System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting real-time voice, facial expression, and gaze data of a user; means for pre-processing the voice, facial expression, and gaze data; means for analyzing the pre-processed data and assessing the user's comprehension, focus, and stress levels; means for generating or selecting learning content based on the assessment results; means for presenting the learning content to the user; and means for collecting and re-preprocessing and analyzing user reaction data.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 conventional educational systems, it has been difficult to accurately grasp the level of understanding and concentration of each learner and provide learning content that corresponds to that level. This has resulted in reduced learning efficiency and made it difficult to provide instruction that meets the individual needs of each learner. The purpose of this invention is to solve these problems and provide a learning experience that is optimized for each learner. [Means for solving the problem]

[0005] The present invention is a system that includes a means for collecting a user's voice, facial expression, and gaze data in real time, a means for preprocessing the voice, facial expression, and gaze data, a means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, a means for generating or selecting learning content based on the evaluation results, a means for presenting the learning content to the user, and a means for collecting, preprocessing, and analyzing user reaction data. The system also includes a means for dynamically adapting the learning content and a speech recognition engine for converting voice data to text data, thereby providing a customized learning path tailored to individual learning needs.

[0006] "User" refers to an individual learner who uses the system to study.

[0007] "Audio data" refers to a digital audio signal that includes the user's speech or voice and its sound wave information.

[0008] "Facial expression data" refers to video information that represents a user's facial expressions and other subtle facial movements.

[0009] "Gaze data" refers to information that represents the user's eye movements and gaze direction.

[0010] "Real-time" refers to immediate response or processing without delay.

[0011] "Preprocessing" refers to operations that convert collected data into a form suitable for analysis, such as signal processing, noise removal, and normalization.

[0012] "Analysis" refers to the process of evaluating and determining pre-processed data using algorithms and machine learning models.

[0013] "Level of understanding" refers to a measure that indicates how accurately the user has understood the learning content.

[0014] "Concentration" refers to a measure that indicates how much a user can concentrate on a particular learning content.

[0015] "Stress level" refers to an index that indicates the degree of tension or strain a user feels while studying.

[0016] "Learning content" refers to educational materials, questions, explanatory videos, etc. that users can use to study.

[0017] "Dynamic adaptation" refers to the continuous adjustment of learning content and learning progress based on the analysis results.

[0018] A "voice recognition engine" refers to the algorithms and software used to convert voice data into text data.

[0019] "Response data" refers to data including user responses and actions to learning content, as well as evaluation results.

[0020] The term "system" refers to the entire hardware and software that integrates and operates the above-mentioned means. [Brief explanation of the drawings]

[0021] [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

[0022] 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.

[0023] First, the terms used in the following description will be explained.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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."

[0029] [First embodiment]

[0030] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0031] 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.

[0032] 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).

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0038] 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.

[0039] 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.

[0040] 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.

[0041] 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."

[0042] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. The following describes in detail an embodiment of the present invention.

[0043] System Overview

[0044] The system consists of three main components: terminals, servers, and users.

[0045] 1. Terminal

[0046] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0047] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0048] 2. Server

[0049] The server receives the preprocessed data sent from the terminal and analyzes it using a voice recognition engine and facial expression analysis software.

[0050] The voice data is converted into text data, and the user's level of understanding, concentration, and stress level are evaluated from facial expression and gaze data.

[0051] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[0052] 3. Users

[0053] The user digests the learning content provided via the terminal and makes the necessary responses.

[0054] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0055] Program processing

[0056] The flow of program processing in this system will be explained below.

[0057] Data Collection Phase

[0058] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0059] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0060] Analysis Phase

[0061] The terminal transmits the preprocessed data to the server.

[0062] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[0063] Learning content customization phase

[0064] Based on the analyzed data, the server evaluates the user's level of understanding and learning progress, and generates or selects customized learning content.

[0065] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0066] Presentation Phase

[0067] The server transmits customized learning content to the terminal.

[0068] The terminal presents the received learning content to the user in real time, and the user proceeds with learning in accordance with the content.

[0069] Feedback Phase

[0070] The user digests the presented learning content and responds to demonstrate the learning effect.

[0071] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0072] Specific examples

[0073] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0074] In this way, the system provides real-time adaptive learning based on the user's level of understanding and concentration.

[0075] The processing flow will be explained below.

[0076] Step 1:

[0077] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0078] Step 2:

[0079] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[0080] Step 3:

[0081] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[0082] Step 4:

[0083] The terminal transmits the preprocessed data to the server.

[0084] Step 5:

[0085] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[0086] Step 6:

[0087] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[0088] Step 7:

[0089] The server analyzes the gaze data and identifies which part the user is focusing on.

[0090] Step 8:

[0091] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[0092] Step 9:

[0093] The server generates or selects customized learning content based on the evaluation results.

[0094] Step 10:

[0095] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[0096] Step 11:

[0097] The server transmits customized learning content to the terminal.

[0098] Step 12:

[0099] The device presents the received learning content to the user in real time, supporting them in their learning.

[0100] Step 13:

[0101] The user digests the presented learning content and takes the necessary responses or actions.

[0102] Step 14:

[0103] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[0104] Step 15:

[0105] The terminal preprocesses the newly collected data and sends it back to the server.

[0106] Step 16:

[0107] The server analyzes the data again and generates the next learning path based on the feedback information.

[0108] Step 17:

[0109] The server sends the generated new learning path to the terminal, which then presents it to the user.

[0110] Step 18:

[0111] Users progress through new learning paths.

[0112] Example 1

[0113] 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."

[0114] Conventional learning systems have had difficulty adjusting content based on the user's real-time status and reactions in order to maximize the user's learning efficiency. This has resulted in the inability to respond immediately when the user stumbles on a particular problem or point, resulting in a decrease in learning efficiency. Therefore, there is a need for a system that can dynamically customize learning content based on the user's real-time data and quickly respond to each user's learning needs.

[0115] 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.

[0116] In this invention, the server includes means for collecting voice, facial expression, and gaze data of a user in real time, means for preprocessing the voice, facial expression, and gaze data, and means for analyzing the preprocessed data to evaluate the user's level of understanding, concentration, and stress level, thereby enabling dynamic customization of learning content based on the user's real-time data to meet individual learning needs.

[0117] "Voice data" refers to data consisting of user utterances, sounds, and linguistic information.

[0118] "Facial expression data" is data that includes information about the user's facial expression.

[0119] "Gaze data" is data that includes information about the user's eye movements and viewpoint.

[0120] "Preprocessing" refers to processing the collected data, such as noise removal and filtering, to make it easier to analyze.

[0121] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[0122] "Learning content" refers to content such as educational materials, questions, and explanatory videos provided to support users' learning.

[0123] "Customized learning content" is learning content that is adapted based on a user's individual needs and learning progress.

[0124] "Batch transmission" is a method of transmitting multiple pieces of data together at regular time intervals.

[0125] "Facial expression analysis software" is software that analyzes a user's facial expressions to evaluate their emotions and state.

[0126] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0127] "Noise reduction" is a process of removing unnecessary noise from audio data and the like.

[0128] "Level of concentration" is a measure of how much the user is concentrating on their studies.

[0129] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[0130] The present invention is a system that collects and analyzes voice, facial expression, and gaze data of a user in real time to improve the user's learning efficiency, and dynamically provides learning content tailored to individual needs. Specific embodiments for implementing the present invention will be described in detail below.

[0131] System Overview

[0132] The system consists of three main components: terminals, servers, and users.

[0133] 1. Terminal

[0134] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0135] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0136] 2. Server

[0137] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The voice data is converted into text data, and the user's level of comprehension, concentration, and stress level are evaluated based on facial expression and gaze data.

[0138] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[0139] 3. Users

[0140] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[0141] Specific examples of programs

[0142] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0143] Hardware and software used

[0144] Speech recognition engine: For example, use the Google Speech-to-Text API to convert voice data into text data.

[0145] Facial expression analysis software: For example, using Microsoft Azure Face API to analyze the user's facial expressions.

[0146] Eye-tracking software: In combination with a camera, it collects and analyzes user gaze data.

[0147] Examples of prompt statements

[0148] For example, enter the following prompt sentence into the generative AI model:

[0149] "Design a system that can detect in real time when you are struggling with a problem and provide you with the appropriate learning content. For example, it would be even better if the system could detect when a student is struggling with a problem and provide supplementary materials."

[0150] summary

[0151] The present invention is a system that dynamically customizes learning content based on users' real-time data to meet their individual learning needs, thereby maximizing the user's learning efficiency and supporting rapid problem solving.

[0152] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0153] Step 1:

[0154] Starting the sensor

[0155] The device activates the camera, microphone, and other sensors to begin collecting data from the user. For example, the device automatically activates each sensor the moment the user starts learning content. The input is the user's action to start the learning content, and the output is each activated sensor.

[0156] Step 2:

[0157] Data collection

[0158] The device collects the user's voice, facial expression, and gaze data in real time. For example, a camera captures the user's facial expressions and a microphone records what the user says. The input is the real-time user data captured by the sensor, and the output is the collected raw data.

[0159] Step 3:

[0160] Temporary storage and preprocessing of data

[0161] The device temporarily stores the collected data in local storage and performs preprocessing such as noise reduction and filtering. For example, a noise reduction algorithm is applied to clear voice data collected in a noisy environment. The input is the collected raw data, and the output is the preprocessed data.

[0162] Step 4:

[0163] Sending preprocessed data

[0164] The terminal sends the pre-processed data to the server, for example, by batch transmission of multiple data at regular intervals. The input is the pre-processed data, and the output is the data sent to the server.

[0165] Step 5:

[0166] Analysis of audio data

[0167] The server converts the voice data into text data using a speech recognition engine (e.g., Google Speech-to-Text API). For example, a user's statement "I don't understand this problem" is converted into text data. The input is the transmitted voice data, and the output is the converted text data.

[0168] Step 6:

[0169] Analysis of facial expression and gaze data

[0170] The server uses facial expression analysis software (e.g., Microsoft Azure Face API) and eye-tracking software to determine the user's level of comprehension, concentration, and stress. For example, it can detect when the user frowns and determine that the user's stress level is high. The input is the transmitted facial expression and eye-gaze data, and the output is the analyzed user's level of comprehension, concentration, and stress.

[0171] Step 7:

[0172] Assessment of understanding and progress

[0173] The server evaluates the user's level of understanding and learning progress based on the analyzed data. For example, it determines the level of understanding based on data such as which questions the user spent the most time on. The input is the analyzed user data, and the output is the evaluation results.

[0174] Step 8:

[0175] Content generation or selection

[0176] The server generates or selects customized learning content and adjusts the content. For example, if a user has difficulty with "factorization," it selects videos or additional practice problems to refresh the knowledge from the basics. The input is the assessment results, and the output is customized learning content.

[0177] Step 9:

[0178] Sending customized content

[0179] The server sends customized learning content to the device. For example, it sends a specific video link or PDF file to the device. The input is the customized learning content, and the output is the data sent to the device.

[0180] Step 10:

[0181] View content

[0182] The device displays the received learning content to the user, and the user proceeds with the learning process. For example, a video for problem solving is displayed on the screen. The input is the transmitted learning content data, and the output is the content displayed to the user.

[0183] Step 11:

[0184] Collecting user learning status

[0185] The user digests the presented learning content and responds. For example, after solving a problem, the user may move on to the next problem or respond by saying "I couldn't solve it." The input is the user's learning behavior, and the output is learning behavior data.

[0186] Step 12:

[0187] Reaction data collection

[0188] The device again collects the user's voice, facial expression, and gaze data, and performs preprocessing and analysis again. For example, if the user makes a new confused expression during learning, that data will also be collected. The input is the newly collected data, and the output is the data that will be preprocessed again.

[0189] (Application example 1)

[0190] 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."

[0191] On manufacturing lines, it is important for workers to quickly acquire skills and improve production efficiency. However, it is difficult to provide training content that is tailored to each worker's level of understanding and progress, and general manuals and videos alone are often insufficient. It is also difficult to grasp in real time which parts of the training workers are struggling with. This reduces the effectiveness of training and leads to issues such as lower production efficiency.

[0192] 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.

[0193] In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for generating or selecting learning content based on the evaluation results, means for presenting the learning content to the user, means for collecting user reaction data and preprocessing and analyzing it again, and means for providing individual training content to production line workers by collecting and analyzing voice, facial expression, and gaze data in real time, thereby enabling the provision of optimal training content tailored to the worker's level of understanding and progress.

[0194] "Collecting user voice, facial expression, and gaze data" refers to sensing and recording the user's voice, facial expression changes, and gaze movements in real time.

[0195] "Preprocessing" refers to a series of operations that remove noise from collected raw data and prepare the data in a usable form.

[0196] "Data analysis" refers to extracting information such as the user's level of understanding, concentration, and stress level from the pre-processed data using specific algorithms or models.

[0197] "Generation or selection of learning content" refers to the dynamic creation or selection of teaching materials or information to provide users with the information or training they need based on the analysis results.

[0198] "Presenting learning content" refers to displaying generated or selected learning content to a user in real time and allowing the user to use the content.

[0199] "Collecting user reaction data" refers to re-detecting and recording the user's behavior, comments, facial expressions, and eye movements while studying.

[0200] "Providing training content to workers on the production line" refers to dynamically presenting content to individual workers working on the manufacturing site to enable them to learn the necessary knowledge and skills on the spot.

[0201] "Individualized training content" refers to training content and learning materials that are customized to the specific needs and level of understanding of the user (worker).

[0202] System Overview

[0203] This invention is a system that collects voice, facial expression, and gaze data of workers in real time and analyzes the data to dynamically provide training content tailored to individual needs in order to improve the learning efficiency of workers on a production line. Specific embodiments for implementing this invention will be described below.

[0204] Main configuration

[0205] The system consists of the following main components:

[0206] 1. Terminal

[0207] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect voice, facial expression, and gaze data of workers in real time.

[0208] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0209] 2. Server

[0210] The server receives the preprocessed data sent from the terminal and analyzes it using a speech recognition engine (for example, the speech_recognition library) or facial expression analysis software (for example, the EmotionRecognition library).

[0211] The voice data is converted into text data, and the worker's level of understanding, concentration, and stress level are evaluated based on facial expression and gaze data.

[0212] Based on the analysis results, training content is generated or selected and dynamically adapted.

[0213] 3. User (operator)

[0214] The worker executes the training content provided via the terminal and makes the necessary responses.

[0215] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0216] Program processing

[0217] The flow of program processing in this system will be explained below.

[0218] Data Collection Phase

[0219] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time.

[0220] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0221] Analysis Phase

[0222] The terminal transmits the preprocessed data to the server.

[0223] The server converts the voice data into text data using a speech recognition engine (speech_recognition library), and uses facial expression and gaze data to determine the worker's level of understanding, concentration, and stress.

[0224] Learning content customization phase

[0225] Based on the analyzed data, the server evaluates the worker's level of understanding and learning progress, and generates or selects customized training content.

[0226] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0227] Presentation Phase

[0228] The server transmits customized learning content to the terminal.

[0229] The terminal presents the received training content to the worker in real time, and the worker follows along with the content.

[0230] Feedback Phase

[0231] The worker performs the presented training content and provides responses that demonstrate the effectiveness of their learning.

[0232] The device again collects data such as the worker's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0233] Specific examples

[0234] For example, if a new worker is confused about a particular procedure on a production line, the device can detect this situation through the camera and microphone. Specifically, it analyzes the worker's confused facial expression and voice, such as "What should I do about this?" Based on this, the server can provide the worker with a training video or additional explanations that are best suited to the worker's basic knowledge. This method can help new workers to proceed with their work smoothly.

[0235] Prompt Sentence Examples

[0236] If a new worker looks confused and says in audio, "What do I do about this?", suggest a training video that will help new workers review the basics.

[0237] This system provides real-time adaptive training that matches the worker's level of understanding and progress, improving the efficiency and quality of the production line.

[0238] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0239] Step 1: Data collection phase

[0240] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time. The input of this data collection is the worker's real-time activity data, and the output is audio files, image data, and gaze tracking data. This data is temporarily stored in local storage and undergoes preprocessing such as noise removal and filtering. For example, background noise is removed from the voice data and unnecessary background is cut out from the image data.

[0241] Step 2: Preprocessing phase

[0242] The device preprocesses the collected data. The input is the voice, facial expression, and gaze data collected in step 1, and the output is data that has been subjected to noise removal and filtering. Specifically, the waveform of the voice data is flattened, and histogram normalization is performed on the image data. The gaze data is also filtered and blurring is corrected.

[0243] Step 3: Data transmission phase

[0244] The device sends the preprocessed data to the server. The input is the preprocessed voice, facial expression, and gaze data, and the output is the data sent to the server. For example, the device divides the data into packets and sends them to the server.

[0245] Step 4: Analysis Phase

[0246] The server receives the transmitted data and converts the voice data into text using a speech recognition engine (e.g., speech_recognition library). The input is the preprocessed data transmitted from the device, and the output is text data and analysis data (level of understanding, concentration, stress level, etc.). Specifically, the server analyzes the facial expression data using facial expression analysis software (e.g., EmotionRecognition library) to determine the user's level of understanding. It also applies an algorithm to evaluate the level of concentration based on gaze data.

