition accuracy in subsequent responses, the quality of support provided to the user can be improved.

[0994] The "means for capturing the user's facial expression" refers to a device or software that uses a camera or image processing device to capture the user's facial expression.

[0995] The "means for collecting the user's voice" refers to a device or software that uses a microphone or voice collection device to record the user's speech or voice.

[0996] "Means for transmitting captured facial expression data and collected voice data to a central processing unit" refers to a device or software that uses a network communication function to transfer collected facial expression data and voice data to a central processing unit (server).

[0997] "Means for recognizing emotions from facial expression data and voice data in a central processing unit" refers to a device or software that executes an algorithm or artificial intelligence model to analyze the collected facial expression data and voice data and identify the user's emotional state.

[0998] The "means for generating an appropriate response based on the recognized emotion" is a program or algorithm for generating a response that is consistent with the identified emotional state.

[0999] The "means for providing the generated response to the user" is a device or software for displaying or audibly outputting the generated response to the user.

[1000] "Means for recollecting user reactions and using them to generate subsequent responses" refers to a device or software that recollects user reactions to the generated response and uses that data to generate the next response or to improve the accuracy of emotion recognition.

[1001] A "storage device" is hardware or software capable of storing data for long or short periods of time.

[1002] The system of this invention recognizes emotions using a user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses. An embodiment of the present invention will be described in detail below.

[1003] Facial expression and voice collection

[1004] The device uses a camera and microphone to collect the user's facial expressions and voice. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and is installed with OpenCV as facial recognition software and Google Speech API as voice collection software, making it possible to acquire data in real time.

[1005] Sending data

[1006] The device encrypts the collected facial expression and voice data and transmits it to a server via a network, using the Advanced Encryption Standard (AES) to ensure data security.

[1007] Data analysis

[1008] The server analyzes the received data, first using image processing technology (e.g., Dlib) to extract facial landmarks (eyes, nose, and mouth positions), and then using audio processing technology (e.g., Librosa) to analyze audio features (pitch, tone, speed, etc.).

[1009] emotion recognition

[1010] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[1011] Response Generation

[1012] The server generates an appropriate response based on the recognized emotion, which is generated in text format and optionally converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[1013] Providing a response

[1014] The device receives the response from the server and provides it to the user. In this case, the method of providing the response may be a screen display or audio playback (e.g., through the device's built-in speaker).

[1015] Collecting feedback

[1016] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected facial expression and voice data is also encrypted and sent.

[1017] Specific use cases

[1018] Example 1: Elderly care in nursing care settings

[1019] 1. An elderly person is relaxing in a care home. A device is set up on the table, with a camera and microphone capturing the elderly person's face and voice.

[1020] 2. The device collects the elderly person's tired facial expression and low-pitched voice.

[1021] 3. The device encrypts this data and sends it to the server.

[1022] 4. The server analyzes the data and extracts facial landmarks and audio features.

[1023] 5. The server's AI model recognizes emotions such as "tired" and "need to relax."

[1024] 6. The server generates the response "Play some relaxing music."

[1025] 7. The device receives the response from the server and plays relaxing music.

[1026] 8. The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[1027] Example 2: Managing stress while working remotely

[1028] 1. An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[1029] 2. The device collects audio of the employee's stern facial expression and fast-paced speech.

[1030] 3. The device encrypts this data and sends it to the server.

[1031] 4. The server analyzes the data and extracts facial landmarks and audio features.

[1032] 5. The server's AI model recognizes emotions such as "stress" and "anxiety."

[1033] 6. The server generates the response "Do some deep breathing exercises."

[1034] 7. The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[1035] 8. The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[1036] Prompt Sentence Examples

[1037] "Design a system that recognizes facial expressions (grim faces) and vocalizations (fast speech) that indicate stress in remote workers and suggests deep breathing exercises."

[1038] "Design a system that generates a response, such as playing relaxing music when an elderly person in a care home looks tired."

[1039] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1040] Step 1:

[1041] The device activates the camera and microphone to collect the user's facial expressions and voice in real time. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and has facial recognition software (e.g., OpenCV) and voice collection software (e.g., Google Speech API) installed. For example, when a user smiles and says "hello," their facial expressions and voice are collected.

[1042] Input: User's facial expressions and voice

[1043] Output: Collected facial expression data and voice data

[1044] Step 2:

[1045] The device encrypts the collected facial expression and voice data and sends it to the server via the network. At this time, the Advanced Encryption Standard (AES) is used to ensure data security. For example, a JSON file containing image data of the user's smile and the voice data of "hello" is generated and encrypted.

[1046] Input: Collected facial expression data and voice data

[1047] Output: Encrypted data

[1048] Step 3:

[1049] The server decrypts the received encrypted data, extracts facial landmarks (the positions of the eyes, nose, and mouth) using image processing technology (e.g., Dlib), and analyzes voice features (pitch, tone, speed, etc.) using voice processing technology (e.g., Librosa). For example, it can identify the degree of upturning of the corners of the eyes and the angle of the mouth from an image of the user's smile and analyze the tone of the voice saying "hello."

[1050] Input: Encrypted data

[1051] Output: Analyzed landmark data and audio feature data

[1052] Step 4:

[1053] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from the analyzed landmark data and voice feature data. This AI model operates based on a pre-trained dataset. For example, it identifies the emotion "joy" from the user's smiling face and bright tone of voice.

[1054] Input: Analyzed landmark data and audio feature data

[1055] Output: Recognized emotion data

[1056] Step 5:

[1057] The server generates an appropriate response based on the recognized emotion. This response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. For example, a response such as "Hello, you look well! How's your day?" can be generated and converted into a voice message.

[1058] Input: Recognized emotion data

[1059] Output: The text and audio data of the generated response

[1060] Step 6:

[1061] The device receives the response data from the server and provides it to the user. This response is output by a method such as displaying it on the screen or playing a voice message. For example, a voice message such as "Hello, you look well! How is your day?" is played from the device's speaker.

[1062] Input: The text and audio data of the generated response

[1063] Output: The response provided to the user

[1064] Step 7:

[1065] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected data is also encrypted and sent. For example, if the user smiles again or says, "Actually, I'm a little tired today," their facial expression and voice will be collected again, encrypted, and sent.

[1066] Input: Facial expression data and voice data based on the user's reaction

[1067] Output: Encrypted re-collected data

[1068] (Application example 1)

[1069] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1070] In brick-and-mortar stores, it is necessary to recognize customers' emotions in real time and provide appropriate responses and assistance. However, conventional systems have difficulty implementing specific responses or promotions based on customers' emotions, posing challenges in improving the quality of customer service and customer satisfaction. In addition, responses based on emotion recognition are one-way, and feedback is not used to improve the accuracy of subsequent responses.

[1071] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1072] In this invention, the server includes means for capturing a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to the server, means for recognizing emotions from the facial expression data and voice data, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, means for providing customer service assistance based on emotion recognition in a physical store, and means for playing a promotional video or offering a special discount in accordance with the generated response. This not only enables specific and appropriate responses based on the emotions of customers in a physical store, but also improves customer satisfaction. Furthermore, by storing the recognized emotion information in a database and improving the accuracy of emotion recognition from the next time onwards, the quality of customer service can be continuously improved.

[1073] A "user" is a person who uses the system.

[1074] "Facial expressions" are expressions that appear on the user's face and indicate their emotions and physical state.

[1075] "Voice" refers to voice data including the voice or speech of the user.

[1076] "Capture" refers to obtaining a user's facial expression and voice using devices such as a camera and microphone.

[1077] A "server" is a computing technology device that analyzes collected facial expression and voice data and generates appropriate responses.

[1078] "Emotion recognition" is the technical process of identifying a user's emotional state from collected facial expression and voice data.

[1079] A "response" is a reply or action to the user that is generated based on the recognized emotion.

[1080] "Brick and mortar store" means a physical store that directly offers goods or services.

[1081] The "customer service assistant" is an interactive system that provides appropriate service to users based on emotion recognition results.

[1082] A "promotional video" is video content used to promote products or services for commercial purposes.

[1083] "Special discounts" are discounted services that are lower than the usual price and are offered under certain conditions or circumstances.

[1084] A "database" is an information technology device for systematically storing and managing collected emotional information.

[1085] "Speech synthesis technology" is a technical method for converting text data into voice data.

[1086] The system for implementing this invention mainly consists of a terminal that collects facial expression and voice data, a server that analyzes this data, and a means for providing responses. The terminal that serves as the user interface is equipped with a camera and microphone, and can capture and collect the user's facial expression and voice in real time.

[1087] The device transmits facial expression data captured by the camera and audio data collected by the microphone to a server over the network, where the data is encrypted for security purposes before being transmitted.

[1088] The server then analyzes the received data. Specifically, it uses image processing technology to extract facial landmarks and audio processing technology to analyze audio features. This analysis process uses open source image processing and audio processing libraries (e.g., OpenCV and librosa).

[1089] On the server, a pre-trained generative AI model is used to recognize user emotions from facial expressions and voice data. The AI ​​model is trained on a wide range of emotion datasets and can identify emotions with high accuracy.

[1090] Based on the recognized emotion, the server generates an appropriate response, which may be generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[1091] The generated response is then sent back over the network to the terminal, which then provides the received response to the user. Typically, the response is displayed on the terminal's display or played as an audio message through a speaker.

