system

Abstract

Provide a system. 【Solution means】 Means for receiving user information and generating an individualized profile based on that information, Means for converting voice input into text data and performing natural language processing on the text, Means for converting the generated text response into voice data and providing it to the user, Means for generating learning content and recording the user's learning status, Means for adjusting the knowledge processing system using the recorded data, Means for recognizing an object from visual information and providing information, Means for selecting educational content based on the conversation and presenting it to the child, A system including the above.

Description

Technical Field

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

Background Art

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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern society, children with busy parents lack sufficient communication opportunities, and as a result, may have difficulty improving their social skills and learning abilities. Also, there is a problem that it is difficult to obtain appropriate support to solve their problems regarding making friends and studying. Furthermore, parents have the problem that it is difficult to provide effective educational support tailored to the growth of their children and to grasp their daily progress.

Means for Solving the Problems

[0005] This invention constructs a system that provides communication and learning support tailored to a child's characteristics by generating personalized user profiles. It enables interaction with children by converting voice input into text and performing natural language processing using a large-scale language model. The system provides generated responses in voice and offers efficient learning support through automatically adjusted learning content. Furthermore, it comprehensively supports a child's development by using regularly collected interaction data to enhance the artificial intelligence model and providing progress reports to parents.

[0006] "User information" refers to the data that forms the basis for creating an individualized profile of a child, including their name, age, hobbies, and learning objectives.

[0007] A "profile" is a personalized data structure generated based on user information, tailored to the child's characteristics and preferences.

[0008] "Voice input" refers to voice data that a user provides to an AI agent as a means of communication.

[0009] "Natural language processing" is a process of analysis performed on text converted from speech input, aimed at generating appropriate responses.

[0010] A "text response" is text data generated as a result of natural language processing and used to reply to the user.

[0011] "Audio data" refers to data generated to provide text responses to users as audio.

[0012] "Learning content" refers to educational materials and activities that are generated in response to the user's learning objectives and progress.

[0013] "Recorded data" refers to data collected and stored regarding user interactions and progress, which is later used for analysis and model adjustment.

[0014] An "artificial intelligence model" is the central algorithm of a system that continuously learns from user interaction data and improves the quality of interactions.

[0015] A "progress report" is a set of information provided to parents that summarizes the user's learning and communication activity history. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] The system of this invention is designed to provide interactive education and communication support to users, namely children and their guardians. This system combines speech recognition, natural language processing, large-scale language models, speech synthesis, and reinforcement learning.

[0038] When a user accesses the system, the terminal requests initial registration and prompts them to enter basic information about their child. This information is sent to the server, which creates an individualized user profile. The profile includes information such as grade level, hobbies, and goals, and all interactions are personalized based on this profile.

[0039] When a child speaks to the AI ​​agent, the device receives the audio and sends the audio data to the server. The server uses a speech recognition engine to convert the audio into text and generates an appropriate response using a large-scale language model. By referring to individualized profile information, it can create a response that is optimal for the child.

[0040] The generated text response is converted into audio data by a speech synthesis engine and played back by the device. This audio response allows the child to continue the conversation with the AI ​​agent. For example, if the child asks, "What should I learn today?", the server will suggest learning content tailored to the child's interests and learning goals.

[0041] Furthermore, learning content is filtered by the server and sent to the device in a difficulty-controlled format. As the user progresses through the learning process using this content, the system records progress information and provides problems tailored to their abilities as needed. This recorded data is periodically analyzed and used for reinforcement learning to improve the AI ​​agent's responses and the learning content.

[0042] Parents can monitor their child's learning progress and interaction history through a dashboard and receive reports from the server as needed, thereby enhancing educational support at home. In this way, the system complements parent-child communication and supports the healthy growth and learning of children.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] The user accesses the system, and a registration screen appears on their device. The user enters necessary information such as the child's name, age, hobbies, and learning goals.

[0046] Step 2:

[0047] The terminal sends the entered user information to the server. The server receives this information, stores it in a database, and then generates an individualized profile.

[0048] Step 3:

[0049] When a user speaks to the AI ​​agent, the device records the audio and sends the audio data to the server.

[0050] Step 4:

[0051] The server uses a speech recognition engine to convert speech data into text. The converted text is then passed to a large-scale language model to generate an appropriate text response.

[0052] Step 5:

[0053] The server passes the generated text response to the speech synthesis engine, which then generates the audio data. The generated audio data is then sent to the terminal.

[0054] Step 6:

[0055] The device plays back the received audio data and provides a response to the user. The user can then continue conversing with the AI ​​agent.

[0056] Step 7:

[0057] When a user requests access to learning content, the device sends that request to the server.

[0058] Step 8:

[0059] The server selects appropriate learning content based on the child's profile information, adjusts the difficulty level, and sends it to the device.

[0060] Step 9:

[0061] The device displays the received learning content and allows the user to progress through the activities. It also records learning progress data.

[0062] Step 10:

[0063] The server periodically receives interaction data from the terminals and updates the artificial intelligence model by applying reinforcement learning algorithms.

[0064] Step 11:

[0065] The server generates progress reports for parents and displays them on their devices. By reviewing these reports, parents can understand their child's learning progress.

[0066] (Example 1)

[0067] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0068] In today's family environment, there are limited means to efficiently and effectively provide education for children and support for parents. Furthermore, it is difficult to provide education tailored to each child's learning style and progress, leading to a heavy burden on parents. There is also a need for systems that effectively utilize AI technology to improve the quality of learning and deepen parent-child communication.