[0247] Step 5: Customizing the learning content

[0248] The server evaluates the worker's level of understanding and learning progress based on the analyzed data, and generates or selects customized training content. The input is data such as the worker's level of understanding, concentration, and stress level obtained in the analysis phase, and the output is training content dynamically created based on this data. Specifically, the server selects video materials and interactive simulations that are optimal for the worker's level of progress.

[0249] Step 6: Content Presentation Phase

[0250] The server sends customized training content to the terminal, which then presents it to the worker. The input is the training content sent from the server, and the output is the training video or simulation that the worker views. Specifically, the terminal displays the received content on a display and presents it to the worker by providing audio guidance.

[0251] Step 7: Feedback Phase

[0252] The user digests the presented training content and responds to demonstrate the learning effect. The input is the user's input (voice response, facial expression, and gaze data), and the output is the collected response data. For example, there are situations where a worker answers questions about the training content by voice. This response data is collected again by the system, and the next learning path is further optimized.

[0253] This process allows workers to receive effective training tailored to their individual needs.

[0254] 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.

[0255] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that also takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[0256] System Overview

[0257] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[0258] 1. Terminal

[0259] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0260] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0261] 2. Server

[0262] The server receives the pre-processed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software to assess the user's level of understanding, concentration, and stress level.

[0263] The server further identifies the user's emotional state using an emotion engine and generates or selects learning content by taking the emotion information into account in the analysis results.

[0264] 3. Emotion Engine

[0265] The emotion engine analyzes the user's facial expression data and voice data to determine the user's emotional state.

[0266] The determined emotions are sent to the server and used to adapt the learning content.

[0267] 4. Users

[0268] The user digests the learning content provided via the terminal and makes the necessary responses.

[0269] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0270] Program processing

[0271] The flow of program processing in this system will be explained below.

[0272] Data Collection Phase

[0273] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0274] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0275] Analysis Phase

[0276] The terminal transmits the preprocessed data to the server.

[0277] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[0278] The server uses an emotion engine to analyze the user's facial expression data and voice data to identify the user's emotional state.

[0279] Learning content customization phase

[0280] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects customized learning content.

[0281] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0282] Presentation Phase

[0283] The server transmits customized learning content to the terminal.

[0284] The device presents the received learning content to the user in real time, supporting them in their learning.

[0285] Feedback Phase

[0286] The user digests the presented learning content and responds to demonstrate the learning effect.

[0287] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0288] Specific examples

[0289] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0290] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[0291] The processing flow will be explained below.

[0292] Step 1:

[0293] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0294] Step 2:

[0295] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[0296] Step 3:

[0297] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[0298] Step 4:

[0299] The terminal transmits the preprocessed data to the server.

[0300] Step 5:

[0301] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[0302] Step 6:

[0303] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[0304] Step 7:

[0305] The server analyzes the gaze data and identifies which part the user is focusing on.

[0306] Step 8:

[0307] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[0308] Step 9:

[0309] The server uses an emotion engine to analyze the facial expression data and voice data to identify the user's emotional state.

[0310] Step 10:

[0311] The server generates or selects customized learning content based on the evaluation results and emotional information.

[0312] Step 11:

[0313] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[0314] Step 12:

[0315] The server transmits customized learning content to the terminal.

[0316] Step 13:

[0317] The device presents the received learning content to the user in real time, supporting them in their learning.

[0318] Step 14:

[0319] The user digests the presented learning content and takes the necessary responses or actions.

[0320] Step 15:

[0321] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[0322] Step 16:

[0323] The terminal preprocesses the newly collected data and sends it back to the server.

[0324] Step 17:

[0325] The server analyzes the data again and generates the next learning path based on the feedback information.

[0326] Step 18:

[0327] The server sends the generated new learning path to the terminal, which then presents it to the user.

[0328] Step 19:

[0329] Users progress through new learning paths.

[0330] Example 2

[0331] 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."

[0332] Conventional learning support systems have difficulty accurately grasping a user's level of understanding and concentration, making it impossible to provide content tailored to individual learning needs. Furthermore, they often provide uniform learning content without considering the user's emotional state, limiting the improvement of learning efficiency. Furthermore, they do not dynamically adapt content to match the user's learning progress, making it difficult to motivate users or provide an effective learning experience.

[0333] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for determining the user's emotional state using the evaluation results and an emotion engine, means for generating or selecting learning content based on the evaluation results and emotion information, means for presenting the learning content to the user, and means for collecting user reaction data and re-preprocessing and analyzing the data. This makes it possible to provide individually customized learning content that takes into account the user's emotional state as well as their level of understanding and concentration.

[0334] "Voice data" refers to data in which the user's voice is recorded in digital format.

[0335] "Facial expression data" refers to data that captures a user's facial expression using a camera or the like and converts it into an analyzable format.

[0336] "Gaze data" refers to data that measures and records the user's eye movements and gaze direction.

[0337] "Preprocessing" is the process of performing processes such as noise removal and filtering on collected raw data to convert it into a format suitable for analysis.

[0338] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[0339] "Evaluation" is the process of determining the user's learning status according to specific criteria based on the information obtained through analysis.

[0340] An "emotion engine" is software or hardware that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[0341] "Learning content" is a general term for teaching materials and educational resources presented to users, and includes text, video, audio, practice questions, and the like.

[0342] "Customization" is the process of optimizing learning content according to each user's needs and learning situation.

[0343] "Presenting" refers to the act of displaying or playing selected or generated learning content to a user.

[0344] "Reaction data" is data that records the user's reactions to learning content, including voice, facial expressions, and gaze.

[0345] "Dynamic adaptation" is the process of adjusting and providing learning content in real time based on the user's learning progress, level of understanding, and emotional state.

[0346] A "voice recognition engine" is software or hardware that converts voice data into text data.

[0347] MODE FOR CARRYING OUT THE INVENTION

[0348] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by using an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[0349] System configuration

[0350] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[0351] Terminal

[0352] The device is equipped with a camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0353] server

[0354] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The server evaluates the user's level of understanding, concentration, and stress level. It also identifies the user's emotional state using an emotion engine, and generates or selects learning content by incorporating emotional information into the analysis results.

[0355] Emotion Engine

[0356] The emotion engine analyzes the user's facial expression and voice data to determine their emotional state. The determined emotions are sent to the server and used to adapt the learning content.

[0357] User

[0358] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[0359] System operation example

[0360] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0361] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[0362] Example prompts to input to the generative AI model

[0363] Below are some example prompts to input to a generative AI model:

[0364] "Based on a scenario where a student is working on a math problem but is having trouble with a particular one, describe a system that analyzes the user's level of understanding and emotional state in real time and provides customized learning content."

[0365] This prompt-based system optimizes the user's learning efficiency and provides advanced learning support that meets individual needs.

[0366] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0367] Step 1: Data collection

[0368] The device activates sensors such as a camera, microphone, and gaze tracker to collect the user's voice, facial expression, and gaze data in real time. This provides basic data for understanding the user's behavior and reactions in real time. For example, a user may have a confused expression and say, "I don't understand this problem." The input data obtained here are voice data, facial expression data, and gaze data, which are collected by the device.

[0369] Step 2: Data Preprocessing

[0370] The device temporarily stores the collected data in local storage and performs preprocessing such as noise removal and filtering. As a result of preprocessing, the data is in a clean format suitable for analysis. For example, background noise is removed from voice data and facial expression data is normalized. The input data is the raw data collected in step 1, and the output data is the preprocessed, clean data.

[0371] Step 3: Send data

[0372] The device divides the preprocessed data into packets and sends them to the server. Highly efficient data transfer is important in this step. The input data is the preprocessed data, and the output is the data sent to the server. For example, the device may use a compression algorithm to reduce the size of the data before sending it to the server.

[0373] Step 4: Convert audio data to text

[0374] The server converts the received voice data into text data using a voice recognition engine. For example, the voice saying "I don't understand this problem" is converted into text "I don't understand this problem." The input data is preprocessed voice data, and the output data is text data.

[0375] Step 5: Analyze facial expression data

[0376] The server uses facial expression analysis software to analyze the user's facial expression data to determine their level of understanding, concentration, and stress. For example, a confused expression can indicate a low level of understanding. The input data is the preprocessed facial expression data, and the output data is the analysis result.

[0377] Step 6: Analyze gaze data

[0378] The server analyzes the gaze tracker data to determine where the user is looking. For example, if the user looks at a particular problem area for a long time, it collects that information. The input data is the preprocessed gaze data, and the output data is the gaze focus information.

[0379] Step 7: Identify your emotional state

[0380] The server uses an emotion engine to identify the user's emotional state based on facial expression and voice data. For example, it determines whether the user is feeling stressed. The input data is the analyzed facial expression and voice data, and the output data is the result of the emotional state determination.

[0381] Step 8: Assess your learning progress

[0382] The server evaluates the user's learning progress based on the collected and analyzed data. For example, it identifies areas where the user is struggling with a particular topic. The input data are the analysis results and the judgment of the user's emotional state, and the output data is the evaluation of the user's learning progress.

[0383] Step 9: Content Selection and Creation

[0384] The server selects or generates optimal learning content for the user, such as basic instructional videos or practice questions. The input data is the evaluation results of the learning progress, and the output data is the selected or generated learning content.

[0385] Step 10: Present content to the user

[0386] The server sends the customized learning content to the device, which then presents it to the user, for example, by displaying and playing a video on the screen. The input data is the selected and generated learning content, and the output data is the presented learning content.

[0387] Step 11: Collect user responses

[0388] The user digests the learning content and responds with quizzes or speech to check their understanding. For example, the user answers questions in a quiz. The input data is the user's response, and the output data is the collected response data.

[0389] Step 12: Recollection and analysis

[0390] The device again collects data such as the user's voice, facial expression, and gaze, and sends it to the server for preprocessing and analysis. For example, it re-analyzes how the user is struggling with a quiz. The input data is the user's voice, facial expression, and gaze data, and the output data is the re-analysis result.

[0391] (Application example 2)

[0392] 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."

[0393] Conventional learning support systems have difficulty in comprehensively evaluating a user's level of understanding, concentration, and emotional state, resulting in problems with the provided learning content not being optimized for each individual user. Furthermore, it is difficult to identify in real time where a user is struggling and provide appropriate feedback. Furthermore, they lack the ability to adaptively generate and select learning content, which reduces the user's learning efficiency.

[0394] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0395] In this invention, the server includes means for collecting a user's voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for identifying the user's emotional state using an emotion engine and using the identified emotional state to adapt content, means for generating or selecting individually customized learning content based on the analysis results, means for presenting the generated or selected learning content to the user and prompting the user to respond to demonstrate learning effectiveness, and means for generating prompts using a generative AI model to generate appropriate learning content. This improves user learning efficiency and enables the provision of content optimized for each individual user in real time.

[0396] "Voice Data" means data in digital form that records a user's speech or other vocal communication.

[0397] "Facial expression data" is digital data acquired in real time based on the movement of the user's facial muscles.

[0398] "Gaze data" is digital data that tracks the user's eye movements and records where they are looking in real time.

[0399] "Preprocessing" refers to the process of converting collected data into a format that is easy to analyze by performing processes such as noise removal, filtering, and normalization on the data.

[0400] "Analysis" refers to data processing to assess the user's level of understanding, concentration, stress level, and emotional state based on the preprocessed data.

[0401] "Level of understanding" is an index that indicates how well a user understands the learning content.

[0402] "Concentration level" is an index that indicates how much a user is concentrating on a learning task.

[0403] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[0404] An "emotion engine" is an algorithm and software that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[0405] "Individually customized learning content" refers to learning information that is appropriately adjusted or generated based on the results of a user's analysis.

[0406] A "generative AI model" is an artificial intelligence model that generates learning content or other output based on a given prompt.

[0407] A "prompt" is a text instruction that can be input into a generative AI model to generate a specific output (e.g., a learning slide or explanatory text).

[0408] This system collects and analyzes users' voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to their individual needs, thereby improving their learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, this system provides highly adaptable learning support that takes into account the user's emotional state.

[0409] System Overview

[0410] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[0411] Terminal

[0412] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization. This results in high-quality data being sent to the server.

[0413] server

[0414] The server receives the preprocessed data sent from the device and analyzes it using a speech recognition engine (e.g., Google Cloud Speech-to-Text) and facial expression analysis software (e.g., Azure Face API) to evaluate the user's comprehension, concentration, and stress level. It also identifies the user's emotional state using an emotion engine (e.g., IBM Watson Tone Analyzer) and generates or selects learning content by incorporating the emotion information into the analysis results.

[0415] Emotion Engine

[0416] An emotion engine is an algorithm and software that analyzes a user's facial expression and voice data to determine their emotional state. The determined emotions are sent to a server and used to adapt the learning content.

[0417] User

[0418] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[0419] Program processing explanation

[0420] Data Collection Phase

[0421] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. This data is temporarily stored in local storage and undergoes pre-processing such as noise reduction and filtering.

[0422] Analysis Phase

[0423] The device sends the preprocessed data to a server, which uses a speech recognition engine to convert the voice data into text, facial expression and gaze data to determine the user's comprehension, concentration, and stress levels, and an emotion engine to identify the user's emotional state.

[0424] Learning content customization phase

[0425] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects individually customized learning content, using prompts to generate appropriate learning content using a generative AI model.

[0426] Presentation Phase

[0427] The server transmits the customized learning content to the terminal, and the terminal presents the received learning content to the user in real time to support the user in progressing with their learning.

[0428] Specific examples

[0429] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[0430] Prompt Sentence Examples

[0431] Examples of prompts to be input to a generative AI model include:

[0432] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[0433] The generated learning content dynamically adapts according to the user's level of understanding, concentration, and emotional state, significantly improving the user's learning efficiency.

[0434] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0435] Step 1:

[0436] Data Collection Phase

[0437] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. The data obtained from each sensor is temporarily stored in local storage. At this time, preprocessing such as noise removal and filtering is performed. The input includes voice, facial expression, and gaze data from the user, which is then output as high-quality data after noise removal and filtering.

[0438] Input: User's voice, facial expression, and gaze data

[0439] Data processing: noise removal, filtering, normalization

[0440] Output: Pre-processed high-quality data

[0441] Step 2:

[0442] Data transmission phase

[0443] The device sends the preprocessed data to the server via a secure communication protocol. The input is the preprocessed high-quality data, which is then sent to the server for the next analysis phase.

[0444] Input: Preprocessed high-quality data

[0445] Data processing: Data transfer, encryption

[0446] Output: Data sent to the server

[0447] Step 3:

[0448] Analysis Phase

[0449] The server receives the preprocessed data and converts it into text using a speech recognition engine. It also analyzes facial expression data using facial expression analysis software and evaluates gaze data, thereby determining the user's level of comprehension, concentration, and stress. The server receives the data as input and outputs it as comprehension, concentration, and stress levels through speech recognition and facial expression analysis.

[0450] Input: Data sent to the server

[0451] Data processing: voice recognition, facial expression analysis, gaze evaluation

[0452] Output: User comprehension, concentration, stress level

[0453] Step 4:

[0454] Sentiment Analysis Phase

[0455] The server uses an emotion engine to identify the user's emotional state from their facial and voice data, which allows it to evaluate the emotions (e.g., stress or anxiety) the user is experiencing during the learning process. The input is facial and voice data, which are analyzed by the emotion engine to output the emotional state.

[0456] Input: facial expression data, voice data

[0457] Data processing: Sentiment analysis

[0458] Output: Emotional state

[0459] Step 5:

[0460] Learning content generation phase

[0461] The server generates or selects individually customized learning content based on the analysis results. It uses a generative AI model to create prompts to generate appropriate learning content, and the content is generated based on these. Inputs include level of comprehension, concentration, stress level, and emotional state, and the server outputs customized learning content based on these.

[0462] Input: Comprehension, concentration, stress level, emotional state

[0463] Data processing: prompt generation, content generation

[0464] Output: Customized learning content

[0465] Step 6:

[0466] Learning content presentation phase

[0467] The server sends the generated learning content to the terminal, which then presents it to the user. The input is the customized learning content, which is presented to the user by the terminal.

[0468] Input: Customized learning content

[0469] Data processing: Data transfer

[0470] Output: Presenting the learning content to the user

[0471] Step 7:

[0472] Feedback gathering phase

[0473] The user digests the learning content and responds. The device again collects the user's voice, facial expression, and gaze data for preprocessing and analysis. This allows the device to continuously evaluate the user's learning effectiveness and further optimize the next learning path. The input is the user's response data, which is then preprocessed and analyzed to output new levels of comprehension, concentration, stress level, and emotional state.

[0474] Input: User response data

[0475] Data processing: noise removal, filtering, voice recognition, facial expression analysis, emotion analysis

[0476] Output: New understanding, focus, stress level, emotional state

[0477] Specific examples

[0478] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[0479] Prompt Sentence Examples

[0480] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[0481] 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.

[0482] 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.

[0483] 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.

[0484] [Second embodiment]

[0485] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0486] 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.

[0487] 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).

[0488] 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.

[0489] 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.

[0490] 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).

[0491] 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.

[0492] 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.

[0493] 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.

[0494] 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.

[0495] 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.