[1092] Furthermore, in physical stores, customer service assistants can be provided based on emotion recognition. For example, if a user is recognized as interested, a promotional video for that product can be played. If a user is determined to be stressed, a special discount can be offered to encourage a purchase.

[1093] As a concrete example, consider a brick-and-mortar store promoting a new product. When a customer enters the store, a camera captures their facial expressions and collects audio data. If the system determines that the customer is interested (e.g., a smiling face, a lively voice), it plays a promotional video related to the product.

[1094] Use the following as an example prompt:

[1095] "What emotion is this customer experiencing? (Choices: 'happy', 'sad', 'neutral') Please determine this based on facial expression and voice data."

[1096] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1097] Step 1:

[1098] Collecting user facial expressions and voice

[1099] The device uses a camera and a microphone to collect the user's facial expression and voice data. As input, it captures the user's real-time video and audio. The camera captures the user's facial expression, and the microphone collects the voice. As output, it generates the captured facial expression data and collected voice data.

[1100] Step 2:

[1101] Sending data

[1102] The device transmits the collected facial expression data and voice data to a server via a network. The data is encrypted before transmission to ensure security. The input is facial expression data and voice data, and the output is a data packet received by the server.

[1103] Step 3:

[1104] Data analysis

[1105] The server analyzes the received facial expression data and audio data. It extracts facial landmarks using an image processing library (e.g., OpenCV) and analyzes audio features using an audio processing library (e.g., librosa). The inputs are facial expression data and audio data, and the analyzed facial landmarks and audio features are generated as outputs.

[1106] Step 4:

[1107] emotion recognition

[1108] The server uses a pre-trained generative AI model to recognize the user's emotions from the analyzed data. The inputs are facial landmarks and audio features, and the output is recognized emotion data. Specifically, the AI ​​model identifies emotions such as "happy," "sad," and "neutral."

[1109] Step 5:

[1110] Response Generation

[1111] The server generates an appropriate response based on the recognized emotion. The response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. The input is the recognized emotion data, and the output is a text response and a voice message.

[1112] Step 6:

[1113] Providing a response

[1114] The terminal provides the user with the response received from the server. The response is communicated to the user by a screen display or loudspeaker playback. The inputs are text responses and voice messages, and the output generates visual and auditory information that is presented to the user.

[1115] Step 7:

[1116] Collecting feedback

[1117] The device again collects the user's reaction after the response and sends the data to the server to help generate subsequent responses. The user's facial expression and voice data are again collected as input, and feedback data is generated as output.

[1118] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1119] The system of this invention combines an emotion engine to recognize a user's emotions at a high level and provide appropriate responses and support. A system using an emotion engine is primarily composed of a server and a terminal, and has comprehensive functions for improving the accuracy of emotion recognition.

[1120] Overall system configuration

[1121] 1. Collecting facial expressions and voice

[1122] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[1123] 2. Data transmission

[1124] The facial expression and voice data collected by the device is converted into a specific format and sent to a server, where it is encrypted to ensure security.

[1125] 3. Data Analysis and Emotion Recognition

[1126] The server analyzes the received data and recognizes the user's emotions using an emotion engine.

[1127] The emotion engine uses machine learning models to provide a fusion model for emotion recognition that integrates both facial expression data and voice data, enabling it to identify a user's emotional state with high accuracy.

[1128] 4. Response Generation

[1129] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[1130] 5. Providing a Response

[1131] The terminal receives the response from the server and provides it to the user, either by displaying it on the screen or by playing back audio.

[1132] 6. Gathering Feedback

[1133] After the response, the terminal collects the user's reaction again and sends the feedback data to the server to help generate subsequent responses.

[1134] Specific use cases

[1135] Example 1: Managing emotions of elderly people receiving home care

[1136] 1. Usage

[1137] An elderly person is receiving care at home. A device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[1138] 2. Processing Flow

[1139] The device collects the elderly person's facial expressions (e.g., depressed expressions) and voice data (slow speech).

[1140] The terminal transmits this data to the server.

[1141] The server analyzes the data and extracts facial landmarks and audio features.

[1142] The server's emotion engine recognizes emotions and gives results such as "depression" or "fatigue."

[1143] The server generates the response "Play some relaxing music."

[1144] The terminal receives the response from the server and plays relaxing music.

[1145] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[1146] Example 2: Managing student focus during remote learning

[1147] 1. Usage

[1148] A student is studying remotely, with a device set up on their desk, and a camera and microphone capturing their face and voice.

[1149] 2. Processing Flow

[1150] The device collects the student's facial expressions (e.g., expression of concentration) and voice data (clear pronunciation).

[1151] The terminal transmits this data to the server.

[1152] The server analyzes the data and extracts facial landmarks and audio features.

[1153] The server's emotion engine recognizes emotions and gives results such as "focused" or "interesting."

[1154] The server generates a response saying "Tell me when to take a break."

[1155] The device receives a response from the server and notifies the user by voice when it is time to take a break.

[1156] The student takes a break according to their response, and the device collects their response again and sends it to the server.

[1157] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and learning efficiency. This system is particularly useful for elderly people receiving care at home and students undergoing remote learning.

[1158] The processing flow will be explained below.

[1159] Step 1:

[1160] The device uses a camera to capture the user's facial expressions, and the facial recognition software installed on the device detects the contours and features of the face and stores them as image data.

[1161] Step 2:

[1162] The device uses a microphone to collect the user's voice, and the voice collection software installed on the device records the user's speech and saves it as audio data.

[1163] Step 3:

[1164] The device converts the captured facial expression data and collected audio data into a certain format (JPEG image, WAV audio file, etc.), which is then encrypted and prepared for transmission.

[1165] Step 4:

[1166] The device sends the encrypted facial expression and voice data to a server over a network, where data integrity and security are ensured.

[1167] Step 5:

[1168] The server analyzes the facial expression data it receives, and the image processing technology installed on the server extracts facial landmarks (the positions of the eyebrows, eyes, mouth, etc.) and detects their movements.

[1169] Step 6:

[1170] The server analyzes the received voice data and uses voice processing technology to extract voice features such as pitch, volume, and speaking rate.

[1171] Step 7:

[1172] The server uses an emotion engine to recognize the user's emotions from the pre-processed facial expression data and voice data. The emotion engine uses a machine learning model to fuse the data and achieve highly accurate emotion recognition.

[1173] Step 8:

[1174] The server stores the recognized emotion data in a database, which is then used as training data to improve the accuracy of future emotion recognition and response generation.

[1175] Step 9:

[1176] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion, which is generated by referencing pre-defined rules and case studies.

[1177] Step 10:

[1178] The server creates the generated response in text format and, if necessary, converts the text message into a voice message using speech synthesis technology.

[1179] Step 11:

[1180] The terminal receives the response message from the server, which is then decrypted and prepared for presentation to the user.

[1181] Step 12:

[1182] The device provides a response message to the user. Specifically, the response message is conveyed to the user through a screen display or audio playback. For example, a voice message saying, "Would you like to play some relaxing music?" is played.

[1183] Step 13:

[1184] The user acts on the response, for example, by following a suggestion from the device and starting to listen to relaxing music.

[1185] Step 14:

[1186] The device then recaptures the user's reaction after the response, collecting changes in facial expressions and voice and saving them as new data.

[1187] Step 15:

[1188] The device sends the newly collected feedback data to the server, where it is re-encrypted and sent over the network to the server.

[1189] Step 16:

[1190] The server analyzes the collected feedback data, which updates the learning database of the AI ​​model, improving the accuracy of emotion recognition and response generation from the next time onwards.

[1191] Example 2

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

[1193] Conventional technologies lack systems that consistently recognize a user's emotions and generate appropriate responses based on those emotions. Furthermore, even if a user's facial expressions and voice are collected, the data is transmitted and analyzed separately, resulting in an inefficient process. Furthermore, there are insufficient mechanisms for improving the accuracy of user emotion recognition and for incorporating the results of previous responses into feedback. This results in an inability to provide appropriate support to users, hindering improvements in their quality of life and learning efficiency.

[1194] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting the user's voice and facial expression, a means for transmitting the collected data, a means for performing data analysis and emotion recognition, a means for generating an appropriate response, a means for providing the generated response, and a means for collecting feedback data. This makes it possible to recognize the user's emotion with high accuracy and provide appropriate support in real time based on the recognition result.

[1195] "User" refers to any individual or entity that uses this system.

[1196] "Means for capturing facial expressions" refers to a device or software for capturing a user's facial expressions as images or videos.

[1197] "Audio collection means" refers to any device or software that records a user's speech or voice.

[1198] "Means for converting into a data format" refers to software for converting the collected facial expression data and voice data into a format that can be sent to a server (e.g., JSON format).

[1199] "Means for encryption and transmission" refers to a method for transmitting data to a server using TLS encryption to ensure data security.

[1200] "Server" refers to a centralized computing device that analyzes collected data, recognizes emotions, and generates responses.

[1201] "Means for recognizing emotions using machine learning models" refers to algorithms that use machine learning frameworks such as TensorFlow and PyTorch to analyze a user's facial expressions and voice data to identify emotions.

[1202] "Means for generating an appropriate response" refers to software for generating feedback to the user in the form of a text or voice message based on the recognized emotion.

[1203] "Means for generating a voice message using speech synthesis technology" refers to technology (e.g., speech synthesis software) for converting the generated text response into speech format.

[1204] "Means for providing a response" refers to an interface (such as a display or speaker) for providing the generated text or audio response to the user.