[0069] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0070] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information; means for converting voice input into text data and performing natural language processing on the text; means for converting the generated text response into voice data and providing it to the user; means for selecting educational content and evaluating the user's learning progress; and means for utilizing recorded data to optimize a machine learning model. This makes it possible to realize a system that provides individualized education and complements support for parents and communication between parents and children.

[0071] "User information" refers to information about a user, including their personal attributes, interests, and learning goals.

[0072] A "personalized profile" is a collection of data that holds attributes and settings specific to each user, based on user information.

[0073] "Voice input" refers to instructions or questions that a user makes to their device using their voice.

[0074] "Text data" refers to string data converted from voice input using speech recognition technology.

[0075] "Natural language processing" is a technology that analyzes text data, understands its intent and meaning, and generates responses.

[0076] A "text response" is character information that represents a reply or reaction generated by natural language processing.

[0077] "Audio data" refers to acoustic signals obtained by converting text responses using speech synthesis technology.

[0078] "Educational content" refers to a collection of learning materials and exercises provided for the purpose of user learning.

[0079] "Learning status" refers to the state of a user's learning progress and level of understanding.

[0080] A "machine learning model" is a software algorithm or system that improves its accuracy by self-improving based on data.

[0081] "Optimization" is the process of adjusting a system or algorithm to achieve a specific goal and extract the maximum effect.

[0082] "Recorded data" refers to information stored as user activity or system operation history.

[0083] This invention is designed to provide users with a personalized educational experience. Specific examples of this system are described below.

[0084] Users first access the system through a device. Available devices include personal computers, smartphones, and tablets. The device displays an initial registration screen where users enter their information. This information includes the child's name, grade level, hobbies, and goals. Once the information is entered and submitted, the device sends it to the server using a secure communication protocol. The server then uses this information to create a user profile in its database.

[0085] Next, when the user speaks to the device, the device receives voice input using its built-in or external microphone. The device sends this voice data to a server. The server uses speech recognition software (e.g., a speech recognition engine) to convert the voice data into text data. The text data is then input into a generative AI model (e.g., a large-scale language model) that performs natural language processing to generate an appropriate text response. This response is personalized by referencing the user profile.

[0086] The generated text response is converted into audio data by speech synthesis software (e.g., a speech synthesis engine) on the server. This audio data is then sent back to the terminal and played back through the terminal's speaker, conveying the message to the user.

[0087] In providing educational content, the server selects the most suitable learning materials based on the user's learning history and profile information. These materials are sent to the device, and the user can use them to progress with their learning. Parents can view progress reports generated by the server and support their child's learning as needed.

[0088] For example, if a user asks the device, "What should I study today?", the server will respond with a voice suggestion based on the user's interests and goals, such as, "Let's learn about space today."

[0089] An example of a prompt might be, "Please suggest some space topics that would interest an 8-year-old child."

[0090] In this way, we can support children's learning by providing beneficial and personalized educational support to users.

[0091] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0092] Step 1:

[0093] The user starts up the terminal and accesses the system. The terminal displays a user information input screen, where the user enters the child's name, grade level, hobbies, and goals. This input information is sent from the user to the terminal and then transmitted from the terminal to the server using a secure protocol. The server receives this information and creates an individualized user profile in its database.

[0094] Step 2:

[0095] When a user speaks into the device, the device receives voice input through its built-in microphone. The received voice is converted into a digital signal and stored in a buffer. This voice data is then sent from the device to the server. The server uses a speech recognition engine to convert the voice data into text data. This converted text data becomes the input for the next process.

[0096] Step 3:

[0097] The server passes the text data obtained through speech recognition to a generative AI model. Here, the AI ​​model creates prompts based on the user's profile information and generates the optimal text response using a large-scale language model. The generated text becomes the output of this process.

[0098] Step 4:

[0099] The text response is input into the server's speech synthesis engine and converted into audio data. This audio data is then sent from the server to the terminal. The terminal plays the received audio data through its speaker, conveying the response to the user.

[0100] Step 5:

[0101] The server analyzes the user's learning history and profile to select the most suitable educational content. The selected content is sent to the device, and the user progresses through their learning using the displayed content. Progress is fed back from the device to the server and recorded.

[0102] Step 6:

[0103] The server analyzes recorded training and interaction data and optimizes the machine learning model based on reinforcement learning. This aims to improve the accuracy of future responses and the quality of educational content.

[0104] (Application Example 1)

[0105] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0106] In educational support systems for children, there is a need for the provision of individualized educational content and the promotion of learning through visual experiences. However, conventional systems have challenges in providing sufficient personalization to meet user interests and in offering interactive experiences that integrate visual and auditory senses. Furthermore, there is a lack of means for parents to accurately understand their child's learning progress and provide appropriate support.

[0107] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0108] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for recognizing objects from visual information and providing information, and means for selecting educational content based on dialogue and presenting it to the child. This enables the provision of individualized educational content and the effective promotion of learning through visual experiences. Furthermore, it creates a system that allows parents to easily grasp progress and support education.

[0109] "User information" refers to basic data about system users and is used to generate individualized profiles.

[0110] A "personalized profile" is a data structure generated based on user information, and serves as a foundation for providing information and services optimized for each user.

[0111] "Voice input" refers to voice data from the user, which is then converted into text data for natural language processing.

[0112] "Natural language processing" is a technology that analyzes input text data and understands its meaning in context.

[0113] "Generated text response" refers to the content of the response to the user obtained as a result of processing by a large-scale language model.

[0114] "Audio data" refers to a data format that enables text responses to be output as speech through speech synthesis.

[0115] "Visual information" refers to image and video data acquired by a system through cameras and sensors, and is used for object recognition.

[0116] "Means for recognizing objects and providing information" refers to technologies that analyze visual information and provide users with information about identified objects.