[0496] 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."

[0497] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. The following describes in detail an embodiment of the present invention.

[0498] System Overview

[0499] The system consists of three main components: terminals, servers, and users.

[0500] 1. Terminal

[0501] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0502] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0503] 2. Server

[0504] The server receives the preprocessed data sent from the terminal and analyzes it using a voice recognition engine and facial expression analysis software.

[0505] The voice data is converted into text data, and the user's level of understanding, concentration, and stress level are evaluated from facial expression and gaze data.

[0506] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[0507] 3. Users

[0508] The user digests the learning content provided via the terminal and makes the necessary responses.

[0509] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0510] Program processing

[0511] The flow of program processing in this system will be explained below.

[0512] Data Collection Phase

[0513] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0514] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0515] Analysis Phase

[0516] The terminal transmits the preprocessed data to the server.

[0517] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[0518] Learning content customization phase

[0519] Based on the analyzed data, the server evaluates the user's level of understanding and learning progress, and generates or selects customized learning content.

[0520] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0521] Presentation Phase

[0522] The server transmits customized learning content to the terminal.

[0523] The terminal presents the received learning content to the user in real time, and the user proceeds with learning in accordance with the content.

[0524] Feedback Phase

[0525] The user digests the presented learning content and responds to demonstrate the learning effect.

[0526] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0527] Specific examples

[0528] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0529] In this way, the system provides real-time adaptive learning based on the user's level of understanding and concentration.

[0530] The processing flow will be explained below.

[0531] Step 1:

[0532] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0533] Step 2:

[0534] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[0535] Step 3:

[0536] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[0537] Step 4:

[0538] The terminal transmits the preprocessed data to the server.

[0539] Step 5:

[0540] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[0541] Step 6:

[0542] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[0543] Step 7:

[0544] The server analyzes the gaze data and identifies which part the user is focusing on.

[0545] Step 8:

[0546] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[0547] Step 9:

[0548] The server generates or selects customized learning content based on the evaluation results.

[0549] Step 10:

[0550] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[0551] Step 11:

[0552] The server transmits customized learning content to the terminal.

[0553] Step 12:

[0554] The device presents the received learning content to the user in real time, supporting them in their learning.

[0555] Step 13:

[0556] The user digests the presented learning content and takes the necessary responses or actions.

[0557] Step 14:

[0558] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[0559] Step 15:

[0560] The terminal preprocesses the newly collected data and sends it back to the server.

[0561] Step 16:

[0562] The server analyzes the data again and generates the next learning path based on the feedback information.

[0563] Step 17:

[0564] The server sends the generated new learning path to the terminal, which then presents it to the user.

[0565] Step 18:

[0566] Users progress through new learning paths.

[0567] Example 1

[0568] 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."

[0569] Conventional learning systems have had difficulty adjusting content based on the user's real-time status and reactions in order to maximize the user's learning efficiency. This has resulted in the inability to respond immediately when the user stumbles on a particular problem or point, resulting in a decrease in learning efficiency. Therefore, there is a need for a system that can dynamically customize learning content based on the user's real-time data and quickly respond to each user's learning needs.

[0570] 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.

[0571] In this invention, the server includes means for collecting voice, facial expression, and gaze data of a user in real time, means for preprocessing the voice, facial expression, and gaze data, and means for analyzing the preprocessed data to evaluate the user's level of understanding, concentration, and stress level, thereby enabling dynamic customization of learning content based on the user's real-time data to meet individual learning needs.

[0572] "Voice data" refers to data consisting of user utterances, sounds, and linguistic information.

[0573] "Facial expression data" is data that includes information about the user's facial expression.

[0574] "Gaze data" is data that includes information about the user's eye movements and viewpoint.

[0575] "Preprocessing" refers to processing the collected data, such as noise removal and filtering, to make it easier to analyze.

[0576] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[0577] "Learning content" refers to content such as educational materials, questions, and explanatory videos provided to support users' learning.

[0578] "Customized learning content" is learning content that is adapted based on a user's individual needs and learning progress.

[0579] "Batch transmission" is a method of transmitting multiple pieces of data together at regular time intervals.

[0580] "Facial expression analysis software" is software that analyzes a user's facial expressions to evaluate their emotions and state.

[0581] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0582] "Noise reduction" is a process of removing unnecessary noise from audio data and the like.

[0583] "Level of concentration" is a measure of how much the user is concentrating on their studies.

[0584] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[0585] The present invention is a system that collects and analyzes voice, facial expression, and gaze data of a user in real time to improve the user's learning efficiency, and dynamically provides learning content tailored to individual needs. Specific embodiments for implementing the present invention will be described in detail below.

[0586] System Overview

[0587] The system consists of three main components: terminals, servers, and users.

[0588] 1. Terminal

[0589] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0590] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0591] 2. Server

[0592] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The voice data is converted into text data, and the user's level of comprehension, concentration, and stress level are evaluated based on facial expression and gaze data.

[0593] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[0594] 3. Users

[0595] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[0596] Specific examples of programs

[0597] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0598] Hardware and software used

[0599] Speech recognition engine: For example, use the Google Speech-to-Text API to convert voice data into text data.

[0600] Facial expression analysis software: For example, using Microsoft Azure Face API to analyze the user's facial expressions.

[0601] Eye-tracking software: In combination with a camera, it collects and analyzes user gaze data.

[0602] Examples of prompt statements

[0603] For example, enter the following prompt sentence into the generative AI model:

[0604] "Design a system that can detect in real time when you are struggling with a problem and provide you with the appropriate learning content. For example, it would be even better if the system could detect when a student is struggling with a problem and provide supplementary materials."

[0605] summary

[0606] The present invention is a system that dynamically customizes learning content based on users' real-time data to meet their individual learning needs, thereby maximizing the user's learning efficiency and supporting rapid problem solving.

[0607] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0608] Step 1:

[0609] Starting the sensor

[0610] The device activates the camera, microphone, and other sensors to begin collecting data from the user. For example, the device automatically activates each sensor the moment the user starts learning content. The input is the user's action to start the learning content, and the output is each activated sensor.

[0611] Step 2:

[0612] Data collection

[0613] The device collects the user's voice, facial expression, and gaze data in real time. For example, a camera captures the user's facial expressions and a microphone records what the user says. The input is the real-time user data captured by the sensor, and the output is the collected raw data.

[0614] Step 3:

[0615] Temporary storage and preprocessing of data

[0616] The device temporarily stores the collected data in local storage and performs preprocessing such as noise reduction and filtering. For example, a noise reduction algorithm is applied to clear voice data collected in a noisy environment. The input is the collected raw data, and the output is the preprocessed data.

[0617] Step 4:

[0618] Sending preprocessed data

[0619] The terminal sends the pre-processed data to the server, for example, by batch transmission of multiple data at regular intervals. The input is the pre-processed data, and the output is the data sent to the server.

[0620] Step 5:

[0621] Analysis of audio data

[0622] The server converts the voice data into text data using a speech recognition engine (e.g., Google Speech-to-Text API). For example, a user's statement "I don't understand this problem" is converted into text data. The input is the transmitted voice data, and the output is the converted text data.

[0623] Step 6:

[0624] Analysis of facial expression and gaze data

[0625] The server uses facial expression analysis software (e.g., Microsoft Azure Face API) and eye-tracking software to determine the user's level of comprehension, concentration, and stress. For example, it can detect when the user frowns and determine that the user's stress level is high. The input is the transmitted facial expression and eye-gaze data, and the output is the analyzed user's level of comprehension, concentration, and stress.

[0626] Step 7:

[0627] Assessment of understanding and progress

[0628] The server evaluates the user's level of understanding and learning progress based on the analyzed data. For example, it determines the level of understanding based on data such as which questions the user spent the most time on. The input is the analyzed user data, and the output is the evaluation results.

[0629] Step 8:

[0630] Content generation or selection

[0631] The server generates or selects customized learning content and adjusts the content. For example, if a user has difficulty with "factorization," it selects videos or additional practice problems to refresh the knowledge from the basics. The input is the assessment results, and the output is customized learning content.

[0632] Step 9:

[0633] Sending customized content

[0634] The server sends customized learning content to the device. For example, it sends a specific video link or PDF file to the device. The input is the customized learning content, and the output is the data sent to the device.

[0635] Step 10:

[0636] View content

[0637] The device displays the received learning content to the user, and the user proceeds with the learning process. For example, a video for problem solving is displayed on the screen. The input is the transmitted learning content data, and the output is the content displayed to the user.

[0638] Step 11:

[0639] Collecting user learning status

[0640] The user digests the presented learning content and responds. For example, after solving a problem, the user may move on to the next problem or respond by saying "I couldn't solve it." The input is the user's learning behavior, and the output is learning behavior data.

[0641] Step 12:

[0642] Reaction data collection

[0643] The device again collects the user's voice, facial expression, and gaze data, and performs preprocessing and analysis again. For example, if the user makes a new confused expression during learning, that data will also be collected. The input is the newly collected data, and the output is the data that will be preprocessed again.

[0644] (Application example 1)

[0645] 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."

[0646] On manufacturing lines, it is important for workers to quickly acquire skills and improve production efficiency. However, it is difficult to provide training content that is tailored to each worker's level of understanding and progress, and general manuals and videos alone are often insufficient. It is also difficult to grasp in real time which parts of the training workers are struggling with. This reduces the effectiveness of training and leads to issues such as lower production efficiency.

[0647] 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.

[0648] In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for generating or selecting learning content based on the evaluation results, means for presenting the learning content to the user, means for collecting user reaction data and preprocessing and analyzing it again, and means for providing individual training content to production line workers by collecting and analyzing voice, facial expression, and gaze data in real time, thereby enabling the provision of optimal training content tailored to the worker's level of understanding and progress.

[0649] "Collecting user voice, facial expression, and gaze data" refers to sensing and recording the user's voice, facial expression changes, and gaze movements in real time.

[0650] "Preprocessing" refers to a series of operations that remove noise from collected raw data and prepare the data in a usable form.

[0651] "Data analysis" refers to extracting information such as the user's level of understanding, concentration, and stress level from the pre-processed data using specific algorithms or models.

[0652] "Generation or selection of learning content" refers to the dynamic creation or selection of teaching materials or information to provide users with the information or training they need based on the analysis results.

[0653] "Presenting learning content" refers to displaying generated or selected learning content to a user in real time and allowing the user to use the content.

[0654] "Collecting user reaction data" refers to re-detecting and recording the user's behavior, comments, facial expressions, and eye movements while studying.

[0655] "Providing training content to workers on the production line" refers to dynamically presenting content to individual workers working on the manufacturing site to enable them to learn the necessary knowledge and skills on the spot.

[0656] "Individualized training content" refers to training content and learning materials that are customized to the specific needs and level of understanding of the user (worker).

[0657] System Overview

[0658] This invention is a system that collects voice, facial expression, and gaze data of workers in real time and analyzes the data to dynamically provide training content tailored to individual needs in order to improve the learning efficiency of workers on a production line. Specific embodiments for implementing this invention will be described below.

[0659] Main configuration

[0660] The system consists of the following main components:

[0661] 1. Terminal

[0662] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect voice, facial expression, and gaze data of workers in real time.

[0663] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0664] 2. Server

[0665] The server receives the preprocessed data sent from the terminal and analyzes it using a speech recognition engine (for example, the speech_recognition library) or facial expression analysis software (for example, the EmotionRecognition library).

[0666] The voice data is converted into text data, and the worker's level of understanding, concentration, and stress level are evaluated based on facial expression and gaze data.

[0667] Based on the analysis results, training content is generated or selected and dynamically adapted.

[0668] 3. User (operator)

[0669] The worker executes the training content provided via the terminal and makes the necessary responses.

[0670] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0671] Program processing

[0672] The flow of program processing in this system will be explained below.

[0673] Data Collection Phase

[0674] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time.

[0675] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0676] Analysis Phase

[0677] The terminal transmits the preprocessed data to the server.

[0678] The server converts the voice data into text data using a speech recognition engine (speech_recognition library), and uses facial expression and gaze data to determine the worker's level of understanding, concentration, and stress.

[0679] Learning content customization phase

[0680] Based on the analyzed data, the server evaluates the worker's level of understanding and learning progress, and generates or selects customized training content.

[0681] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0682] Presentation Phase

[0683] The server transmits customized learning content to the terminal.

[0684] The terminal presents the received training content to the worker in real time, and the worker follows along with the content.

[0685] Feedback Phase

[0686] The worker performs the presented training content and provides responses that demonstrate the effectiveness of their learning.

[0687] The device again collects data such as the worker's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0688] Specific examples

[0689] For example, if a new worker is confused about a particular procedure on a production line, the device can detect this situation through the camera and microphone. Specifically, it analyzes the worker's confused facial expression and voice, such as "What should I do about this?" Based on this, the server can provide the worker with a training video or additional explanations that are best suited to the worker's basic knowledge. This method can help new workers to proceed with their work smoothly.

[0690] Prompt Sentence Examples

[0691] If a new worker looks confused and says in audio, "What do I do about this?", suggest a training video that will help new workers review the basics.

[0692] This system provides real-time adaptive training that matches the worker's level of understanding and progress, improving the efficiency and quality of the production line.

[0693] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0694] Step 1: Data collection phase

[0695] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time. The input of this data collection is the worker's real-time activity data, and the output is audio files, image data, and gaze tracking data. This data is temporarily stored in local storage and undergoes preprocessing such as noise removal and filtering. For example, background noise is removed from the voice data and unnecessary background is cut out from the image data.

[0696] Step 2: Preprocessing phase

[0697] The device preprocesses the collected data. The input is the voice, facial expression, and gaze data collected in step 1, and the output is data that has been subjected to noise removal and filtering. Specifically, the waveform of the voice data is flattened, and histogram normalization is performed on the image data. The gaze data is also filtered and blurring is corrected.

[0698] Step 3: Data transmission phase

[0699] The device sends the preprocessed data to the server. The input is the preprocessed voice, facial expression, and gaze data, and the output is the data sent to the server. For example, the device divides the data into packets and sends them to the server.

[0700] Step 4: Analysis Phase

[0701] The server receives the transmitted data and converts the voice data into text using a speech recognition engine (e.g., speech_recognition library). The input is the preprocessed data transmitted from the device, and the output is text data and analysis data (level of understanding, concentration, stress level, etc.). Specifically, the server analyzes the facial expression data using facial expression analysis software (e.g., EmotionRecognition library) to determine the user's level of understanding. It also applies an algorithm to evaluate the level of concentration based on gaze data.

[0702] Step 5: Customizing the learning content

[0703] The server evaluates the worker's level of understanding and learning progress based on the analyzed data, and generates or selects customized training content. The input is data such as the worker's level of understanding, concentration, and stress level obtained in the analysis phase, and the output is training content dynamically created based on this data. Specifically, the server selects video materials and interactive simulations that are optimal for the worker's level of progress.

[0704] Step 6: Content Presentation Phase

[0705] The server sends customized training content to the terminal, which then presents it to the worker. The input is the training content sent from the server, and the output is the training video or simulation that the worker views. Specifically, the terminal displays the received content on a display and presents it to the worker by providing audio guidance.

[0706] Step 7: Feedback Phase

[0707] The user digests the presented training content and responds to demonstrate the learning effect. The input is the user's input (voice response, facial expression, and gaze data), and the output is the collected response data. For example, there are situations where a worker answers questions about the training content by voice. This response data is collected again by the system, and the next learning path is further optimized.

[0708] This process allows workers to receive effective training tailored to their individual needs.

[0709] 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.

[0710] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that also takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[0711] System Overview

[0712] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[0713] 1. Terminal

[0714] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0715] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0716] 2. Server

[0717] The server receives the pre-processed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software to assess the user's level of understanding, concentration, and stress level.

[0718] The server further identifies the user's emotional state using an emotion engine and generates or selects learning content by taking the emotion information into account in the analysis results.

[0719] 3. Emotion Engine

[0720] The emotion engine analyzes the user's facial expression data and voice data to determine the user's emotional state.

[0721] The determined emotions are sent to the server and used to adapt the learning content.

[0722] 4. Users

[0723] The user digests the learning content provided via the terminal and makes the necessary responses.

[0724] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0725] Program processing

[0726] The flow of program processing in this system will be explained below.

[0727] Data Collection Phase

[0728] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0729] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0730] Analysis Phase

[0731] The terminal transmits the preprocessed data to the server.

[0732] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[0733] The server uses an emotion engine to analyze the user's facial expression data and voice data to identify the user's emotional state.

[0734] Learning content customization phase

[0735] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects customized learning content.

[0736] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0737] Presentation Phase

[0738] The server transmits customized learning content to the terminal.

[0739] The device presents the received learning content to the user in real time, supporting them in their learning.

[0740] Feedback Phase

[0741] The user digests the presented learning content and responds to demonstrate the learning effect.

[0742] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0743] Specific examples

[0744] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0745] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[0746] The processing flow will be explained below.

[0747] Step 1:

[0748] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0749] Step 2:

[0750] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[0751] Step 3:

[0752] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[0753] Step 4:

[0754] The terminal transmits the preprocessed data to the server.