[1205] "Means for collecting feedback data" refers to a method for re-collecting user reactions to the generated responses and using that data to generate future responses and improve the accuracy of emotion recognition.

[1206] This invention relates to a system for recognizing a user's emotions and providing an appropriate response. This system, which is composed of a terminal and a server, improves the accuracy of emotion recognition and provides appropriate support to the user in real time.

[1207] Hardware and software used

[1208] 1. Terminal

[1209] Camera: Used to capture the user's facial expressions. Specifically, a common webcam (e.g., Logitech C920) is used.

[1210] Microphone: Used to collect the user's voice. Specifically, a general USB microphone (e.g., Blue Yeti) is used.

[1211] Software: Utilizes OpenCV and Google Cloud Speech-to-Text APIs for facial recognition and voice data collection.

[1212] 2. Server

[1213] Data analysis: The server uses machine learning frameworks such as TensorFlow and PyTorch to process facial expression and voice data.

[1214] Emotion Recognition: Based on data analysis, a fusion model is used to recognize user emotions with high accuracy.

[1215] Response generation: Generates appropriate responses based on the recognized emotions and generates voice messages using text-to-speech technology (e.g., Amazon Polly) if necessary.

[1216] System operation procedures and examples

[1217] The system works as follows: the device collects the user's facial expressions and voice and sends the data to the server. The server analyzes the data and recognizes the user's emotions. Based on the recognized emotions, the server generates an appropriate response and provides it to the user again via the device. The device then collects the user's feedback and sends it to the server to help improve accuracy next time.

[1218] Specific use cases:

[1219] Example 1: Managing emotions of elderly people receiving home care

[1220] Usage scenario: An elderly person is receiving care at home. The device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[1221] Operational flow: The device collects the elderly person's facial expressions (e.g., a depressed expression) and voice data (speaking slowly) and sends them to the server. The server analyzes the data and recognizes emotions such as "depression" and "fatigue." The server generates a response such as "play relaxing music," and the device plays the music. The elderly person relaxes while listening to the music, and the device collects their responses again and sends them to the server.

[1222] Example 2: Managing student focus during remote learning

[1223] Usage scenario: A student is learning remotely, with the device set up at a desk and the camera and microphone capturing the student's face and voice.

[1224] Operational flow: The device collects the student's facial expressions (e.g., a concentrated expression) and voice data (clear pronunciation) and sends them to the server. The server analyzes the data and recognizes emotions such as "concentration" and "interested." The server generates a response saying "Notify me when it's time to take a break," and the device notifies the student by voice. The student takes an appropriate break, and their response is collected again and sent to the server.

[1225] Prompt Sentence Examples

[1226] Below are some example prompts to be input to the generative AI model:

[1227] "How can we provide appropriate support to elderly people who are depressed while receiving care at home?"

[1228] "What approaches are effective in helping students stay focused while learning remotely?"

[1229] In this way, it is possible to provide appropriate support in real time, especially for elderly people receiving care at home or students undergoing remote learning. This system is extremely effective in improving users' quality of life and learning efficiency.

[1230] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1231] Step 1:

[1232] Facial expression and voice data collection

[1233] The device uses a camera and microphone to collect the user's facial expressions and voice in real time.

[1234] Input: Image data from the camera and audio data from the microphone.

[1235] Processing: Use OpenCV for facial recognition and Google Cloud Speech-to-Text to convert audio data to text.

[1236] Output: Facial expression data (e.g. facial landmark points) and speech text data.

[1237] Specific operation: The camera captures the user's face, and the microphone records what the user says. For example, it collects audio of an elderly person saying, "I'm not feeling well today."

[1238] Step 2:

[1239] Data preprocessing and transmission

[1240] The facial expression data and voice data collected by the device are converted into a fixed data format (JSON format) and sent to the server using TLS encryption.

[1241] Input: Facial expression data, speech text data.

[1242] Processing: Data conversion and TLS encryption.

[1243] Output: A JSON file containing encrypted facial expression and audio data.

[1244] Specific operation: The device converts facial expression data and voice text into JSON format, encrypts it with TLS, and sends it to the server. For example, it contains data such as "Facial landmarks show that the corners of the mouth are down."

[1245] Step 3:

[1246] Data analysis and emotion recognition

[1247] The server analyzes the received data and uses machine learning models to recognize emotions. Specifically, it uses TensorFlow to process facial expression data and voice data.

[1248] Input: Encrypted facial expression and voice data in JSON format.

[1249] Processing: Decryption, data analysis, and emotion recognition using machine learning models.

[1250] Output: Perceived emotion (e.g., "depressed," "tired").

[1251] Specific operation: The server receives the data, extracts facial landmarks and audio features, and inputs them into the fusion model to identify emotions. For example, it extracts features such as "facial landmarks with downturned corners of the mouth" and "negative audio content" and recognizes the emotion as "depressed."

[1252] Step 4:

[1253] Response Generation

[1254] The server generates an appropriate response based on the recognized emotion and creates a voice message using speech synthesis technology.

[1255] Input: Perceived emotion (e.g., "depressed," "tired").

[1256] Processing: Response text generation and speech synthesis.

[1257] Output: Response text and voice message.

[1258] Specific operation: The server generates the text "Playing relaxing music" and synthesizes it to create a voice message.

[1259] Step 5:

[1260] Providing a response

[1261] The device receives the response from the server and provides it to the user, for example by displaying text on the display or playing an audio message over the speaker.

[1262] Input: Response text and voice message.

[1263] Action: Display text or play audio.

[1264] Output: Feedback provided to the user.

[1265] Specific operation: The device will play a voice message saying "Playing relaxing music" and then play relaxing music.

[1266] Step 6:

[1267] Collecting feedback

[1268] The device collects the user's responses and sends them back to the server to help improve accuracy next time.

[1269] Input: User reaction data (facial expressions and voice).

[1270] Processing: Collect and send data again.

[1271] Output: Feedback data.

[1272] Specific operation: The system collects data on whether the elderly person is relaxed while listening to music or if they have positive reactions (e.g., saying "thank you"), and sends the data to a server. For example, this data includes facial landmarks indicating a smile and audio data saying "thank you."

[1273] (Application example 2)

[1274] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1275] In traditional brick-and-mortar stores, it was difficult to immediately grasp customer emotions and reactions and respond accordingly. As a result, it was not possible to provide services that satisfied customers, and the effect of increasing purchasing motivation was limited. It was also difficult to collect feedback based on customer reactions, and this could not be used to improve services.

[1276] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for capturing the user's facial expression, means for collecting the user's voice, and means for transmitting the captured facial expression data and collected voice data to the server. This makes it possible to recognize customer emotions in real time in a physical store and provide appropriate product recommendations or customer service. Furthermore, the customer's reactions can be collected again and transmitted to the server as feedback data, thereby improving the accuracy of emotion recognition from the next time onwards.

[1277] The "means for capturing a user's facial expression" is a device or system that uses a camera or sensor to acquire image data of a user's facial expression.

[1278] The "means for collecting user's voice" refers to a device or system that uses a microphone or a voice recognition device to acquire the user's speech as voice data.

[1279] The "means for transmitting the captured facial expression data and collected voice data to a server" is a communication device or protocol for sending the facial expression data and voice data acquired over a network to a remote server.

[1280] The "means for recognizing emotions from facial expression data and voice data on a server" refers to a technology that uses software and machine learning models installed on the server to analyze received facial expression data and voice data and identify the user's emotions.

[1281] The "means for generating an appropriate response based on the recognized emotion" refers to an algorithm or software that allows the server to create an optimal text or voice message based on the user's emotional information.

[1282] The "means for providing the generated response to the user" refers to a device or system that transmits the generated response message to the user via a display or speaker.

[1283] "A means for recognizing customer emotions in real time in physical stores and making appropriate product suggestions or customer service responses" refers to software and hardware that utilizes emotion recognition technology to dynamically make product suggestions and customer service responses that correspond to the customer's emotional state.

[1284] The "means for recollecting customer reactions and sending them to the server as feedback data" refers to a device or protocol for recollecting the customer's facial expressions and reactions after the interaction and sending that data to an analysis server.

[1285] The system of the present invention recognizes customer emotions at a high level and provides appropriate product suggestions and customer service in a physical store. A specific implementation method thereof will be described in detail below.

[1286] Hardware Configuration

[1287] This system mainly consists of the following hardware:

[1288] 1. Smart glasses or in-store robots: Used to collect customers' facial expressions and voice.

[1289] 2. Camera: Captures the user's facial expressions.

[1290] 3. Microphone: Collects the user's voice.

[1291] 4. Server: Performs data analysis and emotion recognition.

[1292] Software Configuration

[1293] The system includes the following software components:

[1294] 1. OpenCV: A library for collecting and processing facial expressions.

[1295] 2. requests: HTTP request library for sending data.

[1296] 3. Emotion Engine: A machine learning model to analyze the received data and recognize emotions.

[1297] Data processing and calculation

[1298] The system processing flow is as follows:

[1299] 1. Data collection: Using cameras and microphones on smart glasses or robots, customers' facial expressions and voice data are collected in real time, allowing instantaneous acquisition of customer emotional information.

[1300] 2. Data transmission: The collected facial expression and voice data is encrypted and transmitted from the device to the server, ensuring data security.

[1301] 3. Data Analysis: The server analyzes the received data. Specifically, it uses OpenCV to process facial expression data and uses an emotion engine to recognize emotions. The voice data is processed in the same way, and the data is integrated to identify emotions.