[0117] "Learning content" refers to educational materials and activities provided by the system for the purpose of educating users.

[0118] "A means of selecting and presenting educational content to children based on dialogue" refers to a function that selects and provides appropriate learning content through communication with users.

[0119] A "progress report" is a report summarizing the user's learning and activity status, and is provided to parents / guardians.

[0120] A "knowledge processing system" is a system that uses artificial intelligence to analyze information and generate responses to users or select content.

[0121] This invention relates to a system for providing interactive education and communication support to child users and their guardians. The system generates an individualized profile based on user information and interacts with the child by combining speech recognition, natural language processing, large-scale language models, and speech synthesis.

[0122] The server receives voice input from the user and converts the speech to text using a speech recognition engine. The text is then analyzed using natural language processing, and an appropriate response is generated using a large-scale language model (e.g., GPT-3®). In this process, personalized responses are created by referencing profile information. The generated response text is converted into audio data by a speech synthesis engine and played back on the user's device. Specifically, Amazon Polly and Google® Speech-to-Text are used for this purpose.

[0123] The device also acquires visual information through visual sensors and performs object recognition. It generates information about the recognized objects and provides it to the child in an appropriate format. This process uses, for example, a camera and image recognition software.

[0124] As users interact with the system, educational content is automatically selected, and appropriate learning materials are provided. This content selection and delivery is optimized based on the user's learning history and interests. Learning progress is recorded on the server, and progress reports are generated for parents, allowing them to support their child's education based on up-to-date information.

[0125] For example, if a child wearing smart glasses points to a tree in a park and asks, "What kind of tree is this?", the system can analyze the visual information and provide an explanation such as, "This tree is a cherry tree."

[0126] An example of a prompt for a generative AI model is: "A child in the park asked what kind of tree it is. How would you answer through smart glasses?"

[0127] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0128] Step 1:

[0129] The user provides voice input to the device. The device captures this voice data and prepares to send it to the server. At this stage, the input is the user's voice, and the output is data in the form of an audio file.

[0130] Step 2:

[0131] The server receives audio data and converts it into text data using a speech recognition engine. This process analyzes the features of the audio waveform and generates the corresponding text. The input is audio data, and the output is text data.

[0132] Step 3:

[0133] The server performs natural language processing on the generated text data to analyze the user's intent. The input is text data, and the output is context-based analysis results, which are used in the next processing step.

[0134] Step 4:

[0135] Based on the analysis results, the server uses a generative AI model to generate appropriate response text. This process leverages a large-scale language model to create personalized content tailored to the user profile. The input is the analysis results from the previous stage, and the output is the response text.

[0136] Step 5:

[0137] The server uses a speech synthesis engine to convert the response text into audio data. At this stage, the text information is processed into audio information and made playable. The input is the response text, and the output is data in audio file format.

[0138] Step 6:

[0139] The terminal receives native audio from the server and plays it back to the user. At this stage, the user receives an audio response, enabling two-way communication. The input is audio data, and the output is an audio response to the user.

[0140] Step 7:

[0141] When the server performs object recognition using visual data, it analyzes the video data obtained from sensors, identifies the target object, and then provides that information to the user. The input is video data, and the output is information about the recognized object.

[0142] Step 8:

[0143] The server records the user's learning progress and generates progress reports for parents as needed. It also adjusts the knowledge processing system based on the recorded progress to suggest the most suitable learning content for the user. Input is user behavior data, and output is reports and adjusted system parameters.

[0144] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0145] This invention is an educational support system that combines an emotion engine to improve interaction with child users. This system consists of the following elements: user information registration, speech recognition, natural language processing, large-scale language modeling, speech synthesis, emotion recognition, and reinforcement learning.

[0146] When a user first accesses the system, a registration screen appears on their device, prompting them to enter user information. This information is sent to the server, which generates a personalized profile. This profile reflects the user's interests, age, and learning goals, and is used to guide their interaction with the system.

[0147] When a user speaks to the AI ​​agent, the device records the voice and sends the audio data to the server. The server first uses a speech recognition engine to convert the voice into text, and then uses an emotion engine to analyze the emotional state. This emotion analysis information is referenced during the natural language processing stage to generate an appropriate response that takes the child's emotions into account. The generated text response is then converted back into audio data by a speech synthesis engine and sent to the device to be provided to the user.

[0148] For example, if a user speaks in a dejected voice saying, "Today was just bad," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." Because the response changes according to the emotion, it enables more meaningful conversations for children.

[0149] Furthermore, when learning content is requested, the server selects the most suitable learning material based on profile information and emotional state and sends it to the device. As the user progresses through the learning process and the device records its progress, the server stores the child's emotions and learning progress in a database, which is then used for reinforcement learning of the artificial intelligence model.

[0150] Finally, parents can review a progress report that includes their child's learning status and emotional tendencies. This report is generated by the server and presented in an easy-to-understand visual format. This system aims to improve educational effectiveness and support parents by understanding children's emotions and providing appropriate support.

[0151] The following describes the processing flow.

[0152] Step 1:

[0153] The user starts up their device, and the registration screen is displayed. The user enters information such as their name, age, hobbies, and learning goals, and then registers.

[0154] Step 2:

[0155] The terminal sends the entered user information to the server. The server generates an individualized profile based on the received information and stores it in a database.

[0156] Step 3:

[0157] When a user speaks to the AI ​​agent, the device records the audio in real time and sends the audio data to a server.

[0158] Step 4:

[0159] The server processes the audio data through a speech recognition engine and converts it into text. The converted text data is then passed to an emotion engine to analyze the user's emotional state.

[0160] Step 5:

[0161] Based on the emotional information analyzed by the emotion engine, the server uses a natural language processing engine to generate personalized text responses that correspond to the user's emotional state.