[0755] Step 5:

[0756] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[0757] Step 6:

[0758] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[0759] Step 7:

[0760] The server analyzes the gaze data and identifies which part the user is focusing on.

[0761] Step 8:

[0762] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[0763] Step 9:

[0764] The server uses an emotion engine to analyze the facial expression data and voice data to identify the user's emotional state.

[0765] Step 10:

[0766] The server generates or selects customized learning content based on the evaluation results and emotional information.

[0767] Step 11:

[0768] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[0769] Step 12:

[0770] The server transmits customized learning content to the terminal.

[0771] Step 13:

[0772] The device presents the received learning content to the user in real time, supporting them in their learning.

[0773] Step 14:

[0774] The user digests the presented learning content and takes the necessary responses or actions.

[0775] Step 15:

[0776] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[0777] Step 16:

[0778] The terminal preprocesses the newly collected data and sends it back to the server.

[0779] Step 17:

[0780] The server analyzes the data again and generates the next learning path based on the feedback information.

[0781] Step 18:

[0782] The server sends the generated new learning path to the terminal, which then presents it to the user.

[0783] Step 19:

[0784] Users progress through new learning paths.

[0785] Example 2

[0786] 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."

[0787] Conventional learning support systems have difficulty accurately grasping a user's level of understanding and concentration, making it impossible to provide content tailored to individual learning needs. Furthermore, they often provide uniform learning content without considering the user's emotional state, limiting the improvement of learning efficiency. Furthermore, they do not dynamically adapt content to match the user's learning progress, making it difficult to motivate users or provide an effective learning experience.

[0788] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for determining the user's emotional state using the evaluation results and an emotion engine, means for generating or selecting learning content based on the evaluation results and emotion information, means for presenting the learning content to the user, and means for collecting user reaction data and re-preprocessing and analyzing the data. This makes it possible to provide individually customized learning content that takes into account the user's emotional state as well as their level of understanding and concentration.

[0789] "Voice data" refers to data in which the user's voice is recorded in digital format.

[0790] "Facial expression data" refers to data that captures a user's facial expression using a camera or the like and converts it into an analyzable format.

[0791] "Gaze data" refers to data that measures and records the user's eye movements and gaze direction.

[0792] "Preprocessing" is the process of performing processes such as noise removal and filtering on collected raw data to convert it into a format suitable for analysis.

[0793] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[0794] "Evaluation" is the process of determining the user's learning status according to specific criteria based on the information obtained through analysis.

[0795] An "emotion engine" is software or hardware that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[0796] "Learning content" is a general term for teaching materials and educational resources presented to users, and includes text, video, audio, practice questions, and the like.

[0797] "Customization" is the process of optimizing learning content according to each user's needs and learning situation.

[0798] "Presenting" refers to the act of displaying or playing selected or generated learning content to a user.

[0799] "Reaction data" is data that records the user's reactions to learning content, including voice, facial expressions, and gaze.

[0800] "Dynamic adaptation" is the process of adjusting and providing learning content in real time based on the user's learning progress, level of understanding, and emotional state.

[0801] A "voice recognition engine" is software or hardware that converts voice data into text data.

[0802] MODE FOR CARRYING OUT THE INVENTION

[0803] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by using an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[0804] System configuration

[0805] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[0806] Terminal

[0807] The device is equipped with a camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0808] server

[0809] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The server evaluates the user's level of understanding, concentration, and stress level. It also identifies the user's emotional state using an emotion engine, and generates or selects learning content by incorporating emotional information into the analysis results.

[0810] Emotion Engine

[0811] The emotion engine analyzes the user's facial expression and voice data to determine their emotional state. The determined emotions are sent to the server and used to adapt the learning content.

[0812] User

[0813] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[0814] System operation example

[0815] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0816] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[0817] Example prompts to input to the generative AI model

[0818] Below are some example prompts to input to a generative AI model:

[0819] "Based on a scenario where a student is working on a math problem but is having trouble with a particular one, describe a system that analyzes the user's level of understanding and emotional state in real time and provides customized learning content."

[0820] This prompt-based system optimizes the user's learning efficiency and provides advanced learning support that meets individual needs.

[0821] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0822] Step 1: Data collection

[0823] The device activates sensors such as a camera, microphone, and gaze tracker to collect the user's voice, facial expression, and gaze data in real time. This provides basic data for understanding the user's behavior and reactions in real time. For example, a user may have a confused expression and say, "I don't understand this problem." The input data obtained here are voice data, facial expression data, and gaze data, which are collected by the device.

[0824] Step 2: Data Preprocessing

[0825] The device temporarily stores the collected data in local storage and performs preprocessing such as noise removal and filtering. As a result of preprocessing, the data is in a clean format suitable for analysis. For example, background noise is removed from voice data and facial expression data is normalized. The input data is the raw data collected in step 1, and the output data is the preprocessed, clean data.

[0826] Step 3: Send data

[0827] The device divides the preprocessed data into packets and sends them to the server. Highly efficient data transfer is important in this step. The input data is the preprocessed data, and the output is the data sent to the server. For example, the device may use a compression algorithm to reduce the size of the data before sending it to the server.

[0828] Step 4: Convert audio data to text

[0829] The server converts the received voice data into text data using a voice recognition engine. For example, the voice saying "I don't understand this problem" is converted into text "I don't understand this problem." The input data is preprocessed voice data, and the output data is text data.

[0830] Step 5: Analyze facial expression data

[0831] The server uses facial expression analysis software to analyze the user's facial expression data to determine their level of understanding, concentration, and stress. For example, a confused expression can indicate a low level of understanding. The input data is the preprocessed facial expression data, and the output data is the analysis result.

[0832] Step 6: Analyze gaze data

[0833] The server analyzes the gaze tracker data to determine where the user is looking. For example, if the user looks at a particular problem area for a long time, it collects that information. The input data is the preprocessed gaze data, and the output data is the gaze focus information.

[0834] Step 7: Identify your emotional state

[0835] The server uses an emotion engine to identify the user's emotional state based on facial expression and voice data. For example, it determines whether the user is feeling stressed. The input data is the analyzed facial expression and voice data, and the output data is the result of the emotional state determination.

[0836] Step 8: Assess your learning progress

[0837] The server evaluates the user's learning progress based on the collected and analyzed data. For example, it identifies areas where the user is struggling with a particular topic. The input data are the analysis results and the judgment of the user's emotional state, and the output data is the evaluation of the user's learning progress.

[0838] Step 9: Content Selection and Creation

[0839] The server selects or generates optimal learning content for the user, such as basic instructional videos or practice questions. The input data is the evaluation results of the learning progress, and the output data is the selected or generated learning content.

[0840] Step 10: Present content to the user

[0841] The server sends the customized learning content to the device, which then presents it to the user, for example, by displaying and playing a video on the screen. The input data is the selected and generated learning content, and the output data is the presented learning content.

[0842] Step 11: Collect user responses

[0843] The user digests the learning content and responds with quizzes or speech to check their understanding. For example, the user answers questions in a quiz. The input data is the user's response, and the output data is the collected response data.

[0844] Step 12: Recollection and analysis

[0845] The device again collects data such as the user's voice, facial expression, and gaze, and sends it to the server for preprocessing and analysis. For example, it re-analyzes how the user is struggling with a quiz. The input data is the user's voice, facial expression, and gaze data, and the output data is the re-analysis result.

[0846] (Application example 2)

[0847] 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."

[0848] Conventional learning support systems have difficulty in comprehensively evaluating a user's level of understanding, concentration, and emotional state, resulting in problems with the provided learning content not being optimized for each individual user. Furthermore, it is difficult to identify in real time where a user is struggling and provide appropriate feedback. Furthermore, they lack the ability to adaptively generate and select learning content, which reduces the user's learning efficiency.

[0849] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0850] In this invention, the server includes means for collecting a user's voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for identifying the user's emotional state using an emotion engine and using the identified emotional state to adapt content, means for generating or selecting individually customized learning content based on the analysis results, means for presenting the generated or selected learning content to the user and prompting the user to respond to demonstrate learning effectiveness, and means for generating prompts using a generative AI model to generate appropriate learning content. This improves user learning efficiency and enables the provision of content optimized for each individual user in real time.

[0851] "Voice Data" means data in digital form that records a user's speech or other vocal communication.

[0852] "Facial expression data" is digital data acquired in real time based on the movement of the user's facial muscles.

[0853] "Gaze data" is digital data that tracks the user's eye movements and records where they are looking in real time.

[0854] "Preprocessing" refers to the process of converting collected data into a format that is easy to analyze by performing processes such as noise removal, filtering, and normalization on the data.

[0855] "Analysis" refers to data processing to assess the user's level of understanding, concentration, stress level, and emotional state based on the preprocessed data.

[0856] "Level of understanding" is an index that indicates how well a user understands the learning content.

[0857] "Concentration level" is an index that indicates how much a user is concentrating on a learning task.

[0858] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[0859] An "emotion engine" is an algorithm and software that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[0860] "Individually customized learning content" refers to learning information that is appropriately adjusted or generated based on the results of a user's analysis.

[0861] A "generative AI model" is an artificial intelligence model that generates learning content or other output based on a given prompt.

[0862] A "prompt" is a text instruction that can be input into a generative AI model to generate a specific output (e.g., a learning slide or explanatory text).

[0863] This system collects and analyzes users' voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to their individual needs, thereby improving their learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, this system provides highly adaptable learning support that takes into account the user's emotional state.

[0864] System Overview

[0865] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[0866] Terminal

[0867] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization. This results in high-quality data being sent to the server.

[0868] server

[0869] The server receives the preprocessed data sent from the device and analyzes it using a speech recognition engine (e.g., Google Cloud Speech-to-Text) and facial expression analysis software (e.g., Azure Face API) to evaluate the user's comprehension, concentration, and stress level. It also identifies the user's emotional state using an emotion engine (e.g., IBM Watson Tone Analyzer) and generates or selects learning content by incorporating the emotion information into the analysis results.

[0870] Emotion Engine

[0871] An emotion engine is an algorithm and software that analyzes a user's facial expression and voice data to determine their emotional state. The determined emotions are sent to a server and used to adapt the learning content.

[0872] User

[0873] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[0874] Program processing explanation

[0875] Data Collection Phase

[0876] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. This data is temporarily stored in local storage and undergoes pre-processing such as noise reduction and filtering.

[0877] Analysis Phase

[0878] The device sends the preprocessed data to a server, which uses a speech recognition engine to convert the voice data into text, facial expression and gaze data to determine the user's comprehension, concentration, and stress levels, and an emotion engine to identify the user's emotional state.

[0879] Learning content customization phase

[0880] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects individually customized learning content, using prompts to generate appropriate learning content using a generative AI model.

[0881] Presentation Phase

[0882] The server transmits the customized learning content to the terminal, and the terminal presents the received learning content to the user in real time to support the user in progressing with their learning.

[0883] Specific examples

[0884] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[0885] Prompt Sentence Examples

[0886] Examples of prompts to be input to a generative AI model include:

[0887] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[0888] The generated learning content dynamically adapts according to the user's level of understanding, concentration, and emotional state, significantly improving the user's learning efficiency.

[0889] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0890] Step 1:

[0891] Data Collection Phase

[0892] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. The data obtained from each sensor is temporarily stored in local storage. At this time, preprocessing such as noise removal and filtering is performed. The input includes voice, facial expression, and gaze data from the user, which is then output as high-quality data after noise removal and filtering.

[0893] Input: User's voice, facial expression, and gaze data

[0894] Data processing: noise removal, filtering, normalization

[0895] Output: Pre-processed high-quality data

[0896] Step 2:

[0897] Data transmission phase

[0898] The device sends the preprocessed data to the server via a secure communication protocol. The input is the preprocessed high-quality data, which is then sent to the server for the next analysis phase.

[0899] Input: Preprocessed high-quality data

[0900] Data processing: Data transfer, encryption

[0901] Output: Data sent to the server

[0902] Step 3:

[0903] Analysis Phase

[0904] The server receives the preprocessed data and converts it into text using a speech recognition engine. It also analyzes facial expression data using facial expression analysis software and evaluates gaze data, thereby determining the user's level of comprehension, concentration, and stress. The server receives the data as input and outputs it as comprehension, concentration, and stress levels through speech recognition and facial expression analysis.

[0905] Input: Data sent to the server

[0906] Data processing: voice recognition, facial expression analysis, gaze evaluation

[0907] Output: User comprehension, concentration, stress level

[0908] Step 4:

[0909] Sentiment Analysis Phase

[0910] The server uses an emotion engine to identify the user's emotional state from their facial and voice data, which allows it to evaluate the emotions (e.g., stress or anxiety) the user is experiencing during the learning process. The input is facial and voice data, which are analyzed by the emotion engine to output the emotional state.

[0911] Input: facial expression data, voice data

[0912] Data processing: Sentiment analysis

[0913] Output: Emotional state

[0914] Step 5:

[0915] Learning content generation phase

[0916] The server generates or selects individually customized learning content based on the analysis results. It uses a generative AI model to create prompts to generate appropriate learning content, and the content is generated based on these. Inputs include level of comprehension, concentration, stress level, and emotional state, and the server outputs customized learning content based on these.

[0917] Input: Comprehension, concentration, stress level, emotional state

[0918] Data processing: prompt generation, content generation

[0919] Output: Customized learning content

[0920] Step 6:

[0921] Learning content presentation phase

[0922] The server sends the generated learning content to the terminal, which then presents it to the user. The input is the customized learning content, which is presented to the user by the terminal.

[0923] Input: Customized learning content

[0924] Data processing: Data transfer

[0925] Output: Presenting the learning content to the user

[0926] Step 7:

[0927] Feedback gathering phase

[0928] The user digests the learning content and responds. The device again collects the user's voice, facial expression, and gaze data for preprocessing and analysis. This allows the device to continuously evaluate the user's learning effectiveness and further optimize the next learning path. The input is the user's response data, which is then preprocessed and analyzed to output new levels of comprehension, concentration, stress level, and emotional state.

[0929] Input: User response data

[0930] Data processing: noise removal, filtering, voice recognition, facial expression analysis, emotion analysis

[0931] Output: New understanding, focus, stress level, emotional state

[0932] Specific examples

[0933] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[0934] Prompt Sentence Examples

[0935] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[0936] 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.

[0937] 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.

[0938] 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.

[0939] [Third embodiment]

[0940] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0941] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0942] 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).

[0943] 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.

[0944] 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.

[0945] 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).

[0946] 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.

[0947] 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.

[0948] 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.

[0949] 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.

[0950] 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.

[0951] 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."

[0952] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. The following describes in detail an embodiment of the present invention.

[0953] System Overview

[0954] The system consists of three main components: terminals, servers, and users.

[0955] 1. Terminal

[0956] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[0957] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[0958] 2. Server

[0959] The server receives the preprocessed data sent from the terminal and analyzes it using a voice recognition engine and facial expression analysis software.

[0960] The voice data is converted into text data, and the user's level of understanding, concentration, and stress level are evaluated from facial expression and gaze data.

[0961] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[0962] 3. Users

[0963] The user digests the learning content provided via the terminal and makes the necessary responses.

[0964] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[0965] Program processing

[0966] The flow of program processing in this system will be explained below.

[0967] Data Collection Phase

[0968] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0969] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[0970] Analysis Phase

[0971] The terminal transmits the preprocessed data to the server.

[0972] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[0973] Learning content customization phase

[0974] Based on the analyzed data, the server evaluates the user's level of understanding and learning progress, and generates or selects customized learning content.

[0975] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[0976] Presentation Phase

[0977] The server transmits customized learning content to the terminal.

[0978] The terminal presents the received learning content to the user in real time, and the user proceeds with learning in accordance with the content.

[0979] Feedback Phase

[0980] The user digests the presented learning content and responds to demonstrate the learning effect.

[0981] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[0982] Specific examples

[0983] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[0984] In this way, the system provides real-time adaptive learning based on the user's level of understanding and concentration.

[0985] The processing flow will be explained below.

[0986] Step 1:

[0987] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[0988] Step 2:

[0989] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[0990] Step 3:

[0991] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[0992] Step 4:

[0993] The terminal transmits the preprocessed data to the server.

[0994] Step 5:

[0995] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[0996] Step 6:

[0997] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[0998] Step 7:

[0999] The server analyzes the gaze data and identifies which part the user is focusing on.

[1000] Step 8:

[1001] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[1002] Step 9:

[1003] The server generates or selects customized learning content based on the evaluation results.

[1004] Step 10:

[1005] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[1006] Step 11:

[1007] The server transmits customized learning content to the terminal.

[1008] Step 12:

[1009] The device presents the received learning content to the user in real time, supporting them in their learning.

[1010] Step 13:

[1011] The user digests the presented learning content and takes the necessary responses or actions.

[1012] Step 14:

[1013] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[1014] Step 15:

[1015] The terminal preprocesses the newly collected data and sends it back to the server.

[1016] Step 16:

[1017] The server analyzes the data again and generates the next learning path based on the feedback information.

[1018] Step 17:

[1019] The server sends the generated new learning path to the terminal, which then presents it to the user.

[1020] Step 18:

[1021] Users progress through new learning paths.

[1022] Example 1

[1023] 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."