[1302] 4. Response Generation: The server generates appropriate product suggestions and customer service responses based on the recognized emotion data. Responses are created as text or voice messages.

[1303] 5. Response provision: The generated response data is provided to the customer through smart glasses or a robot, thereby enhancing customer interaction.

[1304] 6. Feedback collection: After responding, the system collects the customer's reaction again and sends the data to the server, providing feedback to improve the accuracy of emotion recognition in the future.

[1305] Specific examples

[1306] As a specific example, the following scenario can be considered.

[1307] 1. A customer enters the store: A staff member wearing smart glasses approaches the customer and collects facial expressions and voice recordings.

[1308] 2. Data analysis: The collected data is sent to the server, which analyzes it using an emotion engine. If the result indicates that the customer is interested, the server generates appropriate product suggestions.

[1309] 3. Product proposals and responses: The generated product proposals are transmitted to the staff via smart glasses, who then make the proposals to the customer. The customer's responses are collected again and sent to the server as feedback.

[1310] Prompt Sentence Examples

[1311] Examples of prompts for generative AI models include:

[1312] "Please generate a program for a system that analyzes customer facial expressions and voice data, recognizes emotions, and provides appropriate product suggestions or customer service responses to the customer."

[1313] As described above, this invention utilizes emotion recognition technology to improve the quality of service in brick-and-mortar stores and increase customer satisfaction.

[1314] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1315] Step 1:

[1316] The device uses smart glasses or in-store robots to collect customers' facial expressions and voice in real time.

[1317] Input: Customer's facial expression (video data), voice (audio data)

[1318] How it works: The device's camera captures facial expressions, and the microphone collects audio. The facial expressions are captured as image data using the OpenCV library, and the audio is captured as audio data using a voice recognition device.

[1319] Output: Acquired facial expression data and voice data

[1320] Step 2:

[1321] The facial expression data and voice data collected by the terminal are transmitted to a server.

[1322] Input: Facial expression data, voice data

[1323] How it works: The device encrypts facial expression and voice data and sends it securely to the server using HTTP requests, with the data sent in JSON format.

[1324] Output: Facial expression data and voice data sent to the server

[1325] Step 3:

[1326] The server analyzes the received facial expression data and voice data and recognizes emotions using an emotion engine.

[1327] Input: Facial expression data and voice data sent to the server

[1328] How it works: The server uses OpenCV to analyze facial expression data and extract facial landmarks. It also extracts voice features from voice data. The emotion engine integrates these data and uses machine learning models to recognize emotions.

[1329] Output: Recognized emotion data (e.g., joy, interest, anxiety, etc.)

[1330] Step 4:

[1331] The server generates an appropriate response based on the recognized emotion data.

[1332] Input: Recognized emotion data

[1333] How it works: The server uses a response generation algorithm to generate the optimal response (e.g., product suggestions, customer service messages, etc.) based on the emotion. Responses are constructed as text or voice messages.

[1334] Output: Generated response data (text or audio)

[1335] Step 5:

[1336] The terminal receives the response data from the server and provides it to the client.

[1337] Input: Generated response data

[1338] Operation: The device processes the received response data and displays it on the smart glasses display or plays it aloud through the robot's speaker.

[1339] Output: The response (visual or audio message) provided to the customer

[1340] Step 6:

[1341] The terminal again collects the customer's response and sends it to the server as feedback data.

[1342] Input: New customer facial expression data, voice data

[1343] How it works: The device again collects the customer's facial expressions and voice and sends them to the server. Based on this data, the server performs feedback analysis to improve the accuracy of the emotion recognition model.

[1344] Output: Feedback data (facial expression and voice data) sent to the server

[1345] Through these steps, the system can recognize customer emotions in real time and significantly improve the quality of service in physical stores.

[1346] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1347] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1348] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1349] [Fourth embodiment]

[1350] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1351] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1352] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1353] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1354] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1355] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1356] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1357] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1358] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1359] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1360] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1361] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1362] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1363] The system of this invention recognizes emotions from the user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses.

[1364] Overall system configuration

[1365] 1. Collecting facial expressions and voice

[1366] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[1367] 2. Data transmission

[1368] The facial expression and voice data collected by the device is sent to a server via a network, where it is encrypted for security purposes.

[1369] 3. Data Analysis

[1370] The server analyzes the received data, extracting facial landmarks using image processing technology and analyzing voice features using voice processing technology, thereby identifying the user's emotions from facial expressions and tone of voice.

[1371] 4. Emotion recognition

[1372] The server uses an AI model to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[1373] 5. Response Generation

[1374] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[1375] 6. Providing a Response

[1376] The terminal receives the response from the server and provides it to the user, possibly via a screen display or audio playback.

[1377] 7. Gathering Feedback

[1378] After responding, the terminal collects the user's reaction again and sends it to the server to help generate subsequent responses.

[1379] Specific use cases

[1380] Example 1: Caring for the elderly in nursing care settings

[1381] 1. Usage

[1382] An elderly person is relaxing in a care home with a device set up on a table, its camera and microphone capturing the elderly person's face and voice.

[1383] 2. Processing Flow

[1384] The device collects the elderly person's facial expressions (e.g., tired expression) and voice data (low tone of voice).

[1385] The terminal transmits this data to the server.

[1386] The server analyzes the data and extracts facial landmarks and audio features.

[1387] The server's AI model recognizes emotions and gives results such as "tired" or "needs to relax."

[1388] The server generates the response "Play some relaxing music."

[1389] The terminal receives the response from the server and plays relaxing music.

[1390] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[1391] Example 2: Managing stress while working remotely

[1392] 1. Usage

[1393] An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[1394] 2. Processing Flow

[1395] The device collects the employee's facial expressions (e.g., stern expression) and voice data (fast speaking tone).

[1396] The terminal transmits this data to the server.

[1397] The server analyzes the data and extracts facial landmarks and audio features.

[1398] The server's AI model recognizes emotions and gives results such as "stress" or "anxiety."

[1399] The server generates the response "Do some deep breathing exercises."

[1400] The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[1401] The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[1402] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and work efficiency. This system is particularly useful in nursing homes and remote work environments.

[1403] The processing flow will be explained below.

[1404] Step 1:

[1405] The device uses a camera to capture the user's facial expressions.

[1406] The device uses a microphone to collect the user's voice.

[1407] Step 2:

[1408] The facial expression data captured by the device and the collected voice data are converted into a certain format.

[1409] The device sends the converted data to the server, where it is encrypted for added security.

[1410] Step 3:

[1411] The server parses the data it receives.

[1412] The server uses image processing techniques to extract facial landmarks.

[1413] The server uses voice processing technology to extract features (pitch, volume, speaking speed, etc.) from the voice data.

[1414] Step 4:

[1415] The server uses an AI model to recognize the user's emotions from pre-processed data.

[1416] The server stores the recognized emotion data in a database.

[1417] Step 5:

[1418] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion.

[1419] The server generates a response by referencing pre-configured rules and case studies.

[1420] Step 6:

[1421] The server creates a text version of the generated response.

[1422] The server converts the text message into a voice message using voice synthesis technology as needed.

[1423] Step 7:

[1424] The terminal receives the response message from the server.

[1425] The terminal provides the response message to the user by displaying it on the screen or playing it as an audio message.

[1426] Step 8:

[1427] The terminal captures the user's reaction again after the response and collects feedback data based on this.

[1428] The terminal transmits the collected feedback data to the server.

[1429] Step 9:

[1430] The server analyzes the collected feedback data and updates the database to improve the accuracy of the AI ​​model.

[1431] The server uses this information to improve emotion recognition and response generation performance in future sessions.

[1432] Example 1

[1433] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1434] Conventional emotion recognition systems have had difficulty accurately recognizing users' emotions and providing appropriate real-time responses. They also lacked the feedback functionality to appropriately adjust responses in response to changes in the user's emotions. Furthermore, they lacked a mechanism for utilizing collected data to improve recognition accuracy in future sessions. This resulted in a decline in the quality of support provided to users, making it difficult to improve customer satisfaction.

[1435] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1436] In this invention, the server includes means for capturing images of a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to a central processing unit, means for recognizing emotions from the facial expression data and voice data in the central processing unit, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, and means for recollecting the user's reactions and using them to generate subsequent responses. This makes it possible to accurately recognize the user's emotions and provide appropriate real-time responses. Furthermore, by utilizing the collected data to improve recognition accuracy in subsequent responses, the quality of support provided to the user can be improved.

[1437] The "means for capturing the user's facial expression" refers to a device or software that uses a camera or image processing device to capture the user's facial expression.

[1438] The "means for collecting the user's voice" refers to a device or software that uses a microphone or voice collection device to record the user's speech or voice.

[1439] "Means for transmitting captured facial expression data and collected voice data to a central processing unit" refers to a device or software that uses a network communication function to transfer collected facial expression data and voice data to a central processing unit (server).

[1440] "Means for recognizing emotions from facial expression data and voice data in a central processing unit" refers to a device or software that executes an algorithm or artificial intelligence model to analyze the collected facial expression data and voice data and identify the user's emotional state.

[1441] The "means for generating an appropriate response based on the recognized emotion" is a program or algorithm for generating a response that is consistent with the identified emotional state.

[1442] The "means for providing the generated response to the user" is a device or software for displaying or audibly outputting the generated response to the user.

[1443] "Means for recollecting user reactions and using them to generate subsequent responses" refers to a device or software that recollects user reactions to the generated response and uses that data to generate the next response or to improve the accuracy of emotion recognition.