[0162] Step 6:

[0163] The generated text response is converted into audio data by a speech synthesis engine and sent to the terminal. The terminal then plays this audio data back to the user to provide the response.

[0164] Step 7:

[0165] When a user requests learning content, the device sends the request to the server. The server refers to profile information and sentiment data, selects appropriate learning content, and sends it.

[0166] Step 8:

[0167] The device displays learning content received from the server and prompts the user to engage in activities. The user's learning progress is recorded by the device.

[0168] Step 9:

[0169] The server periodically receives interaction data and emotion data from the terminals and updates the artificial intelligence model using reinforcement learning algorithms.

[0170] Step 10:

[0171] The server generates and displays reports on the device for parents, including information on the child's learning progress and emotional tendencies. Based on this information, parents can gain a detailed understanding of their child's situation.

[0172] (Example 2)

[0173] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0174] Conventional educational support systems often fail to adequately consider user emotions during interaction, making it difficult to maximize learning effectiveness. Furthermore, there is a lack of easy ways for parents to understand their child's learning progress and emotional changes. Therefore, there is a need for flexible responses that respond to user emotions and appropriate management of progress.

[0175] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0176] In this invention, the server includes means for receiving user information and generating personalized datasets based on that information, means for converting voice input into text data and performing natural language processing on the text data, and means for determining the user's emotional state through sentiment analysis. This makes it possible to provide interactions that are tailored to the user's individual characteristics and emotions, thereby enhancing the effectiveness of learning. In addition, parents can understand their child's learning progress and emotional changes in real time through progress reports.

[0177] "User information" refers to data about individual users, including attribute information such as interests, age, and learning goals.

[0178] A "personalized dataset" is a collection of information optimized for each individual user, based on the user information collected.

[0179] "Voice input" is a method by which users transmit information to a system through voice data, which is then converted into a digital format using speech recognition technology.

[0180] "Character data" refers to linguistic information in text format converted from speech input, and is subject to analysis using natural language processing.

[0181] "Natural language processing" is a technology that enables computers to understand and process human language, and it is used to analyze the meaning of text data and generate responses.

[0182] "Emotion analysis" is the process of determining a user's emotional state from their words and actions, enabling the system to respond in accordance with the user's emotions.

[0183] "Text response" refers to character data generated by natural language processing, which is then converted into audio data as feedback to the user.

[0184] "Audio data" refers to information in audio format generated from text responses using speech synthesis technology, and is provided to the user audibly.

[0185] "Learning materials" are educational content selected based on the user's learning goals and interests, and are used during the learning process.

[0186] "Learning progress" refers to the progress a user has made in learning content, and this data is recorded and analyzed.

[0187] A "database" is a system for storing and efficiently managing information, and is used to store learning progress and user information.

[0188] A "machine learning model" is a computer program that learns on its own based on given data, and contributes to improving the performance of a system.

[0189] "Reinforcement learning" is a process of learning the optimal action through trial and error, and as a form of machine learning model, it is used to improve user interaction.

[0190] A "progress report" is a document that summarizes a user's learning progress and emotional changes, and is provided to parents to help them understand their child's learning situation.

[0191] Modes for carrying out the invention

[0192] This educational support system is designed to make user interaction more effective and is achieved by combining technologies such as speech recognition, natural language processing, sentiment analysis, and speech synthesis.

[0193] Hardware and software configuration:

[0194] 1. Terminal:

[0195] It is a device that allows users to interact using voice and text, and is equipped with a microphone and speaker.

[0196] A standard computer or smart device is used to capture and play back audio data.

[0197] Standard software used includes applications for voice recording and communication applications for data transmission and reception.

[0198] 2. Server:

[0199] At the heart of the system, it handles data processing, AI model training, and profile generation.

[0200] Google Cloud Speech-to-Text is used for speech recognition, converting the user's voice into text data.

[0201] Amazon Comprehend is used for sentiment analysis, analyzing user emotions from text data.

[0202] A large-scale language model (e.g., OpenAI® GPT) is used to generate appropriate text responses.

[0203] To convert text responses into speech, use a speech synthesis engine such as Amazon Polly.

[0204] Specific example:

[0205] For example, if a user speaks in a depressed tone, saying "Today was a bad day," the server's emotion analysis engine recognizes the negative emotion. In response to this emotion, the server generates an encouraging response as positive feedback, such as "It's okay. Shall we reflect on your day and talk about what happened?" This response is delivered to the user in real time from the device via speech synthesis.

[0206] Example of a prompt:

[0207] "When a user is expressing negative emotions, what kind of encouraging message should be generated?"

[0208] "Please suggest a method for selecting the most suitable learning content based on the user's age and interests."

[0209] This system aims to support the user's learning process and enhance educational effectiveness through emotionally resonant interactions.

[0210] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0211] Step 1:

[0212] The terminal displays a registration screen to the user. The user enters information such as interests, age, and learning goals. The entered data is sent from the terminal to the server. The terminal converts the user input into digital data and sends data packets to the server using the appropriate communication protocol.

[0213] Step 2:

[0214] The server generates personalized datasets based on the user information it receives. Using user information as input, it analyzes user attribute information through a profile generation algorithm and outputs datasets to enable customized educational support. The server stores this data in its internal database.

[0215] Step 3:

[0216] When a user gives a voice command, the device uses its built-in microphone to capture the audio and sends the audio data to the server. Specific operations are then performed to record the audio data and convert it to a digital audio format.

[0217] Step 4:

[0218] The server uses a speech recognition engine to convert the received audio data into text data. It outputs text data from the audio data as input through the speech recognition process. The server then passes this text data to the next process.