[1024] Conventional learning systems have had difficulty adjusting content based on the user's real-time status and reactions in order to maximize the user's learning efficiency. This has resulted in the inability to respond immediately when the user stumbles on a particular problem or point, resulting in a decrease in learning efficiency. Therefore, there is a need for a system that can dynamically customize learning content based on the user's real-time data and quickly respond to each user's learning needs.

[1025] 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.

[1026] In this invention, the server includes means for collecting voice, facial expression, and gaze data of a user in real time, means for preprocessing the voice, facial expression, and gaze data, and means for analyzing the preprocessed data to evaluate the user's level of understanding, concentration, and stress level, thereby enabling dynamic customization of learning content based on the user's real-time data to meet individual learning needs.

[1027] "Voice data" refers to data consisting of user utterances, sounds, and linguistic information.

[1028] "Facial expression data" is data that includes information about the user's facial expression.

[1029] "Gaze data" is data that includes information about the user's eye movements and viewpoint.

[1030] "Preprocessing" refers to processing the collected data, such as noise removal and filtering, to make it easier to analyze.

[1031] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[1032] "Learning content" refers to content such as educational materials, questions, and explanatory videos provided to support users' learning.

[1033] "Customized learning content" is learning content that is adapted based on a user's individual needs and learning progress.

[1034] "Batch transmission" is a method of transmitting multiple pieces of data together at regular time intervals.

[1035] "Facial expression analysis software" is software that analyzes a user's facial expressions to evaluate their emotions and state.

[1036] A "voice recognition engine" is software or hardware for converting voice data into text data.

[1037] "Noise reduction" is a process of removing unnecessary noise from audio data and the like.

[1038] "Level of concentration" is a measure of how much the user is concentrating on their studies.

[1039] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[1040] The present invention is a system that collects and analyzes voice, facial expression, and gaze data of a user in real time to improve the user's learning efficiency, and dynamically provides learning content tailored to individual needs. Specific embodiments for implementing the present invention will be described in detail below.

[1041] System Overview

[1042] The system consists of three main components: terminals, servers, and users.

[1043] 1. Terminal

[1044] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[1045] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1046] 2. Server

[1047] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The voice data is converted into text data, and the user's level of comprehension, concentration, and stress level are evaluated based on facial expression and gaze data.

[1048] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[1049] 3. Users

[1050] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[1051] Specific examples of programs

[1052] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1053] Hardware and software used

[1054] Speech recognition engine: For example, use the Google Speech-to-Text API to convert voice data into text data.

[1055] Facial expression analysis software: For example, using Microsoft Azure Face API to analyze the user's facial expressions.

[1056] Eye-tracking software: In combination with a camera, it collects and analyzes user gaze data.

[1057] Examples of prompt statements

[1058] For example, enter the following prompt sentence into the generative AI model:

[1059] "Design a system that can detect in real time when you are struggling with a problem and provide you with the appropriate learning content. For example, it would be even better if the system could detect when a student is struggling with a problem and provide supplementary materials."

[1060] summary

[1061] The present invention is a system that dynamically customizes learning content based on users' real-time data to meet their individual learning needs, thereby maximizing the user's learning efficiency and supporting rapid problem solving.

[1062] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1063] Step 1:

[1064] Starting the sensor

[1065] The device activates the camera, microphone, and other sensors to begin collecting data from the user. For example, the device automatically activates each sensor the moment the user starts learning content. The input is the user's action to start the learning content, and the output is each activated sensor.

[1066] Step 2:

[1067] Data collection

[1068] The device collects the user's voice, facial expression, and gaze data in real time. For example, a camera captures the user's facial expressions and a microphone records what the user says. The input is the real-time user data captured by the sensor, and the output is the collected raw data.

[1069] Step 3:

[1070] Temporary storage and preprocessing of data

[1071] The device temporarily stores the collected data in local storage and performs preprocessing such as noise reduction and filtering. For example, a noise reduction algorithm is applied to clear voice data collected in a noisy environment. The input is the collected raw data, and the output is the preprocessed data.

[1072] Step 4:

[1073] Sending preprocessed data

[1074] The terminal sends the pre-processed data to the server, for example, by batch transmission of multiple data at regular intervals. The input is the pre-processed data, and the output is the data sent to the server.

[1075] Step 5:

[1076] Analysis of audio data

[1077] The server converts the voice data into text data using a speech recognition engine (e.g., Google Speech-to-Text API). For example, a user's statement "I don't understand this problem" is converted into text data. The input is the transmitted voice data, and the output is the converted text data.

[1078] Step 6:

[1079] Analysis of facial expression and gaze data

[1080] The server uses facial expression analysis software (e.g., Microsoft Azure Face API) and eye-tracking software to determine the user's level of comprehension, concentration, and stress. For example, it can detect when the user frowns and determine that the user's stress level is high. The input is the transmitted facial expression and eye-gaze data, and the output is the analyzed user's level of comprehension, concentration, and stress.

[1081] Step 7:

[1082] Assessment of understanding and progress

[1083] The server evaluates the user's level of understanding and learning progress based on the analyzed data. For example, it determines the level of understanding based on data such as which questions the user spent the most time on. The input is the analyzed user data, and the output is the evaluation results.

[1084] Step 8:

[1085] Content generation or selection

[1086] The server generates or selects customized learning content and adjusts the content. For example, if a user has difficulty with "factorization," it selects videos or additional practice problems to refresh the knowledge from the basics. The input is the assessment results, and the output is customized learning content.

[1087] Step 9:

[1088] Sending customized content

[1089] The server sends customized learning content to the device. For example, it sends a specific video link or PDF file to the device. The input is the customized learning content, and the output is the data sent to the device.

[1090] Step 10:

[1091] View content

[1092] The device displays the received learning content to the user, and the user proceeds with the learning process. For example, a video for problem solving is displayed on the screen. The input is the transmitted learning content data, and the output is the content displayed to the user.

[1093] Step 11:

[1094] Collecting user learning status

[1095] The user digests the presented learning content and responds. For example, after solving a problem, the user may move on to the next problem or respond by saying "I couldn't solve it." The input is the user's learning behavior, and the output is learning behavior data.

[1096] Step 12:

[1097] Reaction data collection

[1098] The device again collects the user's voice, facial expression, and gaze data, and performs preprocessing and analysis again. For example, if the user makes a new confused expression during learning, that data will also be collected. The input is the newly collected data, and the output is the data that will be preprocessed again.

[1099] (Application example 1)

[1100] 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."

[1101] On manufacturing lines, it is important for workers to quickly acquire skills and improve production efficiency. However, it is difficult to provide training content that is tailored to each worker's level of understanding and progress, and general manuals and videos alone are often insufficient. It is also difficult to grasp in real time which parts of the training workers are struggling with. This reduces the effectiveness of training and leads to issues such as lower production efficiency.

[1102] 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.

[1103] In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for generating or selecting learning content based on the evaluation results, means for presenting the learning content to the user, means for collecting user reaction data and preprocessing and analyzing it again, and means for providing individual training content to production line workers by collecting and analyzing voice, facial expression, and gaze data in real time, thereby enabling the provision of optimal training content tailored to the worker's level of understanding and progress.

[1104] "Collecting user voice, facial expression, and gaze data" refers to sensing and recording the user's voice, facial expression changes, and gaze movements in real time.

[1105] "Preprocessing" refers to a series of operations that remove noise from collected raw data and prepare the data in a usable form.

[1106] "Data analysis" refers to extracting information such as the user's level of understanding, concentration, and stress level from the pre-processed data using specific algorithms or models.

[1107] "Generation or selection of learning content" refers to the dynamic creation or selection of teaching materials or information to provide users with the information or training they need based on the analysis results.

[1108] "Presenting learning content" refers to displaying generated or selected learning content to a user in real time and allowing the user to use the content.

[1109] "Collecting user reaction data" refers to re-detecting and recording the user's behavior, comments, facial expressions, and eye movements while studying.

[1110] "Providing training content to workers on the production line" refers to dynamically presenting content to individual workers working on the manufacturing site to enable them to learn the necessary knowledge and skills on the spot.

[1111] "Individualized training content" refers to training content and learning materials that are customized to the specific needs and level of understanding of the user (worker).

[1112] System Overview

[1113] This invention is a system that collects voice, facial expression, and gaze data of workers in real time and analyzes the data to dynamically provide training content tailored to individual needs in order to improve the learning efficiency of workers on a production line. Specific embodiments for implementing this invention will be described below.

[1114] Main configuration

[1115] The system consists of the following main components:

[1116] 1. Terminal

[1117] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect voice, facial expression, and gaze data of workers in real time.

[1118] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1119] 2. Server

[1120] The server receives the preprocessed data sent from the terminal and analyzes it using a speech recognition engine (for example, the speech_recognition library) or facial expression analysis software (for example, the EmotionRecognition library).

[1121] The voice data is converted into text data, and the worker's level of understanding, concentration, and stress level are evaluated based on facial expression and gaze data.

[1122] Based on the analysis results, training content is generated or selected and dynamically adapted.

[1123] 3. User (operator)

[1124] The worker executes the training content provided via the terminal and makes the necessary responses.

[1125] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[1126] Program processing

[1127] The flow of program processing in this system will be explained below.

[1128] Data Collection Phase

[1129] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time.

[1130] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[1131] Analysis Phase

[1132] The terminal transmits the preprocessed data to the server.

[1133] The server converts the voice data into text data using a speech recognition engine (speech_recognition library), and uses facial expression and gaze data to determine the worker's level of understanding, concentration, and stress.

[1134] Learning content customization phase

[1135] Based on the analyzed data, the server evaluates the worker's level of understanding and learning progress, and generates or selects customized training content.

[1136] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[1137] Presentation Phase

[1138] The server transmits customized learning content to the terminal.

[1139] The terminal presents the received training content to the worker in real time, and the worker follows along with the content.

[1140] Feedback Phase

[1141] The worker performs the presented training content and provides responses that demonstrate the effectiveness of their learning.

[1142] The device again collects data such as the worker's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[1143] Specific examples

[1144] For example, if a new worker is confused about a particular procedure on a production line, the device can detect this situation through the camera and microphone. Specifically, it analyzes the worker's confused facial expression and voice, such as "What should I do about this?" Based on this, the server can provide the worker with a training video or additional explanations that are best suited to the worker's basic knowledge. This method can help new workers to proceed with their work smoothly.

[1145] Prompt Sentence Examples

[1146] If a new worker looks confused and says in audio, "What do I do about this?", suggest a training video that will help new workers review the basics.

[1147] This system provides real-time adaptive training that matches the worker's level of understanding and progress, improving the efficiency and quality of the production line.

[1148] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1149] Step 1: Data collection phase

[1150] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time. The input of this data collection is the worker's real-time activity data, and the output is audio files, image data, and gaze tracking data. This data is temporarily stored in local storage and undergoes preprocessing such as noise removal and filtering. For example, background noise is removed from the voice data and unnecessary background is cut out from the image data.

[1151] Step 2: Preprocessing phase

[1152] The device preprocesses the collected data. The input is the voice, facial expression, and gaze data collected in step 1, and the output is data that has been subjected to noise removal and filtering. Specifically, the waveform of the voice data is flattened, and histogram normalization is performed on the image data. The gaze data is also filtered and blurring is corrected.

[1153] Step 3: Data transmission phase

[1154] The device sends the preprocessed data to the server. The input is the preprocessed voice, facial expression, and gaze data, and the output is the data sent to the server. For example, the device divides the data into packets and sends them to the server.

[1155] Step 4: Analysis Phase

[1156] The server receives the transmitted data and converts the voice data into text using a speech recognition engine (e.g., speech_recognition library). The input is the preprocessed data transmitted from the device, and the output is text data and analysis data (level of understanding, concentration, stress level, etc.). Specifically, the server analyzes the facial expression data using facial expression analysis software (e.g., EmotionRecognition library) to determine the user's level of understanding. It also applies an algorithm to evaluate the level of concentration based on gaze data.

[1157] Step 5: Customizing the learning content

[1158] The server evaluates the worker's level of understanding and learning progress based on the analyzed data, and generates or selects customized training content. The input is data such as the worker's level of understanding, concentration, and stress level obtained in the analysis phase, and the output is training content dynamically created based on this data. Specifically, the server selects video materials and interactive simulations that are optimal for the worker's level of progress.

[1159] Step 6: Content Presentation Phase

[1160] The server sends customized training content to the terminal, which then presents it to the worker. The input is the training content sent from the server, and the output is the training video or simulation that the worker views. Specifically, the terminal displays the received content on a display and presents it to the worker by providing audio guidance.

[1161] Step 7: Feedback Phase

[1162] The user digests the presented training content and responds to demonstrate the learning effect. The input is the user's input (voice response, facial expression, and gaze data), and the output is the collected response data. For example, there are situations where a worker answers questions about the training content by voice. This response data is collected again by the system, and the next learning path is further optimized.

[1163] This process allows workers to receive effective training tailored to their individual needs.

[1164] 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.

[1165] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that also takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[1166] System Overview

[1167] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[1168] 1. Terminal

[1169] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[1170] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1171] 2. Server

[1172] The server receives the pre-processed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software to assess the user's level of understanding, concentration, and stress level.

[1173] The server further identifies the user's emotional state using an emotion engine and generates or selects learning content by taking the emotion information into account in the analysis results.

[1174] 3. Emotion Engine

[1175] The emotion engine analyzes the user's facial expression data and voice data to determine the user's emotional state.

[1176] The determined emotions are sent to the server and used to adapt the learning content.

[1177] 4. Users

[1178] The user digests the learning content provided via the terminal and makes the necessary responses.

[1179] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[1180] Program processing

[1181] The flow of program processing in this system will be explained below.

[1182] Data Collection Phase

[1183] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[1184] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[1185] Analysis Phase

[1186] The terminal transmits the preprocessed data to the server.

[1187] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[1188] The server uses an emotion engine to analyze the user's facial expression data and voice data to identify the user's emotional state.

[1189] Learning content customization phase

[1190] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects customized learning content.

[1191] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[1192] Presentation Phase

[1193] The server transmits customized learning content to the terminal.

[1194] The device presents the received learning content to the user in real time, supporting them in their learning.

[1195] Feedback Phase

[1196] The user digests the presented learning content and responds to demonstrate the learning effect.

[1197] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[1198] Specific examples

[1199] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1200] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[1201] The processing flow will be explained below.

[1202] Step 1:

[1203] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[1204] Step 2:

[1205] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[1206] Step 3:

[1207] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[1208] Step 4:

[1209] The terminal transmits the preprocessed data to the server.

[1210] Step 5:

[1211] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[1212] Step 6:

[1213] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[1214] Step 7:

[1215] The server analyzes the gaze data and identifies which part the user is focusing on.

[1216] Step 8:

[1217] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[1218] Step 9:

[1219] The server uses an emotion engine to analyze the facial expression data and voice data to identify the user's emotional state.

[1220] Step 10:

[1221] The server generates or selects customized learning content based on the evaluation results and emotional information.

[1222] Step 11:

[1223] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[1224] Step 12:

[1225] The server transmits customized learning content to the terminal.

[1226] Step 13:

[1227] The device presents the received learning content to the user in real time, supporting them in their learning.

[1228] Step 14:

[1229] The user digests the presented learning content and takes the necessary responses or actions.

[1230] Step 15:

[1231] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[1232] Step 16:

[1233] The terminal preprocesses the newly collected data and sends it back to the server.

[1234] Step 17:

[1235] The server analyzes the data again and generates the next learning path based on the feedback information.

[1236] Step 18:

[1237] The server sends the generated new learning path to the terminal, which then presents it to the user.

[1238] Step 19:

[1239] Users progress through new learning paths.

[1240] Example 2

[1241] 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."

[1242] Conventional learning support systems have difficulty accurately grasping a user's level of understanding and concentration, making it impossible to provide content tailored to individual learning needs. Furthermore, they often provide uniform learning content without considering the user's emotional state, limiting the improvement of learning efficiency. Furthermore, they do not dynamically adapt content to match the user's learning progress, making it difficult to motivate users or provide an effective learning experience.

[1243] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for determining the user's emotional state using the evaluation results and an emotion engine, means for generating or selecting learning content based on the evaluation results and emotion information, means for presenting the learning content to the user, and means for collecting user reaction data and re-preprocessing and analyzing the data. This makes it possible to provide individually customized learning content that takes into account the user's emotional state as well as their level of understanding and concentration.

[1244] "Voice data" refers to data in which the user's voice is recorded in digital format.

[1245] "Facial expression data" refers to data that captures a user's facial expression using a camera or the like and converts it into an analyzable format.

[1246] "Gaze data" refers to data that measures and records the user's eye movements and gaze direction.

[1247] "Preprocessing" is the process of performing processes such as noise removal and filtering on collected raw data to convert it into a format suitable for analysis.

[1248] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[1249] "Evaluation" is the process of determining the user's learning status according to specific criteria based on the information obtained through analysis.

[1250] An "emotion engine" is software or hardware that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[1251] "Learning content" is a general term for teaching materials and educational resources presented to users, and includes text, video, audio, practice questions, and the like.

[1252] "Customization" is the process of optimizing learning content according to each user's needs and learning situation.

[1253] "Presenting" refers to the act of displaying or playing selected or generated learning content to a user.