[1444] A "storage device" is hardware or software capable of storing data for long or short periods of time.

[1445] The system of this invention recognizes emotions using a user's facial expressions and voice, and provides appropriate responses and support. The system is mainly composed of a server and a terminal, and the terminal with which the user directly interfaces works in conjunction with the server that analyzes data and generates responses. An embodiment of the present invention will be described in detail below.

[1446] Facial expression and voice collection

[1447] The device uses a camera and microphone to collect the user's facial expressions and voice. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and is installed with OpenCV as facial recognition software and Google Speech API as voice collection software, making it possible to acquire data in real time.

[1448] Sending data

[1449] The device encrypts the collected facial expression and voice data and transmits it to a server via a network, using the Advanced Encryption Standard (AES) to ensure data security.

[1450] Data analysis

[1451] The server analyzes the received data, first using image processing technology (e.g., Dlib) to extract facial landmarks (eyes, nose, and mouth positions), and then using audio processing technology (e.g., Librosa) to analyze audio features (pitch, tone, speed, etc.).

[1452] emotion recognition

[1453] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from facial expressions and voice data, which operates on a pre-trained dataset.

[1454] Response Generation

[1455] The server generates an appropriate response based on the recognized emotion, which is generated in text format and optionally converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[1456] Providing a response

[1457] The device receives the response from the server and provides it to the user. In this case, the method of providing the response may be a screen display or audio playback (e.g., through the device's built-in speaker).

[1458] Collecting feedback

[1459] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected facial expression and voice data is also encrypted and sent.

[1460] Specific use cases

[1461] Example 1: Elderly care in nursing care settings

[1462] 1. An elderly person is relaxing in a care home. A device is set up on the table, with a camera and microphone capturing the elderly person's face and voice.

[1463] 2. The device collects the elderly person's tired facial expression and low-pitched voice.

[1464] 3. The device encrypts this data and sends it to the server.

[1465] 4. The server analyzes the data and extracts facial landmarks and audio features.

[1466] 5. The server's AI model recognizes emotions such as "tired" and "need to relax."

[1467] 6. The server generates the response "Play some relaxing music."

[1468] 7. The device receives the response from the server and plays relaxing music.

[1469] 8. The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[1470] Example 2: Managing stress while working remotely

[1471] 1. An employee working from home is working in front of a computer, which is set up on a desktop and has a camera and microphone capturing the employee's face and voice.

[1472] 2. The device collects audio of the employee's stern facial expression and fast-paced speech.

[1473] 3. The device encrypts this data and sends it to the server.

[1474] 4. The server analyzes the data and extracts facial landmarks and audio features.

[1475] 5. The server's AI model recognizes emotions such as "stress" and "anxiety."

[1476] 6. The server generates the response "Do some deep breathing exercises."

[1477] 7. The device receives a response from the server and provides audio instructions on how to perform deep breathing exercises.

[1478] 8. The employee follows the instructions to take a deep breath, and the device again collects their response and sends it to the server.

[1479] Prompt Sentence Examples

[1480] "Design a system that recognizes facial expressions (grim faces) and vocalizations (fast speech) that indicate stress in remote workers and suggests deep breathing exercises."

[1481] "Design a system that generates a response, such as playing relaxing music when an elderly person in a care home looks tired."

[1482] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1483] Step 1:

[1484] The device activates the camera and microphone to collect the user's facial expressions and voice in real time. The device is equipped with a high-resolution camera and a high-sensitivity microphone, and has facial recognition software (e.g., OpenCV) and voice collection software (e.g., Google Speech API) installed. For example, when a user smiles and says "hello," their facial expressions and voice are collected.

[1485] Input: User's facial expressions and voice

[1486] Output: Collected facial expression data and voice data

[1487] Step 2:

[1488] The device encrypts the collected facial expression and voice data and sends it to the server via the network. At this time, the Advanced Encryption Standard (AES) is used to ensure data security. For example, a JSON file containing image data of the user's smile and the voice data of "hello" is generated and encrypted.

[1489] Input: Collected facial expression data and voice data

[1490] Output: Encrypted data

[1491] Step 3:

[1492] The server decrypts the received encrypted data, extracts facial landmarks (the positions of the eyes, nose, and mouth) using image processing technology (e.g., Dlib), and analyzes voice features (pitch, tone, speed, etc.) using voice processing technology (e.g., Librosa). For example, it can identify the degree of upturning of the corners of the eyes and the angle of the mouth from an image of the user's smile and analyze the tone of the voice saying "hello."

[1493] Input: Encrypted data

[1494] Output: Analyzed landmark data and audio feature data

[1495] Step 4:

[1496] The server uses an AI model (e.g., a convolutional neural network using TensorFlow) to recognize the user's emotions from the analyzed landmark data and voice feature data. This AI model operates based on a pre-trained dataset. For example, it identifies the emotion "joy" from the user's smiling face and bright tone of voice.

[1497] Input: Analyzed landmark data and audio feature data

[1498] Output: Recognized emotion data

[1499] Step 5:

[1500] The server generates an appropriate response based on the recognized emotion. This response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. For example, a response such as "Hello, you look well! How's your day?" can be generated and converted into a voice message.

[1501] Input: Recognized emotion data

[1502] Output: The text and audio data of the generated response

[1503] Step 6:

[1504] The device receives the response data from the server and provides it to the user. This response is output by a method such as displaying it on the screen or playing a voice message. For example, a voice message such as "Hello, you look well! How is your day?" is played from the device's speaker.

[1505] Input: The text and audio data of the generated response

[1506] Output: The response provided to the user

[1507] Step 7:

[1508] After responding, the device collects the user's reaction again and sends it to the server to help generate future responses. The newly collected data is also encrypted and sent. For example, if the user smiles again or says, "Actually, I'm a little tired today," their facial expression and voice will be collected again, encrypted, and sent.

[1509] Input: Facial expression data and voice data based on the user's reaction

[1510] Output: Encrypted re-collected data

[1511] (Application example 1)

[1512] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1513] In brick-and-mortar stores, it is necessary to recognize customers' emotions in real time and provide appropriate responses and assistance. However, conventional systems have difficulty implementing specific responses or promotions based on customers' emotions, posing challenges in improving the quality of customer service and customer satisfaction. In addition, responses based on emotion recognition are one-way, and feedback is not used to improve the accuracy of subsequent responses.

[1514] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1515] In this invention, the server includes means for capturing a user's facial expression, means for collecting the user's voice, means for transmitting the captured facial expression data and collected voice data to the server, means for recognizing emotions from the facial expression data and voice data, means for generating an appropriate response based on the recognized emotion, means for providing the generated response to the user, means for providing customer service assistance based on emotion recognition in a physical store, and means for playing a promotional video or offering a special discount in accordance with the generated response. This not only enables specific and appropriate responses based on the emotions of customers in a physical store, but also improves customer satisfaction. Furthermore, by storing the recognized emotion information in a database and improving the accuracy of emotion recognition from the next time onwards, the quality of customer service can be continuously improved.

[1516] A "user" is a person who uses the system.

[1517] "Facial expressions" are expressions that appear on the user's face and indicate their emotions and physical state.

[1518] "Voice" refers to voice data including the voice or speech of the user.

[1519] "Capture" refers to obtaining a user's facial expression and voice using devices such as a camera and microphone.

[1520] A "server" is a computing technology device that analyzes collected facial expression and voice data and generates appropriate responses.

[1521] "Emotion recognition" is the technical process of identifying a user's emotional state from collected facial expression and voice data.

[1522] A "response" is a reply or action to the user that is generated based on the recognized emotion.

[1523] "Brick and mortar store" means a physical store that directly offers goods or services.

[1524] The "customer service assistant" is an interactive system that provides appropriate service to users based on emotion recognition results.

[1525] A "promotional video" is video content used to promote products or services for commercial purposes.

[1526] "Special discounts" are discounted services that are lower than the usual price and are offered under certain conditions or circumstances.

[1527] A "database" is an information technology device for systematically storing and managing collected emotional information.

[1528] "Speech synthesis technology" is a technical method for converting text data into voice data.

[1529] The system for implementing this invention mainly consists of a terminal that collects facial expression and voice data, a server that analyzes this data, and a means for providing responses. The terminal that serves as the user interface is equipped with a camera and microphone, and can capture and collect the user's facial expression and voice in real time.

[1530] The device transmits facial expression data captured by the camera and audio data collected by the microphone to a server over the network, where the data is encrypted for security purposes before being transmitted.

[1531] The server then analyzes the received data. Specifically, it uses image processing technology to extract facial landmarks and audio processing technology to analyze audio features. This analysis process uses open source image processing and audio processing libraries (e.g., OpenCV and librosa).

[1532] On the server, a pre-trained generative AI model is used to recognize user emotions from facial expressions and voice data. The AI ​​model is trained on a wide range of emotion datasets and can identify emotions with high accuracy.

[1533] Based on the recognized emotion, the server generates an appropriate response, which may be generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API).

[1534] The generated response is then sent back over the network to the terminal, which then provides the received response to the user. Typically, the response is displayed on the terminal's display or played as an audio message through a speaker.

[1535] Furthermore, in physical stores, customer service assistants can be provided based on emotion recognition. For example, if a user is recognized as interested, a promotional video for that product can be played. If a user is determined to be stressed, a special discount can be offered to encourage a purchase.