[0219] Step 5:

[0220] The server uses an emotion analysis engine to analyze the emotional state of text. It applies emotion analysis techniques to the input text data and outputs an emotional state such as positive, negative, or neutral. This emotional information is then used to generate the next response.

[0221] Step 6:

[0222] The server uses a natural language processing engine to combine text data with sentiment analysis results to generate responses. Based on the input text and its emotional state, it outputs an appropriate text response. The generated response is then formatted using a generative AI model to ensure contextual accuracy.

[0223] Step 7:

[0224] The server passes the generated text response to the speech synthesis engine, which converts it into audio data. It converts the input text response into an audio file and outputs the specific audio to be delivered to the user. The server then sends this audio data to the terminal.

[0225] Step 8:

[0226] The terminal plays audio data received from the server and provides a response to the user. Audio playback is performed using a speaker, providing feedback to the user as auditory information rather than visual.

[0227] Step 9:

[0228] When a user requests learning content, the server selects the most suitable learning materials based on the user profile and sentiment analysis data. It selects appropriate learning materials from the user information and sentiment data as input and provides them to the terminal as output.

[0229] Step 10:

[0230] The terminal displays the supplied learning content to the user and records the learning progress. It presents learning materials via the display device and saves the user's learning behavior as a log. The server collects the log data and uses it as data for reinforcement learning.

[0231] (Application Example 2)

[0232] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0233] In recent years, educational support systems have been required to incorporate technologies that enhance the effectiveness of children's learning activities. However, conventional systems lack the ability to interact with individual users while considering their emotions. Therefore, a key challenge is to create a system that can appropriately respond to the emotions children exhibit during learning and improve the quality of their learning.

[0234] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0235] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for converting voice input into text data and performing natural language processing on the text, and means for acquiring visual input and analyzing the user's emotions. This enables appropriate learning support and feedback in response to the emotions expressed by the child user.

[0236] "User information" refers to basic personal data obtained from users, and is used for profile generation and the provision of personalized services.

[0237] A "personalized profile" is a collection of information customized based on each user's interests, age, learning goals, etc., and is used to optimize the system's response and learning content.

[0238] "Voice input" refers to information that a user provides to a system via voice, which is then converted into text data using speech recognition technology.

[0239] "Natural language processing" is a technology that enables computers to understand and process human language, and it is a process for analyzing user intent and generating appropriate responses.

[0240] "Visual input" refers to acquiring visual information such as a user's facial expressions using devices like cameras, and this data is useful for analyzing the user's emotions.

[0241] "Means of analyzing emotions" refers to technologies that determine a user's emotional state based on their voice and visual information and provide appropriate feedback.

[0242] "Learning content" refers to educational materials and information provided by the system to achieve specific educational goals, and is selected appropriately according to the user's learning progress.

[0243] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and generates optimal responses and suggestions based on user behavior and emotions.

[0244] This invention is a system that provides educational support while taking the user's emotions into consideration. The system includes smart glasses used by the user, which are equipped with a microphone and a camera. When the user, a child, speaks through the glasses, the voice input is collected by the microphone in the glasses, and visual information is captured by the camera. This data is transmitted from the terminal to a server.

[0245] The server converts the provided audio data into text data using Google Speech Recognition. Visual information is processed using OpenCV to analyze the user's facial expressions and estimate their emotions. Then, based on the text data and emotion information, a natural language processing process using Hugging Face Transformers is performed to generate an appropriate response according to the user's emotions. The generated response is converted into speech using gTTS (Google Text-to-Speech) and fed back to the user through the device.

[0246] For example, if a user speaks in a depressed tone saying, "Today was a bad day," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." This allows for emotionally responsive support for children, improving the quality of their learning.

[0247] The following are examples of prompts to input into a generative AI model:

[0248] Child's comment: "Today was unlucky."

[0249] Detected emotion: Negative

[0250] Required response generation: Words of encouragement and support

[0251] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0252] Step 1:

[0253] The user speaks to the system via voice through smart glasses. This input is collected by a microphone built into the smart glasses. Specifically, the smart glasses capture the voice data and save it as a digital audio file. The input is the user's spoken words, and the output is digital audio data.

[0254] Step 2:

[0255] The device sends this audio data to the server. The server uses Google Speech Recognition to convert the audio data into text format. It analyzes the audio waveform data and converts it into a string of words using a speech recognition algorithm. The input is audio data, and the output is text data.

[0256] Step 3:

[0257] Visual information is collected by a camera, and the terminal sends the video data to a server. The server processes this video data using OpenCV and analyzes the user's facial expressions. Specifically, an image processing algorithm extracts facial feature points and determines the emotional state. The input is video data, and the output is emotion data.

[0258] Step 4:

[0259] The server performs natural language processing based on the acquired text and sentiment data. It uses Hugging Face Transformers to generate appropriate responses corresponding to emotions. The generative AI model analyzes text time-series data and sentiment characteristics, processing the data to output appropriate language responses. The input is text data and sentiment data, and the output is a text-based response.

[0260] Step 5:

[0261] The server converts the generated text response into audio data using gTTS. By converting natural language text into synthesized speech, the output is made easier for the user to understand. The input is a text-based response, and the output is audio data.

[0262] Step 6:

[0263] The terminal receives audio data transmitted from the server and plays it back to the user. The audio is played through the smart glasses' speaker, and the user receives the system's response audibly. The input is audio data, and the output is audio output to the user.

[0264] This series of processes allows users to receive emotionally responsive interactions.