[1254] "Reaction data" is data that records the user's reactions to learning content, including voice, facial expressions, and gaze.

[1255] "Dynamic adaptation" is the process of adjusting and providing learning content in real time based on the user's learning progress, level of understanding, and emotional state.

[1256] A "voice recognition engine" is software or hardware that converts voice data into text data.

[1257] MODE FOR CARRYING OUT THE INVENTION

[1258] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by using an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[1259] System configuration

[1260] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[1261] Terminal

[1262] The device is equipped with a camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1263] server

[1264] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The server evaluates the user's level of understanding, concentration, and stress level. It also identifies the user's emotional state using an emotion engine, and generates or selects learning content by incorporating emotional information into the analysis results.

[1265] Emotion Engine

[1266] The emotion engine analyzes the user's facial expression and voice data to determine their emotional state. The determined emotions are sent to the server and used to adapt the learning content.

[1267] User

[1268] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[1269] System operation example

[1270] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1271] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[1272] Example prompts to input to the generative AI model

[1273] Below are some example prompts to input to a generative AI model:

[1274] "Based on a scenario where a student is working on a math problem but is having trouble with a particular one, describe a system that analyzes the user's level of understanding and emotional state in real time and provides customized learning content."

[1275] This prompt-based system optimizes the user's learning efficiency and provides advanced learning support that meets individual needs.

[1276] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1277] Step 1: Data collection

[1278] The device activates sensors such as a camera, microphone, and gaze tracker to collect the user's voice, facial expression, and gaze data in real time. This provides basic data for understanding the user's behavior and reactions in real time. For example, a user may have a confused expression and say, "I don't understand this problem." The input data obtained here are voice data, facial expression data, and gaze data, which are collected by the device.

[1279] Step 2: Data Preprocessing

[1280] The device temporarily stores the collected data in local storage and performs preprocessing such as noise removal and filtering. As a result of preprocessing, the data is in a clean format suitable for analysis. For example, background noise is removed from voice data and facial expression data is normalized. The input data is the raw data collected in step 1, and the output data is the preprocessed, clean data.

[1281] Step 3: Send data

[1282] The device divides the preprocessed data into packets and sends them to the server. Highly efficient data transfer is important in this step. The input data is the preprocessed data, and the output is the data sent to the server. For example, the device may use a compression algorithm to reduce the size of the data before sending it to the server.

[1283] Step 4: Convert audio data to text

[1284] The server converts the received voice data into text data using a voice recognition engine. For example, the voice saying "I don't understand this problem" is converted into text "I don't understand this problem." The input data is preprocessed voice data, and the output data is text data.

[1285] Step 5: Analyze facial expression data

[1286] The server uses facial expression analysis software to analyze the user's facial expression data to determine their level of understanding, concentration, and stress. For example, a confused expression can indicate a low level of understanding. The input data is the preprocessed facial expression data, and the output data is the analysis result.

[1287] Step 6: Analyze gaze data

[1288] The server analyzes the gaze tracker data to determine where the user is looking. For example, if the user looks at a particular problem area for a long time, it collects that information. The input data is the preprocessed gaze data, and the output data is the gaze focus information.

[1289] Step 7: Identify your emotional state

[1290] The server uses an emotion engine to identify the user's emotional state based on facial expression and voice data. For example, it determines whether the user is feeling stressed. The input data is the analyzed facial expression and voice data, and the output data is the result of the emotional state determination.

[1291] Step 8: Assess your learning progress

[1292] The server evaluates the user's learning progress based on the collected and analyzed data. For example, it identifies areas where the user is struggling with a particular topic. The input data are the analysis results and the judgment of the user's emotional state, and the output data is the evaluation of the user's learning progress.

[1293] Step 9: Content Selection and Creation

[1294] The server selects or generates optimal learning content for the user, such as basic instructional videos or practice questions. The input data is the evaluation results of the learning progress, and the output data is the selected or generated learning content.

[1295] Step 10: Present content to the user

[1296] The server sends the customized learning content to the device, which then presents it to the user, for example, by displaying and playing a video on the screen. The input data is the selected and generated learning content, and the output data is the presented learning content.

[1297] Step 11: Collect user responses

[1298] The user digests the learning content and responds with quizzes or speech to check their understanding. For example, the user answers questions in a quiz. The input data is the user's response, and the output data is the collected response data.

[1299] Step 12: Recollection and analysis

[1300] The device again collects data such as the user's voice, facial expression, and gaze, and sends it to the server for preprocessing and analysis. For example, it re-analyzes how the user is struggling with a quiz. The input data is the user's voice, facial expression, and gaze data, and the output data is the re-analysis result.

[1301] (Application example 2)

[1302] 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."

[1303] Conventional learning support systems have difficulty in comprehensively evaluating a user's level of understanding, concentration, and emotional state, resulting in problems with the provided learning content not being optimized for each individual user. Furthermore, it is difficult to identify in real time where a user is struggling and provide appropriate feedback. Furthermore, they lack the ability to adaptively generate and select learning content, which reduces the user's learning efficiency.

[1304] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1305] In this invention, the server includes means for collecting a user's voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for identifying the user's emotional state using an emotion engine and using the identified emotional state to adapt content, means for generating or selecting individually customized learning content based on the analysis results, means for presenting the generated or selected learning content to the user and prompting the user to respond to demonstrate learning effectiveness, and means for generating prompts using a generative AI model to generate appropriate learning content. This improves user learning efficiency and enables the provision of content optimized for each individual user in real time.

[1306] "Voice Data" means data in digital form that records a user's speech or other vocal communication.

[1307] "Facial expression data" is digital data acquired in real time based on the movement of the user's facial muscles.

[1308] "Gaze data" is digital data that tracks the user's eye movements and records where they are looking in real time.

[1309] "Preprocessing" refers to the process of converting collected data into a format that is easy to analyze by performing processes such as noise removal, filtering, and normalization on the data.

[1310] "Analysis" refers to data processing to assess the user's level of understanding, concentration, stress level, and emotional state based on the preprocessed data.

[1311] "Level of understanding" is an index that indicates how well a user understands the learning content.

[1312] "Concentration level" is an index that indicates how much a user is concentrating on a learning task.

[1313] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[1314] An "emotion engine" is an algorithm and software that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[1315] "Individually customized learning content" refers to learning information that is appropriately adjusted or generated based on the results of a user's analysis.

[1316] A "generative AI model" is an artificial intelligence model that generates learning content or other output based on a given prompt.

[1317] A "prompt" is a text instruction that can be input into a generative AI model to generate a specific output (e.g., a learning slide or explanatory text).

[1318] This system collects and analyzes users' voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to their individual needs, thereby improving their learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, this system provides highly adaptable learning support that takes into account the user's emotional state.

[1319] System Overview

[1320] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[1321] Terminal

[1322] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization. This results in high-quality data being sent to the server.

[1323] server

[1324] The server receives the preprocessed data sent from the device and analyzes it using a speech recognition engine (e.g., Google Cloud Speech-to-Text) and facial expression analysis software (e.g., Azure Face API) to evaluate the user's comprehension, concentration, and stress level. It also identifies the user's emotional state using an emotion engine (e.g., IBM Watson Tone Analyzer) and generates or selects learning content by incorporating the emotion information into the analysis results.

[1325] Emotion Engine

[1326] An emotion engine is an algorithm and software that analyzes a user's facial expression and voice data to determine their emotional state. The determined emotions are sent to a server and used to adapt the learning content.

[1327] User

[1328] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[1329] Program processing explanation

[1330] Data Collection Phase

[1331] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. This data is temporarily stored in local storage and undergoes pre-processing such as noise reduction and filtering.

[1332] Analysis Phase

[1333] The device sends the preprocessed data to a server, which uses a speech recognition engine to convert the voice data into text, facial expression and gaze data to determine the user's comprehension, concentration, and stress levels, and an emotion engine to identify the user's emotional state.

[1334] Learning content customization phase

[1335] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects individually customized learning content, using prompts to generate appropriate learning content using a generative AI model.

[1336] Presentation Phase

[1337] The server transmits the customized learning content to the terminal, and the terminal presents the received learning content to the user in real time to support the user in progressing with their learning.

[1338] Specific examples

[1339] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[1340] Prompt Sentence Examples

[1341] Examples of prompts to be input to a generative AI model include:

[1342] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[1343] The generated learning content dynamically adapts according to the user's level of understanding, concentration, and emotional state, significantly improving the user's learning efficiency.

[1344] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1345] Step 1:

[1346] Data Collection Phase

[1347] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. The data obtained from each sensor is temporarily stored in local storage. At this time, preprocessing such as noise removal and filtering is performed. The input includes voice, facial expression, and gaze data from the user, which is then output as high-quality data after noise removal and filtering.

[1348] Input: User's voice, facial expression, and gaze data

[1349] Data processing: noise removal, filtering, normalization

[1350] Output: Pre-processed high-quality data

[1351] Step 2:

[1352] Data transmission phase

[1353] The device sends the preprocessed data to the server via a secure communication protocol. The input is the preprocessed high-quality data, which is then sent to the server for the next analysis phase.

[1354] Input: Preprocessed high-quality data

[1355] Data processing: Data transfer, encryption

[1356] Output: Data sent to the server

[1357] Step 3:

[1358] Analysis Phase

[1359] The server receives the preprocessed data and converts it into text using a speech recognition engine. It also analyzes facial expression data using facial expression analysis software and evaluates gaze data, thereby determining the user's level of comprehension, concentration, and stress. The server receives the data as input and outputs it as comprehension, concentration, and stress levels through speech recognition and facial expression analysis.

[1360] Input: Data sent to the server

[1361] Data processing: voice recognition, facial expression analysis, gaze evaluation

[1362] Output: User comprehension, concentration, stress level

[1363] Step 4:

[1364] Sentiment Analysis Phase

[1365] The server uses an emotion engine to identify the user's emotional state from their facial and voice data, which allows it to evaluate the emotions (e.g., stress or anxiety) the user is experiencing during the learning process. The input is facial and voice data, which are analyzed by the emotion engine to output the emotional state.

[1366] Input: facial expression data, voice data

[1367] Data processing: Sentiment analysis

[1368] Output: Emotional state

[1369] Step 5:

[1370] Learning content generation phase

[1371] The server generates or selects individually customized learning content based on the analysis results. It uses a generative AI model to create prompts to generate appropriate learning content, and the content is generated based on these. Inputs include level of comprehension, concentration, stress level, and emotional state, and the server outputs customized learning content based on these.

[1372] Input: Comprehension, concentration, stress level, emotional state

[1373] Data processing: prompt generation, content generation

[1374] Output: Customized learning content

[1375] Step 6:

[1376] Learning content presentation phase

[1377] The server sends the generated learning content to the terminal, which then presents it to the user. The input is the customized learning content, which is presented to the user by the terminal.

[1378] Input: Customized learning content

[1379] Data processing: Data transfer

[1380] Output: Presenting the learning content to the user

[1381] Step 7:

[1382] Feedback gathering phase

[1383] The user digests the learning content and responds. The device again collects the user's voice, facial expression, and gaze data for preprocessing and analysis. This allows the device to continuously evaluate the user's learning effectiveness and further optimize the next learning path. The input is the user's response data, which is then preprocessed and analyzed to output new levels of comprehension, concentration, stress level, and emotional state.

[1384] Input: User response data

[1385] Data processing: noise removal, filtering, voice recognition, facial expression analysis, emotion analysis

[1386] Output: New understanding, focus, stress level, emotional state

[1387] Specific examples

[1388] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[1389] Prompt Sentence Examples

[1390] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[1391] 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.

[1392] 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.

[1393] 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.

[1394] [Fourth embodiment]

[1395] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1396] 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.

[1397] 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).

[1398] 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.

[1399] 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.

[1400] 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).

[1401] 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.

[1402] 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.

[1403] 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.

[1404] 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.

[1405] 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.

[1406] 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.

[1407] 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."

[1408] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. The following describes in detail an embodiment of the present invention.

[1409] System Overview

[1410] The system consists of three main components: terminals, servers, and users.

[1411] 1. Terminal

[1412] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[1413] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1414] 2. Server

[1415] The server receives the preprocessed data sent from the terminal and analyzes it using a voice recognition engine and facial expression analysis software.

[1416] The voice data is converted into text data, and the user's level of understanding, concentration, and stress level are evaluated from facial expression and gaze data.

[1417] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[1418] 3. Users

[1419] The user digests the learning content provided via the terminal and makes the necessary responses.

[1420] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[1421] Program processing

[1422] The flow of program processing in this system will be explained below.

[1423] Data Collection Phase

[1424] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[1425] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[1426] Analysis Phase

[1427] The terminal transmits the preprocessed data to the server.

[1428] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[1429] Learning content customization phase

[1430] Based on the analyzed data, the server evaluates the user's level of understanding and learning progress, and generates or selects customized learning content.

[1431] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[1432] Presentation Phase

[1433] The server transmits customized learning content to the terminal.

[1434] The terminal presents the received learning content to the user in real time, and the user proceeds with learning in accordance with the content.

[1435] Feedback Phase

[1436] The user digests the presented learning content and responds to demonstrate the learning effect.

[1437] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[1438] Specific examples

[1439] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1440] In this way, the system provides real-time adaptive learning based on the user's level of understanding and concentration.

[1441] The processing flow will be explained below.

[1442] Step 1:

[1443] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[1444] Step 2:

[1445] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[1446] Step 3:

[1447] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[1448] Step 4:

[1449] The terminal transmits the preprocessed data to the server.

[1450] Step 5:

[1451] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[1452] Step 6:

[1453] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[1454] Step 7:

[1455] The server analyzes the gaze data and identifies which part the user is focusing on.

[1456] Step 8:

[1457] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[1458] Step 9:

[1459] The server generates or selects customized learning content based on the evaluation results.

[1460] Step 10:

[1461] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[1462] Step 11:

[1463] The server transmits customized learning content to the terminal.

[1464] Step 12:

[1465] The device presents the received learning content to the user in real time, supporting them in their learning.

[1466] Step 13:

[1467] The user digests the presented learning content and takes the necessary responses or actions.

[1468] Step 14:

[1469] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[1470] Step 15:

[1471] The terminal preprocesses the newly collected data and sends it back to the server.

[1472] Step 16:

[1473] The server analyzes the data again and generates the next learning path based on the feedback information.

[1474] Step 17:

[1475] The server sends the generated new learning path to the terminal, which then presents it to the user.

[1476] Step 18:

[1477] Users progress through new learning paths.

[1478] Example 1

[1479] 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."

[1480] Conventional learning systems have had difficulty adjusting content based on the user's real-time status and reactions in order to maximize the user's learning efficiency. This has resulted in the inability to respond immediately when the user stumbles on a particular problem or point, resulting in a decrease in learning efficiency. Therefore, there is a need for a system that can dynamically customize learning content based on the user's real-time data and quickly respond to each user's learning needs.

[1481] 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.

[1482] In this invention, the server includes means for collecting voice, facial expression, and gaze data of a user in real time, means for preprocessing the voice, facial expression, and gaze data, and means for analyzing the preprocessed data to evaluate the user's level of understanding, concentration, and stress level, thereby enabling dynamic customization of learning content based on the user's real-time data to meet individual learning needs.

[1483] "Voice data" refers to data consisting of user utterances, sounds, and linguistic information.

[1484] "Facial expression data" is data that includes information about the user's facial expression.

[1485] "Gaze data" is data that includes information about the user's eye movements and viewpoint.

[1486] "Preprocessing" refers to processing the collected data, such as noise removal and filtering, to make it easier to analyze.

[1487] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[1488] "Learning content" refers to content such as educational materials, questions, and explanatory videos provided to support users' learning.

[1489] "Customized learning content" is learning content that is adapted based on a user's individual needs and learning progress.

[1490] "Batch transmission" is a method of transmitting multiple pieces of data together at regular time intervals.

[1491] "Facial expression analysis software" is software that analyzes a user's facial expressions to evaluate their emotions and state.

[1492] A "voice recognition engine" is software or hardware for converting voice data into text data.

[1493] "Noise reduction" is a process of removing unnecessary noise from audio data and the like.

[1494] "Level of concentration" is a measure of how much the user is concentrating on their studies.

[1495] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[1496] The present invention is a system that collects and analyzes voice, facial expression, and gaze data of a user in real time to improve the user's learning efficiency, and dynamically provides learning content tailored to individual needs. Specific embodiments for implementing the present invention will be described in detail below.

[1497] System Overview

[1498] The system consists of three main components: terminals, servers, and users.

[1499] 1. Terminal

[1500] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[1501] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1502] 2. Server

[1503] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The voice data is converted into text data, and the user's level of comprehension, concentration, and stress level are evaluated based on facial expression and gaze data.

[1504] Based on the analysis results, learning content is generated or selected and dynamically adapted.

[1505] 3. Users

[1506] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[1507] Specific examples of programs

[1508] For example, let's say Student A is struggling with a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies the part where Student A is struggling. The server then determines that the identified part requires a reconfirmation of basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1509] Hardware and software used

[1510] Speech recognition engine: For example, use the Google Speech-to-Text API to convert voice data into text data.

[1511] Facial expression analysis software: For example, using Microsoft Azure Face API to analyze the user's facial expressions.