[1536] As a concrete example, consider a brick-and-mortar store promoting a new product. When a customer enters the store, a camera captures their facial expressions and collects audio data. If the system determines that the customer is interested (e.g., a smiling face, a lively voice), it plays a promotional video related to the product.

[1537] Use the following as an example prompt:

[1538] "What emotion is this customer experiencing? (Choices: 'happy', 'sad', 'neutral') Please determine this based on facial expression and voice data."

[1539] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1540] Step 1:

[1541] Collecting user facial expressions and voice

[1542] The device uses a camera and a microphone to collect the user's facial expression and voice data. As input, it captures the user's real-time video and audio. The camera captures the user's facial expression, and the microphone collects the voice. As output, it generates the captured facial expression data and collected voice data.

[1543] Step 2:

[1544] Sending data

[1545] The device transmits the collected facial expression data and voice data to a server via a network. The data is encrypted before transmission to ensure security. The input is facial expression data and voice data, and the output is a data packet received by the server.

[1546] Step 3:

[1547] Data analysis

[1548] The server analyzes the received facial expression data and audio data. It extracts facial landmarks using an image processing library (e.g., OpenCV) and analyzes audio features using an audio processing library (e.g., librosa). The inputs are facial expression data and audio data, and the analyzed facial landmarks and audio features are generated as outputs.

[1549] Step 4:

[1550] emotion recognition

[1551] The server uses a pre-trained generative AI model to recognize the user's emotions from the analyzed data. The inputs are facial landmarks and audio features, and the output is recognized emotion data. Specifically, the AI ​​model identifies emotions such as "happy," "sad," and "neutral."

[1552] Step 5:

[1553] Response Generation

[1554] The server generates an appropriate response based on the recognized emotion. The response is generated in text format and converted into a voice message using speech synthesis technology (e.g., Google Text-to-Speech API) if necessary. The input is the recognized emotion data, and the output is a text response and a voice message.

[1555] Step 6:

[1556] Providing a response

[1557] The terminal provides the user with the response received from the server. The response is communicated to the user by a screen display or loudspeaker playback. The inputs are text responses and voice messages, and the output generates visual and auditory information that is presented to the user.

[1558] Step 7:

[1559] Collecting feedback

[1560] The device again collects the user's reaction after the response and sends the data to the server to help generate subsequent responses. The user's facial expression and voice data are again collected as input, and feedback data is generated as output.

[1561] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1562] The system of this invention combines an emotion engine to recognize a user's emotions at a high level and provide appropriate responses and support. A system using an emotion engine is primarily composed of a server and a terminal, and has comprehensive functions for improving the accuracy of emotion recognition.

[1563] Overall system configuration

[1564] 1. Collecting facial expressions and voice

[1565] The device uses a camera and microphone to capture the user's facial expressions and voice. The device is equipped with facial recognition software and voice capture software, which are used to obtain data in real time.

[1566] 2. Data transmission

[1567] The facial expression and voice data collected by the device is converted into a specific format and sent to a server, where it is encrypted to ensure security.

[1568] 3. Data Analysis and Emotion Recognition

[1569] The server analyzes the received data and recognizes the user's emotions using an emotion engine.

[1570] The emotion engine uses machine learning models to provide a fusion model for emotion recognition that integrates both facial expression data and voice data, enabling it to identify a user's emotional state with high accuracy.

[1571] 4. Response Generation

[1572] The server generates an appropriate response based on the recognized emotion, which is generated in text format and converted into a voice message using speech synthesis technology if necessary.

[1573] 5. Providing a Response

[1574] The terminal receives the response from the server and provides it to the user, either by displaying it on the screen or by playing back audio.

[1575] 6. Gathering Feedback

[1576] After the response, the terminal collects the user's reaction again and sends the feedback data to the server to help generate subsequent responses.

[1577] Specific use cases

[1578] Example 1: Managing emotions of elderly people receiving home care

[1579] 1. Usage

[1580] An elderly person is receiving care at home. A device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[1581] 2. Processing Flow

[1582] The device collects the elderly person's facial expressions (e.g., depressed expressions) and voice data (slow speech).

[1583] The terminal transmits this data to the server.

[1584] The server analyzes the data and extracts facial landmarks and audio features.

[1585] The server's emotion engine recognizes emotions and gives results such as "depression" or "fatigue."

[1586] The server generates the response "Play some relaxing music."

[1587] The terminal receives the response from the server and plays relaxing music.

[1588] The elderly person relaxes while listening to music, and the device again collects their reactions and sends them to the server.

[1589] Example 2: Managing student focus during remote learning

[1590] 1. Usage

[1591] A student is studying remotely, with a device set up on their desk, and a camera and microphone capturing their face and voice.

[1592] 2. Processing Flow

[1593] The device collects the student's facial expressions (e.g., expression of concentration) and voice data (clear pronunciation).

[1594] The terminal transmits this data to the server.

[1595] The server analyzes the data and extracts facial landmarks and audio features.

[1596] The server's emotion engine recognizes emotions and gives results such as "focused" or "interesting."

[1597] The server generates a response saying "Tell me when to take a break."

[1598] The device receives a response from the server and notifies the user by voice when it is time to take a break.

[1599] The student takes a break according to their response, and the device collects their response again and sends it to the server.

[1600] These embodiments allow users to receive appropriate support in real time according to their emotional state, improving their quality of life and learning efficiency. This system is particularly useful for elderly people receiving care at home and students undergoing remote learning.

[1601] The processing flow will be explained below.

[1602] Step 1:

[1603] The device uses a camera to capture the user's facial expressions, and the facial recognition software installed on the device detects the contours and features of the face and stores them as image data.

[1604] Step 2:

[1605] The device uses a microphone to collect the user's voice, and the voice collection software installed on the device records the user's speech and saves it as audio data.

[1606] Step 3:

[1607] The device converts the captured facial expression data and collected audio data into a certain format (JPEG image, WAV audio file, etc.), which is then encrypted and prepared for transmission.

[1608] Step 4:

[1609] The device sends the encrypted facial expression and voice data to a server over a network, where data integrity and security are ensured.

[1610] Step 5:

[1611] The server analyzes the facial expression data it receives, and the image processing technology installed on the server extracts facial landmarks (the positions of the eyebrows, eyes, mouth, etc.) and detects their movements.

[1612] Step 6:

[1613] The server analyzes the received voice data and uses voice processing technology to extract voice features such as pitch, volume, and speaking rate.

[1614] Step 7:

[1615] The server uses an emotion engine to recognize the user's emotions from the pre-processed facial expression data and voice data. The emotion engine uses a machine learning model to fuse the data and achieve highly accurate emotion recognition.

[1616] Step 8:

[1617] The server stores the recognized emotion data in a database, which is then used as training data to improve the accuracy of future emotion recognition and response generation.

[1618] Step 9:

[1619] The server compares the data with past data in a database and selects an appropriate response based on the perceived emotion, which is generated by referencing pre-defined rules and case studies.

[1620] Step 10:

[1621] The server creates the generated response in text format and, if necessary, converts the text message into a voice message using speech synthesis technology.

[1622] Step 11:

[1623] The terminal receives the response message from the server, which is then decrypted and prepared for presentation to the user.

[1624] Step 12:

[1625] The device provides a response message to the user. Specifically, the response message is conveyed to the user through a screen display or audio playback. For example, a voice message saying, "Would you like to play some relaxing music?" is played.

[1626] Step 13:

[1627] The user acts on the response, for example, by following a suggestion from the device and starting to listen to relaxing music.

[1628] Step 14:

[1629] The device then recaptures the user's reaction after the response, collecting changes in facial expressions and voice and saving them as new data.

[1630] Step 15:

[1631] The device sends the newly collected feedback data to the server, where it is re-encrypted and sent over the network to the server.

[1632] Step 16:

[1633] The server analyzes the collected feedback data, which updates the learning database of the AI ​​model, improving the accuracy of emotion recognition and response generation from the next time onwards.

[1634] Example 2

[1635] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1636] Conventional technologies lack systems that consistently recognize a user's emotions and generate appropriate responses based on those emotions. Furthermore, even if a user's facial expressions and voice are collected, the data is transmitted and analyzed separately, resulting in an inefficient process. Furthermore, there are insufficient mechanisms for improving the accuracy of user emotion recognition and for incorporating the results of previous responses into feedback. This results in an inability to provide appropriate support to users, hindering improvements in their quality of life and learning efficiency.

[1637] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting the user's voice and facial expression, a means for transmitting the collected data, a means for performing data analysis and emotion recognition, a means for generating an appropriate response, a means for providing the generated response, and a means for collecting feedback data. This makes it possible to recognize the user's emotion with high accuracy and provide appropriate support in real time based on the recognition result.

[1638] "User" refers to any individual or entity that uses this system.

[1639] "Means for capturing facial expressions" refers to a device or software for capturing a user's facial expressions as images or videos.

[1640] "Audio collection means" refers to any device or software that records a user's speech or voice.

[1641] "Means for converting into a data format" refers to software for converting the collected facial expression data and voice data into a format that can be sent to a server (e.g., JSON format).

[1642] "Means for encryption and transmission" refers to a method for transmitting data to a server using TLS encryption to ensure data security.

[1643] "Server" refers to a centralized computing device that analyzes collected data, recognizes emotions, and generates responses.

[1644] "Means for recognizing emotions using machine learning models" refers to algorithms that use machine learning frameworks such as TensorFlow and PyTorch to analyze a user's facial expressions and voice data to identify emotions.