[0265] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0266] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0267] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0268] [Second Embodiment]

[0269] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0270] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0271] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0272] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0273] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0274] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0275] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0276] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0277] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0278] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0279] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0280] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0281] The system of this invention is designed to provide interactive education and communication support to users, namely children and their guardians. This system combines speech recognition, natural language processing, large-scale language models, speech synthesis, and reinforcement learning.

[0282] When a user accesses the system, the terminal requests initial registration and prompts them to enter basic information about their child. This information is sent to the server, which creates an individualized user profile. The profile includes information such as grade level, hobbies, and goals, and all interactions are personalized based on this profile.

[0283] When a child talks to the AI agent, the terminal receives the voice and sends the voice data to the server. The server uses a speech recognition engine to convert the voice into text and uses a large language model to generate an appropriate response. At this time, by referring to the individualized profile information, an optimal response for the child can be created.

[0284] The generated text response is converted into voice data by a text-to-speech engine and played back by the terminal. With this voice response, the child can continue the conversation with the AI agent. As a specific example, when a child asks "What should I learn today?", the server proposes learning content according to the child's interests and learning goals.

[0285] Furthermore, the learning content is screened by the server and sent to the terminal in a form with the difficulty level managed. When the user uses this content to proceed with learning, the system records the progress information and provides questions suitable for the ability as needed. The recorded data is periodically analyzed and used for reinforcement learning to improve the responses of the AI agent and the learning content.

[0286] Parents can monitor the child's learning progress and interaction history through the dashboard and receive reports from the server as needed, thereby strengthening the educational support at home. In this way, this system complements parent-child communication and supports the healthy growth and learning of children.

[0287] The following describes the process flow.

[0288] Step 1:

[0289] The user accesses the system, and a registration screen is displayed on the terminal. The user inputs necessary information such as the child's name, age, hobbies, and learning goals.

[0290] Step 2:

[0291] The terminal sends the entered user information to the server. The server receives this information, stores it in a database, and then generates an individualized profile.

[0292] Step 3:

[0293] When a user speaks to the AI ​​agent, the device records the audio and sends the audio data to the server.

[0294] Step 4:

[0295] The server uses a speech recognition engine to convert speech data into text. The converted text is then passed to a large-scale language model to generate an appropriate text response.

[0296] Step 5:

[0297] The server passes the generated text response to the speech synthesis engine, which then generates the audio data. The generated audio data is then sent to the terminal.

[0298] Step 6:

[0299] The device plays back the received audio data and provides a response to the user. The user can then continue conversing with the AI ​​agent.

[0300] Step 7:

[0301] When a user requests access to learning content, the device sends that request to the server.

[0302] Step 8:

[0303] The server selects appropriate learning content based on the child's profile information, adjusts the difficulty level, and sends it to the device.

[0304] Step 9:

[0305] The terminal displays the received learning content to enable the user to proceed with activities. It also records learning progress data.

[0306] Step 10:

[0307] The server periodically receives interaction data from the terminal and applies a reinforcement learning algorithm to update the artificial intelligence model.

[0308] Step 11:

[0309] The server generates a progress report for the guardian and displays it through the terminal. By checking this, the guardian can grasp the child's learning situation.

[0310] (Example 1)

[0311] Next, Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0312] In the modern home environment, there are limited means to efficiently and effectively carry out children's education and guardian support. Also, it is difficult to provide education according to each child's learning style and progress, and there is an issue that the burden on guardians increases. Furthermore, there is a demand for a system that effectively utilizes AI technology to improve the quality of learning and deepen parent-child communication.

[0313] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0314] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information; means for converting voice input into text data and performing natural language processing on the text; means for converting the generated text response into voice data and providing it to the user; means for selecting educational content and evaluating the user's learning progress; and means for utilizing recorded data to optimize a machine learning model. This makes it possible to realize a system that provides individualized education and complements support for parents and communication between parents and children.

[0315] "User information" refers to information about a user, including their personal attributes, interests, and learning goals.

[0316] A "personalized profile" is a collection of data that holds attributes and settings specific to each user, based on user information.

[0317] "Voice input" refers to instructions or questions that a user makes to their device using their voice.

[0318] "Text data" refers to string data converted from voice input using speech recognition technology.

[0319] "Natural language processing" is a technology that analyzes text data, understands its intent and meaning, and generates responses.

[0320] A "text response" is character information that represents a reply or reaction generated by natural language processing.

[0321] "Audio data" refers to acoustic signals obtained by converting text responses using speech synthesis technology.

[0322] "Educational content" refers to a collection of learning materials and exercises provided for the purpose of user learning.

[0323] "Learning status" refers to the state of a user's learning progress and level of understanding.

[0324] A "machine learning model" is a software algorithm or system that improves its accuracy by self-improving based on data.

[0325] "Optimization" is the process of adjusting a system or algorithm to achieve a specific goal and extract the maximum effect.

[0326] "Recorded data" refers to information stored as user activity or system operation history.

[0327] This invention is designed to provide users with a personalized educational experience. Specific examples of this system are described below.

[0328] Users first access the system through a device. Available devices include personal computers, smartphones, and tablets. The device displays an initial registration screen where users enter their information. This information includes the child's name, grade level, hobbies, and goals. Once the information is entered and submitted, the device sends it to the server using a secure communication protocol. The server then uses this information to create a user profile in its database.

[0329] Next, when the user speaks to the device, the device receives voice input using its built-in or external microphone. The device sends this voice data to a server. The server uses speech recognition software (e.g., a speech recognition engine) to convert the voice data into text data. The text data is then input into a generative AI model (e.g., a large-scale language model) that performs natural language processing to generate an appropriate text response. This response is personalized by referencing the user profile.

[0330] The generated text response is converted into audio data by speech synthesis software (e.g., a speech synthesis engine) on the server. This audio data is then sent back to the terminal and played back through the terminal's speaker, conveying the message to the user.