[1512] Eye-tracking software: In combination with a camera, it collects and analyzes user gaze data.

[1513] Examples of prompt statements

[1514] For example, enter the following prompt sentence into the generative AI model:

[1515] "Design a system that can detect in real time when you are struggling with a problem and provide you with the appropriate learning content. For example, it would be even better if the system could detect when a student is struggling with a problem and provide supplementary materials."

[1516] summary

[1517] The present invention is a system that dynamically customizes learning content based on users' real-time data to meet their individual learning needs, thereby maximizing the user's learning efficiency and supporting rapid problem solving.

[1518] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1519] Step 1:

[1520] Starting the sensor

[1521] The device activates the camera, microphone, and other sensors to begin collecting data from the user. For example, the device automatically activates each sensor the moment the user starts learning content. The input is the user's action to start the learning content, and the output is each activated sensor.

[1522] Step 2:

[1523] Data collection

[1524] The device collects the user's voice, facial expression, and gaze data in real time. For example, a camera captures the user's facial expressions and a microphone records what the user says. The input is the real-time user data captured by the sensor, and the output is the collected raw data.

[1525] Step 3:

[1526] Temporary storage and preprocessing of data

[1527] The device temporarily stores the collected data in local storage and performs preprocessing such as noise reduction and filtering. For example, a noise reduction algorithm is applied to clear voice data collected in a noisy environment. The input is the collected raw data, and the output is the preprocessed data.

[1528] Step 4:

[1529] Sending preprocessed data

[1530] The terminal sends the pre-processed data to the server, for example, by batch transmission of multiple data at regular intervals. The input is the pre-processed data, and the output is the data sent to the server.

[1531] Step 5:

[1532] Analysis of audio data

[1533] The server converts the voice data into text data using a speech recognition engine (e.g., Google Speech-to-Text API). For example, a user's statement "I don't understand this problem" is converted into text data. The input is the transmitted voice data, and the output is the converted text data.

[1534] Step 6:

[1535] Analysis of facial expression and gaze data

[1536] The server uses facial expression analysis software (e.g., Microsoft Azure Face API) and eye-tracking software to determine the user's level of comprehension, concentration, and stress. For example, it can detect when the user frowns and determine that the user's stress level is high. The input is the transmitted facial expression and eye-gaze data, and the output is the analyzed user's level of comprehension, concentration, and stress.

[1537] Step 7:

[1538] Assessment of understanding and progress

[1539] The server evaluates the user's level of understanding and learning progress based on the analyzed data. For example, it determines the level of understanding based on data such as which questions the user spent the most time on. The input is the analyzed user data, and the output is the evaluation results.

[1540] Step 8:

[1541] Content generation or selection

[1542] The server generates or selects customized learning content and adjusts the content. For example, if a user has difficulty with "factorization," it selects videos or additional practice problems to refresh the knowledge from the basics. The input is the assessment results, and the output is customized learning content.

[1543] Step 9:

[1544] Sending customized content

[1545] The server sends customized learning content to the device. For example, it sends a specific video link or PDF file to the device. The input is the customized learning content, and the output is the data sent to the device.

[1546] Step 10:

[1547] View content

[1548] The device displays the received learning content to the user, and the user proceeds with the learning process. For example, a video for problem solving is displayed on the screen. The input is the transmitted learning content data, and the output is the content displayed to the user.

[1549] Step 11:

[1550] Collecting user learning status

[1551] The user digests the presented learning content and responds. For example, after solving a problem, the user may move on to the next problem or respond by saying "I couldn't solve it." The input is the user's learning behavior, and the output is learning behavior data.

[1552] Step 12:

[1553] Reaction data collection

[1554] The device again collects the user's voice, facial expression, and gaze data, and performs preprocessing and analysis again. For example, if the user makes a new confused expression during learning, that data will also be collected. The input is the newly collected data, and the output is the data that will be preprocessed again.

[1555] (Application example 1)

[1556] 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."

[1557] On manufacturing lines, it is important for workers to quickly acquire skills and improve production efficiency. However, it is difficult to provide training content that is tailored to each worker's level of understanding and progress, and general manuals and videos alone are often insufficient. It is also difficult to grasp in real time which parts of the training workers are struggling with. This reduces the effectiveness of training and leads to issues such as lower production efficiency.

[1558] 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.

[1559] In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for generating or selecting learning content based on the evaluation results, means for presenting the learning content to the user, means for collecting user reaction data and preprocessing and analyzing it again, and means for providing individual training content to production line workers by collecting and analyzing voice, facial expression, and gaze data in real time, thereby enabling the provision of optimal training content tailored to the worker's level of understanding and progress.

[1560] "Collecting user voice, facial expression, and gaze data" refers to sensing and recording the user's voice, facial expression changes, and gaze movements in real time.

[1561] "Preprocessing" refers to a series of operations that remove noise from collected raw data and prepare the data in a usable form.

[1562] "Data analysis" refers to extracting information such as the user's level of understanding, concentration, and stress level from the pre-processed data using specific algorithms or models.

[1563] "Generation or selection of learning content" refers to the dynamic creation or selection of teaching materials or information to provide users with the information or training they need based on the analysis results.

[1564] "Presenting learning content" refers to displaying generated or selected learning content to a user in real time and allowing the user to use the content.

[1565] "Collecting user reaction data" refers to re-detecting and recording the user's behavior, comments, facial expressions, and eye movements while studying.

[1566] "Providing training content to workers on the production line" refers to dynamically presenting content to individual workers working on the manufacturing site to enable them to learn the necessary knowledge and skills on the spot.

[1567] "Individualized training content" refers to training content and learning materials that are customized to the specific needs and level of understanding of the user (worker).

[1568] System Overview

[1569] This invention is a system that collects voice, facial expression, and gaze data of workers in real time and analyzes the data to dynamically provide training content tailored to individual needs in order to improve the learning efficiency of workers on a production line. Specific embodiments for implementing this invention will be described below.

[1570] Main configuration

[1571] The system consists of the following main components:

[1572] 1. Terminal

[1573] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect voice, facial expression, and gaze data of workers in real time.

[1574] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1575] 2. Server

[1576] The server receives the preprocessed data sent from the terminal and analyzes it using a speech recognition engine (for example, the speech_recognition library) or facial expression analysis software (for example, the EmotionRecognition library).

[1577] The voice data is converted into text data, and the worker's level of understanding, concentration, and stress level are evaluated based on facial expression and gaze data.

[1578] Based on the analysis results, training content is generated or selected and dynamically adapted.

[1579] 3. User (operator)

[1580] The worker executes the training content provided via the terminal and makes the necessary responses.

[1581] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[1582] Program processing

[1583] The flow of program processing in this system will be explained below.

[1584] Data Collection Phase

[1585] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time.

[1586] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[1587] Analysis Phase

[1588] The terminal transmits the preprocessed data to the server.

[1589] The server converts the voice data into text data using a speech recognition engine (speech_recognition library), and uses facial expression and gaze data to determine the worker's level of understanding, concentration, and stress.

[1590] Learning content customization phase

[1591] Based on the analyzed data, the server evaluates the worker's level of understanding and learning progress, and generates or selects customized training content.

[1592] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[1593] Presentation Phase

[1594] The server transmits customized learning content to the terminal.

[1595] The terminal presents the received training content to the worker in real time, and the worker follows along with the content.

[1596] Feedback Phase

[1597] The worker performs the presented training content and provides responses that demonstrate the effectiveness of their learning.

[1598] The device again collects data such as the worker's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[1599] Specific examples

[1600] For example, if a new worker is confused about a particular procedure on a production line, the device can detect this situation through the camera and microphone. Specifically, it analyzes the worker's confused facial expression and voice, such as "What should I do about this?" Based on this, the server can provide the worker with a training video or additional explanations that are best suited to the worker's basic knowledge. This method can help new workers to proceed with their work smoothly.

[1601] Prompt Sentence Examples

[1602] If a new worker looks confused and says in audio, "What do I do about this?", suggest a training video that will help new workers review the basics.

[1603] This system provides real-time adaptive training that matches the worker's level of understanding and progress, improving the efficiency and quality of the production line.

[1604] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1605] Step 1: Data collection phase

[1606] The device activates the camera, microphone, and other sensors to collect the worker's voice, facial expression, and gaze data in real time. The input of this data collection is the worker's real-time activity data, and the output is audio files, image data, and gaze tracking data. This data is temporarily stored in local storage and undergoes preprocessing such as noise removal and filtering. For example, background noise is removed from the voice data and unnecessary background is cut out from the image data.

[1607] Step 2: Preprocessing phase

[1608] The device preprocesses the collected data. The input is the voice, facial expression, and gaze data collected in step 1, and the output is data that has been subjected to noise removal and filtering. Specifically, the waveform of the voice data is flattened, and histogram normalization is performed on the image data. The gaze data is also filtered and blurring is corrected.

[1609] Step 3: Data transmission phase

[1610] The device sends the preprocessed data to the server. The input is the preprocessed voice, facial expression, and gaze data, and the output is the data sent to the server. For example, the device divides the data into packets and sends them to the server.

[1611] Step 4: Analysis Phase

[1612] The server receives the transmitted data and converts the voice data into text using a speech recognition engine (e.g., speech_recognition library). The input is the preprocessed data transmitted from the device, and the output is text data and analysis data (level of understanding, concentration, stress level, etc.). Specifically, the server analyzes the facial expression data using facial expression analysis software (e.g., EmotionRecognition library) to determine the user's level of understanding. It also applies an algorithm to evaluate the level of concentration based on gaze data.

[1613] Step 5: Customizing the learning content

[1614] The server evaluates the worker's level of understanding and learning progress based on the analyzed data, and generates or selects customized training content. The input is data such as the worker's level of understanding, concentration, and stress level obtained in the analysis phase, and the output is training content dynamically created based on this data. Specifically, the server selects video materials and interactive simulations that are optimal for the worker's level of progress.

[1615] Step 6: Content Presentation Phase

[1616] The server sends customized training content to the terminal, which then presents it to the worker. The input is the training content sent from the server, and the output is the training video or simulation that the worker views. Specifically, the terminal displays the received content on a display and presents it to the worker by providing audio guidance.

[1617] Step 7: Feedback Phase

[1618] The user digests the presented training content and responds to demonstrate the learning effect. The input is the user's input (voice response, facial expression, and gaze data), and the output is the collected response data. For example, there are situations where a worker answers questions about the training content by voice. This response data is collected again by the system, and the next learning path is further optimized.

[1619] This process allows workers to receive effective training tailored to their individual needs.

[1620] 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.

[1621] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that also takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[1622] System Overview

[1623] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[1624] 1. Terminal

[1625] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time.

[1626] The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1627] 2. Server

[1628] The server receives the pre-processed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software to assess the user's level of understanding, concentration, and stress level.

[1629] The server further identifies the user's emotional state using an emotion engine and generates or selects learning content by taking the emotion information into account in the analysis results.

[1630] 3. Emotion Engine

[1631] The emotion engine analyzes the user's facial expression data and voice data to determine the user's emotional state.

[1632] The determined emotions are sent to the server and used to adapt the learning content.

[1633] 4. Users

[1634] The user digests the learning content provided via the terminal and makes the necessary responses.

[1635] Reaction data, such as speech and facial expressions, is collected again and continuously evaluated and adjusted by the system.

[1636] Program processing

[1637] The flow of program processing in this system will be explained below.

[1638] Data Collection Phase

[1639] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[1640] The collected data is temporarily stored in local storage and undergoes pre-processing such as noise removal and filtering.

[1641] Analysis Phase

[1642] The terminal transmits the preprocessed data to the server.

[1643] The server converts the voice data into text data using a voice recognition engine, and uses facial expression and gaze data to determine the user's level of understanding, concentration, and stress.

[1644] The server uses an emotion engine to analyze the user's facial expression data and voice data to identify the user's emotional state.

[1645] Learning content customization phase

[1646] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects customized learning content.

[1647] Content will be provided with adjustments to the difficulty level and additional explanations as needed.

[1648] Presentation Phase

[1649] The server transmits customized learning content to the terminal.

[1650] The device presents the received learning content to the user in real time, supporting them in their learning.

[1651] Feedback Phase

[1652] The user digests the presented learning content and responds to demonstrate the learning effect.

[1653] The device again collects data such as the user's voice, facial expressions, and gaze, and then preprocesses and analyzes it again.

[1654] Specific examples

[1655] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1656] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[1657] The processing flow will be explained below.

[1658] Step 1:

[1659] The device activates the camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time.

[1660] Step 2:

[1661] The audio data collected by the device is temporarily stored in local storage and pre-processed to remove noise.

[1662] Step 3:

[1663] The facial expression and gaze data collected by the device is filtered and normalized as a pre-processing step.

[1664] Step 4:

[1665] The terminal transmits the preprocessed data to the server.

[1666] Step 5:

[1667] In order for the server to analyze the received data, the voice data is sent to a voice recognition engine and converted into text data.

[1668] Step 6:

[1669] The server uses a facial expression analysis algorithm to analyze the facial expression data and determine the user's concentration level and stress level.

[1670] Step 7:

[1671] The server analyzes the gaze data and identifies which part the user is focusing on.

[1672] Step 8:

[1673] The server evaluates the user's level of understanding and learning progress based on the analysis results.

[1674] Step 9:

[1675] The server uses an emotion engine to analyze the facial expression data and voice data to identify the user's emotional state.

[1676] Step 10:

[1677] The server generates or selects customized learning content based on the evaluation results and emotional information.

[1678] Step 11:

[1679] The server adjusts the difficulty level of the content as needed and includes additional instructions.

[1680] Step 12:

[1681] The server transmits customized learning content to the terminal.

[1682] Step 13:

[1683] The device presents the received learning content to the user in real time, supporting them in their learning.

[1684] Step 14:

[1685] The user digests the presented learning content and takes the necessary responses or actions.

[1686] Step 15:

[1687] The device again collects reaction data such as the user's voice, facial expressions, and gaze.

[1688] Step 16:

[1689] The terminal preprocesses the newly collected data and sends it back to the server.

[1690] Step 17:

[1691] The server analyzes the data again and generates the next learning path based on the feedback information.

[1692] Step 18:

[1693] The server sends the generated new learning path to the terminal, which then presents it to the user.

[1694] Step 19:

[1695] Users progress through new learning paths.

[1696] Example 2

[1697] 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."

[1698] Conventional learning support systems have difficulty accurately grasping a user's level of understanding and concentration, making it impossible to provide content tailored to individual learning needs. Furthermore, they often provide uniform learning content without considering the user's emotional state, limiting the improvement of learning efficiency. Furthermore, they do not dynamically adapt content to match the user's learning progress, making it difficult to motivate users or provide an effective learning experience.

[1699] The identification processing by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for determining the user's emotional state using the evaluation results and an emotion engine, means for generating or selecting learning content based on the evaluation results and emotion information, means for presenting the learning content to the user, and means for collecting user reaction data and re-preprocessing and analyzing the data. This makes it possible to provide individually customized learning content that takes into account the user's emotional state as well as their level of understanding and concentration.

[1700] "Voice data" refers to data in which the user's voice is recorded in digital format.

[1701] "Facial expression data" refers to data that captures a user's facial expression using a camera or the like and converts it into an analyzable format.

[1702] "Gaze data" refers to data that measures and records the user's eye movements and gaze direction.

[1703] "Preprocessing" is the process of performing processes such as noise removal and filtering on collected raw data to convert it into a format suitable for analysis.

[1704] "Analysis" is the process of evaluating the user's level of understanding, concentration, stress level, etc. based on preprocessed data.

[1705] "Evaluation" is the process of determining the user's learning status according to specific criteria based on the information obtained through analysis.

[1706] An "emotion engine" is software or hardware that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[1707] "Learning content" is a general term for teaching materials and educational resources presented to users, and includes text, video, audio, practice questions, and the like.

[1708] "Customization" is the process of optimizing learning content according to each user's needs and learning situation.

[1709] "Presenting" refers to the act of displaying or playing selected or generated learning content to a user.

[1710] "Reaction data" is data that records the user's reactions to learning content, including voice, facial expressions, and gaze.

[1711] "Dynamic adaptation" is the process of adjusting and providing learning content in real time based on the user's learning progress, level of understanding, and emotional state.

[1712] A "voice recognition engine" is software or hardware that converts voice data into text data.

[1713] MODE FOR CARRYING OUT THE INVENTION

[1714] The present invention is a system that collects and analyzes a user's voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to individual needs in order to improve the user's learning efficiency. Furthermore, by using an emotion engine that recognizes the user's emotions, the system provides highly adaptable learning support that takes into account the user's emotional state. The following describes in detail the embodiments of the present invention.

[1715] System configuration

[1716] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[1717] Terminal

[1718] The device is equipped with a camera, microphone, and other sensors to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization.

[1719] server

[1720] The server receives the preprocessed data sent from the device and analyzes it using a voice recognition engine and facial expression analysis software. The server evaluates the user's level of understanding, concentration, and stress level. It also identifies the user's emotional state using an emotion engine, and generates or selects learning content by incorporating emotional information into the analysis results.