[1645] "Means for generating an appropriate response" refers to software for generating feedback to the user in the form of a text or voice message based on the recognized emotion.

[1646] "Means for generating a voice message using speech synthesis technology" refers to technology (e.g., speech synthesis software) for converting the generated text response into speech format.

[1647] "Means for providing a response" refers to an interface (such as a display or speaker) for providing the generated text or audio response to the user.

[1648] "Means for collecting feedback data" refers to a method for re-collecting user reactions to the generated responses and using that data to generate future responses and improve the accuracy of emotion recognition.

[1649] This invention relates to a system for recognizing a user's emotions and providing an appropriate response. This system, which is composed of a terminal and a server, improves the accuracy of emotion recognition and provides appropriate support to the user in real time.

[1650] Hardware and software used

[1651] 1. Terminal

[1652] Camera: Used to capture the user's facial expressions. Specifically, a common webcam (e.g., Logitech C920) is used.

[1653] Microphone: Used to collect the user's voice. Specifically, a general USB microphone (e.g., Blue Yeti) is used.

[1654] Software: Utilizes OpenCV and Google Cloud Speech-to-Text APIs for facial recognition and voice data collection.

[1655] 2. Server

[1656] Data analysis: The server uses machine learning frameworks such as TensorFlow and PyTorch to process facial expression and voice data.

[1657] Emotion Recognition: Based on data analysis, a fusion model is used to recognize user emotions with high accuracy.

[1658] Response generation: Generates appropriate responses based on the recognized emotions and generates voice messages using text-to-speech technology (e.g., Amazon Polly) if necessary.

[1659] System operation procedures and examples

[1660] The system works as follows: the device collects the user's facial expressions and voice and sends the data to the server. The server analyzes the data and recognizes the user's emotions. Based on the recognized emotions, the server generates an appropriate response and provides it to the user again via the device. The device then collects the user's feedback and sends it to the server to help improve accuracy next time.

[1661] Specific use cases:

[1662] Example 1: Managing emotions of elderly people receiving home care

[1663] Usage scenario: An elderly person is receiving care at home. The device is installed in the living room, and the camera and microphone capture the elderly person's face and voice.

[1664] Operational flow: The device collects the elderly person's facial expressions (e.g., a depressed expression) and voice data (speaking slowly) and sends them to the server. The server analyzes the data and recognizes emotions such as "depression" and "fatigue." The server generates a response such as "play relaxing music," and the device plays the music. The elderly person relaxes while listening to the music, and the device collects their responses again and sends them to the server.

[1665] Example 2: Managing student focus during remote learning

[1666] Usage scenario: A student is learning remotely, with the device set up at a desk and the camera and microphone capturing the student's face and voice.

[1667] Operational flow: The device collects the student's facial expressions (e.g., a concentrated expression) and voice data (clear pronunciation) and sends them to the server. The server analyzes the data and recognizes emotions such as "concentration" and "interested." The server generates a response saying "Notify me when it's time to take a break," and the device notifies the student by voice. The student takes an appropriate break, and their response is collected again and sent to the server.

[1668] Prompt Sentence Examples

[1669] Below are some example prompts to be input to the generative AI model:

[1670] "How can we provide appropriate support to elderly people who are depressed while receiving care at home?"

[1671] "What approaches are effective in helping students stay focused while learning remotely?"

[1672] In this way, it is possible to provide appropriate support in real time, especially for elderly people receiving care at home or students undergoing remote learning. This system is extremely effective in improving users' quality of life and learning efficiency.

[1673] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1674] Step 1:

[1675] Facial expression and voice data collection

[1676] The device uses a camera and microphone to collect the user's facial expressions and voice in real time.

[1677] Input: Image data from the camera and audio data from the microphone.

[1678] Processing: Use OpenCV for facial recognition and Google Cloud Speech-to-Text to convert audio data to text.

[1679] Output: Facial expression data (e.g. facial landmark points) and speech text data.

[1680] Specific operation: The camera captures the user's face, and the microphone records what the user says. For example, it collects audio of an elderly person saying, "I'm not feeling well today."

[1681] Step 2:

[1682] Data preprocessing and transmission

[1683] The facial expression data and voice data collected by the device are converted into a fixed data format (JSON format) and sent to the server using TLS encryption.

[1684] Input: Facial expression data, speech text data.

[1685] Processing: Data conversion and TLS encryption.

[1686] Output: A JSON file containing encrypted facial expression and audio data.

[1687] Specific operation: The device converts facial expression data and voice text into JSON format, encrypts it with TLS, and sends it to the server. For example, it contains data such as "Facial landmarks show that the corners of the mouth are down."

[1688] Step 3:

[1689] Data analysis and emotion recognition

[1690] The server analyzes the received data and uses machine learning models to recognize emotions. Specifically, it uses TensorFlow to process facial expression data and voice data.

[1691] Input: Encrypted facial expression and voice data in JSON format.

[1692] Processing: Decryption, data analysis, and emotion recognition using machine learning models.

[1693] Output: Perceived emotion (e.g., "depressed," "tired").

[1694] Specific operation: The server receives the data, extracts facial landmarks and audio features, and inputs them into the fusion model to identify emotions. For example, it extracts features such as "facial landmarks with downturned corners of the mouth" and "negative audio content" and recognizes the emotion as "depressed."

[1695] Step 4:

[1696] Response Generation

[1697] The server generates an appropriate response based on the recognized emotion and creates a voice message using speech synthesis technology.

[1698] Input: Perceived emotion (e.g., "depressed," "tired").

[1699] Processing: Response text generation and speech synthesis.

[1700] Output: Response text and voice message.

[1701] Specific operation: The server generates the text "Playing relaxing music" and synthesizes it to create a voice message.

[1702] Step 5:

[1703] Providing a response

[1704] The device receives the response from the server and provides it to the user, for example by displaying text on the display or playing an audio message over the speaker.

[1705] Input: Response text and voice message.

[1706] Action: Display text or play audio.

[1707] Output: Feedback provided to the user.

[1708] Specific operation: The device will play a voice message saying "Playing relaxing music" and then play relaxing music.

[1709] Step 6:

[1710] Collecting feedback

[1711] The device collects the user's responses and sends them back to the server to help improve accuracy next time.

[1712] Input: User reaction data (facial expressions and voice).

[1713] Processing: Collect and send data again.

[1714] Output: Feedback data.

[1715] Specific operation: The system collects data on whether the elderly person is relaxed while listening to music or if they have positive reactions (e.g., saying "thank you"), and sends the data to a server. For example, this data includes facial landmarks indicating a smile and audio data saying "thank you."

[1716] (Application example 2)

[1717] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1718] In traditional brick-and-mortar stores, it was difficult to immediately grasp customer emotions and reactions and respond accordingly. As a result, it was not possible to provide services that satisfied customers, and the effect of increasing purchasing motivation was limited. It was also difficult to collect feedback based on customer reactions, and this could not be used to improve services.

[1719] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for capturing the user's facial expression, means for collecting the user's voice, and means for transmitting the captured facial expression data and collected voice data to the server. This makes it possible to recognize customer emotions in real time in a physical store and provide appropriate product recommendations or customer service. Furthermore, the customer's reactions can be collected again and transmitted to the server as feedback data, thereby improving the accuracy of emotion recognition from the next time onwards.

[1720] The "means for capturing a user's facial expression" is a device or system that uses a camera or sensor to acquire image data of a user's facial expression.

[1721] The "means for collecting user's voice" refers to a device or system that uses a microphone or a voice recognition device to acquire the user's speech as voice data.

[1722] The "means for transmitting the captured facial expression data and collected voice data to a server" is a communication device or protocol for sending the facial expression data and voice data acquired over a network to a remote server.

[1723] The "means for recognizing emotions from facial expression data and voice data on a server" refers to a technology that uses software and machine learning models installed on the server to analyze received facial expression data and voice data and identify the user's emotions.

[1724] The "means for generating an appropriate response based on the recognized emotion" refers to an algorithm or software that allows the server to create an optimal text or voice message based on the user's emotional information.

[1725] The "means for providing the generated response to the user" refers to a device or system that transmits the generated response message to the user via a display or speaker.

[1726] "A means for recognizing customer emotions in real time in physical stores and making appropriate product suggestions or customer service responses" refers to software and hardware that utilizes emotion recognition technology to dynamically make product suggestions and customer service responses that correspond to the customer's emotional state.

[1727] The "means for recollecting customer reactions and sending them to the server as feedback data" refers to a device or protocol for recollecting the customer's facial expressions and reactions after the interaction and sending that data to an analysis server.

[1728] The system of the present invention recognizes customer emotions at a high level and provides appropriate product suggestions and customer service in a physical store. A specific implementation method thereof will be described in detail below.

[1729] Hardware Configuration

[1730] This system mainly consists of the following hardware:

[1731] 1. Smart glasses or in-store robots: Used to collect customers' facial expressions and voice.

[1732] 2. Camera: Captures the user's facial expressions.

[1733] 3. Microphone: Collects the user's voice.

[1734] 4. Server: Performs data analysis and emotion recognition.

[1735] Software Configuration

[1736] The system includes the following software components:

[1737] 1. OpenCV: A library for collecting and processing facial expressions.

[1738] 2. requests: HTTP request library for sending data.

[1739] 3. Emotion Engine: A machine learning model to analyze the received data and recognize emotions.