[0331] In providing educational content, the server selects the most suitable learning materials based on the user's learning history and profile information. These materials are sent to the device, and the user can use them to progress with their learning. Parents can view progress reports generated by the server and support their child's learning as needed.

[0332] For example, if a user asks the device, "What should I study today?", the server will respond with a voice suggestion based on the user's interests and goals, such as, "Let's learn about space today."

[0333] An example of a prompt might be, "Please suggest some space topics that would interest an 8-year-old child."

[0334] In this way, we can support children's learning by providing beneficial and personalized educational support to users.

[0335] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0336] Step 1:

[0337] The user starts up the terminal and accesses the system. The terminal displays a user information input screen, where the user enters the child's name, grade level, hobbies, and goals. This input information is sent from the user to the terminal and then transmitted from the terminal to the server using a secure protocol. The server receives this information and creates an individualized user profile in its database.

[0338] Step 2:

[0339] When a user speaks into the device, the device receives voice input through its built-in microphone. The received voice is converted into a digital signal and stored in a buffer. This voice data is then sent from the device to the server. The server uses a speech recognition engine to convert the voice data into text data. This converted text data becomes the input for the next process.

[0340] Step 3:

[0341] The server passes the text data obtained through speech recognition to a generative AI model. Here, the AI ​​model creates prompts based on the user's profile information and generates the optimal text response using a large-scale language model. The generated text becomes the output of this process.

[0342] Step 4:

[0343] The text response is input into the server's speech synthesis engine and converted into audio data. This audio data is then sent from the server to the terminal. The terminal plays the received audio data through its speaker, conveying the response to the user.

[0344] Step 5:

[0345] The server analyzes the user's learning history and profile to select the most suitable educational content. The selected content is sent to the device, and the user progresses through their learning using the displayed content. Progress is fed back from the device to the server and recorded.

[0346] Step 6:

[0347] The server analyzes recorded training and interaction data and optimizes the machine learning model based on reinforcement learning. This aims to improve the accuracy of future responses and the quality of educational content.

[0348] (Application Example 1)

[0349] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0350] In educational support systems for children, there is a need for the provision of individualized educational content and the promotion of learning through visual experiences. However, conventional systems have challenges in providing sufficient personalization to meet user interests and in offering interactive experiences that integrate visual and auditory senses. Furthermore, there is a lack of means for parents to accurately understand their child's learning progress and provide appropriate support.

[0351] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0352] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for recognizing objects from visual information and providing information, and means for selecting educational content based on dialogue and presenting it to the child. This enables the provision of individualized educational content and the effective promotion of learning through visual experiences. Furthermore, it creates a system that allows parents to easily grasp progress and support education.

[0353] "User information" refers to basic data about system users and is used to generate individualized profiles.

[0354] A "personalized profile" is a data structure generated based on user information, and serves as a foundation for providing information and services optimized for each user.

[0355] "Voice input" refers to voice data from the user, which is then converted into text data for natural language processing.

[0356] "Natural language processing" is a technology that analyzes input text data and understands its meaning in context.

[0357] "Generated text response" refers to the content of the response to the user obtained as a result of processing by a large-scale language model.

[0358] "Audio data" refers to a data format that enables text responses to be output as speech through speech synthesis.

[0359] "Visual information" refers to image and video data acquired by a system through cameras and sensors, and is used for object recognition.

[0360] "Means for recognizing objects and providing information" refers to technologies that analyze visual information and provide users with information about identified objects.

[0361] "Learning content" refers to educational materials and activities provided by the system for the purpose of educating users.

[0362] "A means of selecting and presenting educational content to children based on dialogue" refers to a function that selects and provides appropriate learning content through communication with users.

[0363] A "progress report" is a report summarizing the user's learning and activity status, and is provided to parents / guardians.

[0364] A "knowledge processing system" is a system that uses artificial intelligence to analyze information and generate responses to users or select content.

[0365] This invention relates to a system for providing interactive education and communication support to child users and their guardians. The system generates an individualized profile based on user information and interacts with the child by combining speech recognition, natural language processing, large-scale language models, and speech synthesis.

[0366] The server receives voice input from the user and converts the speech to text using a speech recognition engine. The text is then analyzed using natural language processing, and an appropriate response is generated using a large-scale language model (e.g., GPT-3). In this process, personalized responses are created by referencing profile information. The generated response text is converted into audio data by a speech synthesis engine and played back on the user's device. Specifically, Amazon Polly and Google Speech-to-Text are used for this purpose.

[0367] The device also acquires visual information through visual sensors and performs object recognition. It generates information about the recognized objects and provides it to the child in an appropriate format. This process uses, for example, a camera and image recognition software.

[0368] As users interact with the system, educational content is automatically selected, and appropriate learning materials are provided. This content selection and delivery is optimized based on the user's learning history and interests. Learning progress is recorded on the server, and progress reports are generated for parents, allowing them to support their child's education based on up-to-date information.

[0369] For example, if a child wearing smart glasses points to a tree in a park and asks, "What kind of tree is this?", the system can analyze the visual information and provide an explanation such as, "This tree is a cherry tree."

[0370] An example of a prompt for a generative AI model is: "A child in the park asked what kind of tree it is. How would you answer through smart glasses?"

[0371] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0372] Step 1:

[0373] The user provides voice input to the device. The device captures this voice data and prepares to send it to the server. At this stage, the input is the user's voice, and the output is data in the form of an audio file.

[0374] Step 2:

[0375] The server receives audio data and converts it into text data using a speech recognition engine. This process analyzes the features of the audio waveform and generates the corresponding text. The input is audio data, and the output is text data.