[1721] Emotion Engine

[1722] The emotion engine analyzes the user's facial expression and voice data to determine their emotional state. The determined emotions are sent to the server and used to adapt the learning content.

[1723] User

[1724] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[1725] System operation example

[1726] For example, suppose Student A gets stuck on a particular part of a math problem. In this case, the device detects a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is struggling with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that Student A needs to review his or her basic knowledge, selects relevant explanatory videos and practice problems, and sends them to the device. The device displays these to Student A, who can use them to deepen his or her understanding. As Student A digests the new learning content, his or her reactions are again fed back into the system, further optimizing the next learning path.

[1727] In this way, the system provides real-time adaptive learning that takes into account the user's level of understanding, concentration, and even emotional state.

[1728] Example prompts to input to the generative AI model

[1729] Below are some example prompts to input to a generative AI model:

[1730] "Based on a scenario where a student is working on a math problem but is having trouble with a particular one, describe a system that analyzes the user's level of understanding and emotional state in real time and provides customized learning content."

[1731] This prompt-based system optimizes the user's learning efficiency and provides advanced learning support that meets individual needs.

[1732] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1733] Step 1: Data collection

[1734] The device activates sensors such as a camera, microphone, and gaze tracker to collect the user's voice, facial expression, and gaze data in real time. This provides basic data for understanding the user's behavior and reactions in real time. For example, a user may have a confused expression and say, "I don't understand this problem." The input data obtained here are voice data, facial expression data, and gaze data, which are collected by the device.

[1735] Step 2: Data Preprocessing

[1736] The device temporarily stores the collected data in local storage and performs preprocessing such as noise removal and filtering. As a result of preprocessing, the data is in a clean format suitable for analysis. For example, background noise is removed from voice data and facial expression data is normalized. The input data is the raw data collected in step 1, and the output data is the preprocessed, clean data.

[1737] Step 3: Send data

[1738] The device divides the preprocessed data into packets and sends them to the server. Highly efficient data transfer is important in this step. The input data is the preprocessed data, and the output is the data sent to the server. For example, the device may use a compression algorithm to reduce the size of the data before sending it to the server.

[1739] Step 4: Convert audio data to text

[1740] The server converts the received voice data into text data using a voice recognition engine. For example, the voice saying "I don't understand this problem" is converted into text "I don't understand this problem." The input data is preprocessed voice data, and the output data is text data.

[1741] Step 5: Analyze facial expression data

[1742] The server uses facial expression analysis software to analyze the user's facial expression data to determine their level of understanding, concentration, and stress. For example, a confused expression can indicate a low level of understanding. The input data is the preprocessed facial expression data, and the output data is the analysis result.

[1743] Step 6: Analyze gaze data

[1744] The server analyzes the gaze tracker data to determine where the user is looking. For example, if the user looks at a particular problem area for a long time, it collects that information. The input data is the preprocessed gaze data, and the output data is the gaze focus information.

[1745] Step 7: Identify your emotional state

[1746] The server uses an emotion engine to identify the user's emotional state based on facial expression and voice data. For example, it determines whether the user is feeling stressed. The input data is the analyzed facial expression and voice data, and the output data is the result of the emotional state determination.

[1747] Step 8: Assess your learning progress

[1748] The server evaluates the user's learning progress based on the collected and analyzed data. For example, it identifies areas where the user is struggling with a particular topic. The input data are the analysis results and the judgment of the user's emotional state, and the output data is the evaluation of the user's learning progress.

[1749] Step 9: Content Selection and Creation

[1750] The server selects or generates optimal learning content for the user, such as basic instructional videos or practice questions. The input data is the evaluation results of the learning progress, and the output data is the selected or generated learning content.

[1751] Step 10: Present content to the user

[1752] The server sends the customized learning content to the device, which then presents it to the user, for example, by displaying and playing a video on the screen. The input data is the selected and generated learning content, and the output data is the presented learning content.

[1753] Step 11: Collect user responses

[1754] The user digests the learning content and responds with quizzes or speech to check their understanding. For example, the user answers questions in a quiz. The input data is the user's response, and the output data is the collected response data.

[1755] Step 12: Recollection and analysis

[1756] The device again collects data such as the user's voice, facial expression, and gaze, and sends it to the server for preprocessing and analysis. For example, it re-analyzes how the user is struggling with a quiz. The input data is the user's voice, facial expression, and gaze data, and the output data is the re-analysis result.

[1757] (Application example 2)

[1758] 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."

[1759] Conventional learning support systems have difficulty in comprehensively evaluating a user's level of understanding, concentration, and emotional state, resulting in problems with the provided learning content not being optimized for each individual user. Furthermore, it is difficult to identify in real time where a user is struggling and provide appropriate feedback. Furthermore, they lack the ability to adaptively generate and select learning content, which reduces the user's learning efficiency.

[1760] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1761] In this invention, the server includes means for collecting a user's voice, facial expression, and gaze data in real time, means for preprocessing the voice, facial expression, and gaze data, means for analyzing the preprocessed data and evaluating the user's level of understanding, concentration, and stress level, means for identifying the user's emotional state using an emotion engine and using the identified emotional state to adapt content, means for generating or selecting individually customized learning content based on the analysis results, means for presenting the generated or selected learning content to the user and prompting the user to respond to demonstrate learning effectiveness, and means for generating prompts using a generative AI model to generate appropriate learning content. This improves user learning efficiency and enables the provision of content optimized for each individual user in real time.

[1762] "Voice Data" means data in digital form that records a user's speech or other vocal communication.

[1763] "Facial expression data" is digital data acquired in real time based on the movement of the user's facial muscles.

[1764] "Gaze data" is digital data that tracks the user's eye movements and records where they are looking in real time.

[1765] "Preprocessing" refers to the process of converting collected data into a format that is easy to analyze by performing processes such as noise removal, filtering, and normalization on the data.

[1766] "Analysis" refers to data processing to assess the user's level of understanding, concentration, stress level, and emotional state based on the preprocessed data.

[1767] "Level of understanding" is an index that indicates how well a user understands the learning content.

[1768] "Concentration level" is an index that indicates how much a user is concentrating on a learning task.

[1769] The "stress level" is an index that indicates the degree of stress that the user is feeling.

[1770] An "emotion engine" is an algorithm and software that analyzes a user's facial expression data and voice data to determine the user's emotional state.

[1771] "Individually customized learning content" refers to learning information that is appropriately adjusted or generated based on the results of a user's analysis.

[1772] A "generative AI model" is an artificial intelligence model that generates learning content or other output based on a given prompt.

[1773] A "prompt" is a text instruction that can be input into a generative AI model to generate a specific output (e.g., a learning slide or explanatory text).

[1774] This system collects and analyzes users' voice, facial expression, and gaze data in real time to dynamically provide learning content tailored to their individual needs, thereby improving their learning efficiency. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, this system provides highly adaptable learning support that takes into account the user's emotional state.

[1775] System Overview

[1776] The system consists of four main components: a terminal, a server, an emotion engine, and a user.

[1777] Terminal

[1778] The device is equipped with a camera, microphone, and other sensors, and has the ability to collect the user's voice, facial expression, and gaze data in real time. The collected data is temporarily stored in local storage and undergoes preprocessing such as noise removal and normalization. This results in high-quality data being sent to the server.

[1779] server

[1780] The server receives the preprocessed data sent from the device and analyzes it using a speech recognition engine (e.g., Google Cloud Speech-to-Text) and facial expression analysis software (e.g., Azure Face API) to evaluate the user's comprehension, concentration, and stress level. It also identifies the user's emotional state using an emotion engine (e.g., IBM Watson Tone Analyzer) and generates or selects learning content by incorporating the emotion information into the analysis results.

[1781] Emotion Engine

[1782] An emotion engine is an algorithm and software that analyzes a user's facial expression and voice data to determine their emotional state. The determined emotions are sent to a server and used to adapt the learning content.

[1783] User

[1784] The user digests the learning content provided via the device and responds as needed. Response data, such as speech and facial expressions, is collected and continuously evaluated and adjusted by the system.

[1785] Program processing explanation

[1786] Data Collection Phase

[1787] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. This data is temporarily stored in local storage and undergoes pre-processing such as noise reduction and filtering.

[1788] Analysis Phase

[1789] The device sends the preprocessed data to a server, which uses a speech recognition engine to convert the voice data into text, facial expression and gaze data to determine the user's comprehension, concentration, and stress levels, and an emotion engine to identify the user's emotional state.

[1790] Learning content customization phase

[1791] The server evaluates the user's level of understanding and learning progress based on the analyzed data and emotional information, and generates or selects individually customized learning content, using prompts to generate appropriate learning content using a generative AI model.

[1792] Presentation Phase

[1793] The server transmits the customized learning content to the terminal, and the terminal presents the received learning content to the user in real time to support the user in progressing with their learning.

[1794] Specific examples

[1795] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[1796] Prompt Sentence Examples

[1797] Examples of prompts to be input to a generative AI model include:

[1798] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[1799] The generated learning content dynamically adapts according to the user's level of understanding, concentration, and emotional state, significantly improving the user's learning efficiency.

[1800] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1801] Step 1:

[1802] Data Collection Phase

[1803] The device activates the camera and microphone to collect the user's voice, facial expression, and gaze data in real time. The data obtained from each sensor is temporarily stored in local storage. At this time, preprocessing such as noise removal and filtering is performed. The input includes voice, facial expression, and gaze data from the user, which is then output as high-quality data after noise removal and filtering.

[1804] Input: User's voice, facial expression, and gaze data

[1805] Data processing: noise removal, filtering, normalization

[1806] Output: Pre-processed high-quality data

[1807] Step 2:

[1808] Data transmission phase

[1809] The device sends the preprocessed data to the server via a secure communication protocol. The input is the preprocessed high-quality data, which is then sent to the server for the next analysis phase.

[1810] Input: Preprocessed high-quality data

[1811] Data processing: Data transfer, encryption

[1812] Output: Data sent to the server

[1813] Step 3:

[1814] Analysis Phase

[1815] The server receives the preprocessed data and converts it into text using a speech recognition engine. It also analyzes facial expression data using facial expression analysis software and evaluates gaze data, thereby determining the user's level of comprehension, concentration, and stress. The server receives the data as input and outputs it as comprehension, concentration, and stress levels through speech recognition and facial expression analysis.

[1816] Input: Data sent to the server

[1817] Data processing: voice recognition, facial expression analysis, gaze evaluation

[1818] Output: User comprehension, concentration, stress level

[1819] Step 4:

[1820] Sentiment Analysis Phase

[1821] The server uses an emotion engine to identify the user's emotional state from their facial and voice data, which allows it to evaluate the emotions (e.g., stress or anxiety) the user is experiencing during the learning process. The input is facial and voice data, which are analyzed by the emotion engine to output the emotional state.

[1822] Input: facial expression data, voice data

[1823] Data processing: Sentiment analysis

[1824] Output: Emotional state

[1825] Step 5:

[1826] Learning content generation phase

[1827] The server generates or selects individually customized learning content based on the analysis results. It uses a generative AI model to create prompts to generate appropriate learning content, and the content is generated based on these. Inputs include level of comprehension, concentration, stress level, and emotional state, and the server outputs customized learning content based on these.

[1828] Input: Comprehension, concentration, stress level, emotional state

[1829] Data processing: prompt generation, content generation

[1830] Output: Customized learning content

[1831] Step 6:

[1832] Learning content presentation phase

[1833] The server sends the generated learning content to the terminal, which then presents it to the user. The input is the customized learning content, which is presented to the user by the terminal.

[1834] Input: Customized learning content

[1835] Data processing: Data transfer

[1836] Output: Presenting the learning content to the user

[1837] Step 7:

[1838] Feedback gathering phase

[1839] The user digests the learning content and responds. The device again collects the user's voice, facial expression, and gaze data for preprocessing and analysis. This allows the device to continuously evaluate the user's learning effectiveness and further optimize the next learning path. The input is the user's response data, which is then preprocessed and analyzed to output new levels of comprehension, concentration, stress level, and emotional state.

[1840] Input: User response data

[1841] Data processing: noise removal, filtering, voice recognition, facial expression analysis, emotion analysis

[1842] Output: New understanding, focus, stress level, emotional state

[1843] Specific examples

[1844] For example, if Student A has trouble with a particular part of a math problem, the device will detect a change in Student A's facial expression and a voice message saying, "I don't understand this problem." The server analyzes this and identifies which part Student A is having trouble with. In addition, the emotion engine recognizes emotions such as stress and anxiety from Student A's facial expression. Next, based on the identified part and emotional information, the server determines that basic knowledge needs to be reviewed, and selects and sends relevant explanatory videos and practice problems to the device. The device then displays these to Student A, who can use them to deepen his or her understanding. By repeating this process, Student A's learning is optimized.

[1845] Prompt Sentence Examples

[1846] "Generate a slide that explains in detail the math concept 'quadratic equations' that Student A does not understand."

[1847] 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.

[1848] 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.

[1849] 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.

[1850] 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.

[1851] 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.

[1852] 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.

[1853] 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).

[1854] 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.

[1855] 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."

[1856] 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.

[1857] 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).

[1858] 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.

[1859] 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.

[1860] 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.

[1861] 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.

[1862] 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.

[1863] 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.

[1864] 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.

[1865] 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.

[1866] 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.

[1867] 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.

[1868] The following is further disclosed regarding the above embodiment.

[1869] (Claim 1)

[1870] means for collecting user voice, facial expression, and gaze data in real time;

[1871] means for pre-processing the speech, facial expression, and gaze data;

[1872] means for analyzing the pre-processed data to assess the user's comprehension, concentration, and stress level;

[1873] a means for generating or selecting learning content based on the assessment results;

[1874] means for presenting the learning content to a user;

[1875] The system includes means for collecting user response data and again preprocessing and analyzing it.

[1876] (Claim 2)

[1877] 10. The system of claim 1, further comprising means for dynamically adapting the learning content.

[1878] (Claim 3)

[1879] 2. The system of claim 1, wherein the analyzing means includes a speech recognition engine that converts speech data into text data.

[1880] "Example 1"

[1881] (Claim 1)

[1882] means for collecting user voice, facial expression, and gaze data in real time;

[1883] means for pre-processing the speech, facial expression, and gaze data;

[1884] means for analyzing the pre-processed data to assess the user's comprehension, concentration, and stress level;

[1885] a means for generating or selecting learning content based on the assessment results;

[1886] means for presenting the learning content to a user;

[1887] a means for collecting, preprocessing and analyzing user reaction data;

[1888] A means for evaluating the user's learning progress based on the analyzed data and providing customized learning content;

[1889] means for sending the preprocessed data in a batch transmission;

[1890] A system that includes a means to adjust the difficulty level of learning content and include additional explanations.

[1891] (Claim 2)

[1892] 10. The system of claim 1, further comprising means for dynamically adapting the learning content.

[1893] (Claim 3)

[1894] 2. The system of claim 1, wherein the analyzing means includes a speech recognition engine that converts speech data into text data.

[1895] "Application Example 1"

[1896] (Claim 1)

[1897] means for collecting user voice, facial expression, and gaze data in real time;

[1898] means for pre-processing the speech, facial expression, and gaze data;

[1899] means for analyzing the pre-processed data to assess the user's comprehension, concentration, and stress level;

[1900] a means for generating or selecting learning content based on the assessment results;

[1901] means for presenting the learning content to a user;

[1902] a means for collecting, preprocessing and analyzing user reaction data;

[1903] A system that includes a means for providing individualized training content to workers on a production line by collecting and analyzing voice, facial expression, and gaze data in real time.

[1904] (Claim 2)

[1905] 10. The system of claim 1, further comprising means for dynamically adapting the learning content.

[1906] (Claim 3)

[1907] 2. The system of claim 1, wherein the analyzing means includes a speech recognition engine that converts speech data into text data.

[1908] "Example 2: Combining Emotion Engines"

[1909] (Claim 1)

[1910] means for collecting user voice, facial expression, and gaze data in real time;

[1911] means for pre-processing the speech, facial expression, and gaze data;

[1912] means for analyzing the pre-processed data to assess the user's comprehension, concentration, and stress level;

[1913] means for determining a user's emotional state using the evaluation results and an emotion engine;

[1914] means for generating or selecting learning content based on the evaluation results and the emotional information;

[1915] means for presenting the learning content to a user;

[1916] The system includes means for collecting user response data and again preprocessing and analyzing it.

[1917] (Claim 2)

[1918] 10. The system of claim 1, further comprising means for dynamically adapting the learning content.

[1919] (Claim 3)

[1920] 2. The system of claim 1, wherein the analyzing means includes a speech recognition engine that converts speech data into text data.

[1921] "Application example 2 when combining emotion engines"

[1922] (Claim 1) ...

Claims

1. means for collecting user voice, facial expression, and gaze data in real time; means for pre-processing the speech, facial expression, and gaze data; means for analyzing the pre-processed data to assess the user's comprehension, concentration, and stress level; a means for generating or selecting learning content based on the assessment results; means for presenting the learning content to a user; The system includes means for collecting user response data and again preprocessing and analyzing it.

2. The system of claim 1 further comprising means for dynamically adapting the learning content.

3. 2. The system of claim 1, wherein said analyzing means includes a speech recognition engine for converting speech data into text data.

Images