[1740] Data processing and calculation

[1741] The system processing flow is as follows:

[1742] 1. Data collection: Using cameras and microphones on smart glasses or robots, customers' facial expressions and voice data are collected in real time, allowing instantaneous acquisition of customer emotional information.

[1743] 2. Data transmission: The collected facial expression and voice data is encrypted and transmitted from the device to the server, ensuring data security.

[1744] 3. Data Analysis: The server analyzes the received data. Specifically, it uses OpenCV to process facial expression data and uses an emotion engine to recognize emotions. The voice data is processed in the same way, and the data is integrated to identify emotions.

[1745] 4. Response Generation: The server generates appropriate product suggestions and customer service responses based on the recognized emotion data. Responses are created as text or voice messages.

[1746] 5. Response provision: The generated response data is provided to the customer through smart glasses or a robot, thereby enhancing customer interaction.

[1747] 6. Feedback collection: After responding, the system collects the customer's reaction again and sends the data to the server, providing feedback to improve the accuracy of emotion recognition in the future.

[1748] Specific examples

[1749] As a specific example, the following scenario can be considered.

[1750] 1. A customer enters the store: A staff member wearing smart glasses approaches the customer and collects facial expressions and voice recordings.

[1751] 2. Data analysis: The collected data is sent to the server, which analyzes it using an emotion engine. If the result indicates that the customer is interested, the server generates appropriate product suggestions.

[1752] 3. Product proposals and responses: The generated product proposals are transmitted to the staff via smart glasses, who then make the proposals to the customer. The customer's responses are collected again and sent to the server as feedback.

[1753] Prompt Sentence Examples

[1754] Examples of prompts for generative AI models include:

[1755] "Please generate a program for a system that analyzes customer facial expressions and voice data, recognizes emotions, and provides appropriate product suggestions or customer service responses to the customer."

[1756] As described above, this invention utilizes emotion recognition technology to improve the quality of service in brick-and-mortar stores and increase customer satisfaction.

[1757] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1758] Step 1:

[1759] The device uses smart glasses or in-store robots to collect customers' facial expressions and voice in real time.

[1760] Input: Customer's facial expression (video data), voice (audio data)

[1761] How it works: The device's camera captures facial expressions, and the microphone collects audio. The facial expressions are captured as image data using the OpenCV library, and the audio is captured as audio data using a voice recognition device.

[1762] Output: Acquired facial expression data and voice data

[1763] Step 2:

[1764] The facial expression data and voice data collected by the terminal are transmitted to a server.

[1765] Input: Facial expression data, voice data

[1766] How it works: The device encrypts facial expression and voice data and sends it securely to the server using HTTP requests, with the data sent in JSON format.

[1767] Output: Facial expression data and voice data sent to the server

[1768] Step 3:

[1769] The server analyzes the received facial expression data and voice data and recognizes emotions using an emotion engine.

[1770] Input: Facial expression data and voice data sent to the server

[1771] How it works: The server uses OpenCV to analyze facial expression data and extract facial landmarks. It also extracts voice features from voice data. The emotion engine integrates these data and uses machine learning models to recognize emotions.

[1772] Output: Recognized emotion data (e.g., joy, interest, anxiety, etc.)

[1773] Step 4:

[1774] The server generates an appropriate response based on the recognized emotion data.

[1775] Input: Recognized emotion data

[1776] How it works: The server uses a response generation algorithm to generate the optimal response (e.g., product suggestions, customer service messages, etc.) based on the emotion. Responses are constructed as text or voice messages.

[1777] Output: Generated response data (text or audio)

[1778] Step 5:

[1779] The terminal receives the response data from the server and provides it to the client.

[1780] Input: Generated response data

[1781] Operation: The device processes the received response data and displays it on the smart glasses display or plays it aloud through the robot's speaker.

[1782] Output: The response (visual or audio message) provided to the customer

[1783] Step 6:

[1784] The terminal again collects the customer's response and sends it to the server as feedback data.

[1785] Input: New customer facial expression data, voice data

[1786] How it works: The device again collects the customer's facial expressions and voice and sends them to the server. Based on this data, the server performs feedback analysis to improve the accuracy of the emotion recognition model.

[1787] Output: Feedback data (facial expression and voice data) sent to the server

[1788] Through these steps, the system can recognize customer emotions in real time and significantly improve the quality of service in physical stores.

[1789] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1790] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1791] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1792] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1793] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1794] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1795] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1796] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1797] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1798] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1799] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1800] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1801] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1802] 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.

[1803] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1804] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1805] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1806] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1807] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1808] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1809] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1810] The following is further disclosed regarding the above embodiment.

[1811] (Claim 1)

[1812] means for capturing a user's facial expression;

[1813] means for collecting the user's voice;

[1814] means for transmitting the captured facial expression data and collected voice data to a server;

[1815] means for recognizing emotions from the facial expression data and voice data in the server;

[1816] means for generating an appropriate response based on the recognized emotion;

[1817] means for providing the generated response to a user;

[1818] A system including:

[1819] (Claim 2)

[1820] 2. The system according to claim 1, further comprising means for storing the recognized emotion information in a database in the server and improving the accuracy of subsequent emotion recognitions based on the stored information.

[1821] (Claim 3)

[1822] 10. The system of claim 1, wherein the response provided to the user comprises a voice message using speech synthesis technology.

[1823] "Example 1"

[1824] (Claim 1)

[1825] A means for capturing a user's facial expression;

[1826] means for collecting the user's voice;

[1827] means for transmitting the captured facial expression data and collected voice data to a central processing unit;

[1828] means for recognizing emotions from the facial expression data and voice data in the central processing unit;

[1829] means for generating an appropriate response based on the recognized emotion;

[1830] means for providing the generated response to a user;

[1831] A means for collecting user responses again and using them to generate responses in the future;

[1832] A system including:

[1833] (Claim 2)

[1834] 2. The system according to claim 1, further comprising means in said central processing unit for storing said recognized emotion information in a storage device and for improving the accuracy of subsequent emotion recognitions based on said stored emotion information.

[1835] (Claim 3)

[1836] 10. The system of claim 1, wherein the response provided to the user comprises a voice message using speech synthesis technology.

[1837] "Application Example 1"

[1838] (Claim 1)

[1839] means for capturing a user's facial expression;

[1840] means for collecting the user's voice;

[1841] means for transmitting the captured facial expression data and collected voice data to a server;

[1842] means for recognizing emotions from the facial expression data and voice data in the server;

[1843] means for generating an appropriate response based on the recognized emotion;

[1844] means for providing the generated response to a user;

[1845] A means for providing a customer service assistant based on emotion recognition in the physical store; and a means for playing a promotional video or offering a special discount in response to the generated response.

[1846] A system including:

[1847] (Claim 2)

[1848] 2. The system according to claim 1, further comprising means for storing the recognized emotion information in a database in the server and improving the accuracy of subsequent emotion recognitions based on the stored information.

[1849] (Claim 3)

[1850] 10. The system of claim 1, wherein the response provided to the user comprises a voice message using speech synthesis technology.

[1851] "Example 2: Combining Emotion Engines"

[1852] (Claim 1)

[1853] means for capturing a user's facial expression;

[1854] means for collecting the user's voice;

[1855] means for converting the captured facial expression data and collected voice data into a data format, encrypting the data, and transmitting the data to a server;

[1856] means for recognizing emotions from the facial expression data and voice data using a machine learning model in the server;

[1857] means for generating an appropriate response based on the recognized emotion and, if necessary, generating a voice message using voice synthesis technology;

[1858] means for providing the generated response to the user, and collecting the user's reaction again and transmitting the feedback data to the server;

[1859] A system including:

[1860] (Claim 2)

[1861] 2. The system according to claim 1, further comprising means for storing the recognized emotion information in a database in the server and improving the accuracy of subsequent emotion recognitions based on the stored information.

[1862] (Claim 3)

[1863] 10. The system of claim 1, wherein the response provided to the user comprises a voice message using speech synthesis technology.

[1864] "Application example 2 when combining emotion engines"

[1865] (Claim 1)

[1866] means for capturing a user's facial expression;

[1867] means for collecting the user's voice;

[1868] means for transmitting the captured facial expression data and collected voice data to a server;

[1869] means for recognizing emotions from the facial expression data and voice data in the server;

[1870] means for generating an appropriate response based on the recognized emotion;

[1871] means for providing the generated response to a user;

[1872] A means to recognize customer emotions in real time in physical stores and provide appropriate product suggestions or customer service responses;

[1873] A means for collecting customer responses again and transmitting them to the server as feedback data;

[1874] A system including:

[1875] (Claim 2)

[1876] 2. The system according to claim 1, further comprising means for storing the recognized emotion information in a database in the server and improving the accuracy of subsequent emotion recognitions based on the stored information.

[1877] (Claim 3)

[1878] 10. The system of claim 1, wherein the response provided to the user comprises a voice message using speech synthesis technology. [Explanation of symbols]

[1879] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for capturing a user's facial expression; means for collecting the user's voice; means for transmitting the captured facial expression data and collected voice data to a server; means for recognizing emotions from the facial expression data and voice data in the server; means for generating an appropriate response based on the recognized emotion; means for providing the generated response to a user; A system including:

2. The system according to claim 1 , further comprising means for storing the recognized emotion information in a database in the server and for improving the accuracy of subsequent emotion recognitions based on the stored information.

3. The system of claim 1 , wherein the response provided to the user comprises a voice message using speech synthesis technology.

Images