[0376] Step 3:

[0377] The server performs natural language processing on the generated text data to analyze the user's intent. The input is text data, and the output is context-based analysis results, which are used in the next processing step.

[0378] Step 4:

[0379] Based on the analysis results, the server uses a generative AI model to generate appropriate response text. This process leverages a large-scale language model to create personalized content tailored to the user profile. The input is the analysis results from the previous stage, and the output is the response text.

[0380] Step 5:

[0381] The server uses a speech synthesis engine to convert the response text into audio data. At this stage, the text information is processed into audio information and made playable. The input is the response text, and the output is data in audio file format.

[0382] Step 6:

[0383] The terminal receives native audio from the server and plays it back to the user. At this stage, the user receives an audio response, enabling two-way communication. The input is audio data, and the output is an audio response to the user.

[0384] Step 7:

[0385] When the server performs object recognition using visual data, it analyzes the video data obtained from sensors, identifies the target object, and then provides that information to the user. The input is video data, and the output is information about the recognized object.

[0386] Step 8:

[0387] The server records the user's learning progress and generates progress reports for parents as needed. It also adjusts the knowledge processing system based on the recorded progress to suggest the most suitable learning content for the user. Input is user behavior data, and output is reports and adjusted system parameters.

[0388] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0389] This invention is an educational support system that combines an emotion engine to improve interaction with child users. This system consists of the following elements: user information registration, speech recognition, natural language processing, large-scale language modeling, speech synthesis, emotion recognition, and reinforcement learning.

[0390] When a user first accesses the system, a registration screen appears on their device, prompting them to enter user information. This information is sent to the server, which generates a personalized profile. This profile reflects the user's interests, age, and learning goals, and is used to guide their interaction with the system.

[0391] When a user speaks to the AI ​​agent, the device records the voice and sends the audio data to the server. The server first uses a speech recognition engine to convert the voice into text, and then uses an emotion engine to analyze the emotional state. This emotion analysis information is referenced during the natural language processing stage to generate an appropriate response that takes the child's emotions into account. The generated text response is then converted back into audio data by a speech synthesis engine and sent to the device to be provided to the user.

[0392] For example, if a user speaks in a dejected voice saying, "Today was just bad," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." Because the response changes according to the emotion, it enables more meaningful conversations for children.

[0393] Furthermore, when learning content is requested, the server selects the most suitable learning material based on profile information and emotional state and sends it to the device. As the user progresses through the learning process and the device records its progress, the server stores the child's emotions and learning progress in a database, which is then used for reinforcement learning of the artificial intelligence model.

[0394] Finally, parents can review a progress report that includes their child's learning status and emotional tendencies. This report is generated by the server and presented in an easy-to-understand visual format. This system aims to improve educational effectiveness and support parents by understanding children's emotions and providing appropriate support.

[0395] The following describes the processing flow.

[0396] Step 1:

[0397] The user starts up their device, and the registration screen is displayed. The user enters information such as their name, age, hobbies, and learning goals, and then registers.

[0398] Step 2:

[0399] The terminal sends the entered user information to the server. The server generates an individualized profile based on the received information and stores it in a database.

[0400] Step 3:

[0401] When a user speaks to the AI ​​agent, the device records the audio in real time and sends the audio data to a server.

[0402] Step 4:

[0403] The server processes the audio data through a speech recognition engine and converts it into text. The converted text data is then passed to an emotion engine to analyze the user's emotional state.

[0404] Step 5:

[0405] Based on the emotional information analyzed by the emotion engine, the server uses a natural language processing engine to generate personalized text responses that correspond to the user's emotional state.

[0406] Step 6:

[0407] The generated text response is converted into audio data by a speech synthesis engine and sent to the terminal. The terminal then plays this audio data back to the user to provide the response.

[0408] Step 7:

[0409] When a user requests learning content, the device sends the request to the server. The server refers to profile information and sentiment data, selects appropriate learning content, and sends it.

[0410] Step 8:

[0411] The device displays learning content received from the server and prompts the user to engage in activities. The user's learning progress is recorded by the device.

[0412] Step 9:

[0413] The server periodically receives interaction data and emotion data from the terminals and updates the artificial intelligence model using reinforcement learning algorithms.

[0414] Step 10:

[0415] The server generates and displays reports on the device for parents, including information on the child's learning progress and emotional tendencies. Based on this information, parents can gain a detailed understanding of their child's situation.

[0416] (Example 2)

[0417] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0418] Conventional educational support systems often fail to adequately consider user emotions during interaction, making it difficult to maximize learning effectiveness. Furthermore, there is a lack of easy ways for parents to understand their child's learning progress and emotional changes. Therefore, there is a need for flexible responses that respond to user emotions and appropriate management of progress.

[0419] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0420] In this invention, the server includes means for receiving user information and generating personalized datasets based on that information, means for converting voice input into text data and performing natural language processing on the text data, and means for determining the user's emotional state through sentiment analysis. This makes it possible to provide interactions that are tailored to the user's individual characteristics and emotions, thereby enhancing the effectiveness of learning. In addition, parents can understand their child's learning progress and emotional changes in real time through progress reports.

[0421] "User information" refers to data about individual users, including attribute information such as interests, age, and learning goals.

[0422] A "personalized dataset" is a collection of information optimized for each individual user, based on the user information collected.

[0423] "Voice input" is a method by which users transmit information to a system through voice data, which is then converted into a digital format using speech recognition technology.

[0424] "Character data" refers to linguistic information in text format converted from speech input, and is subject to analysis using natural language processing.

[0425] "Natural language processing" is a technology that enables computers to understand and process human language, and it is used to analyze the meaning of text data and generate responses.

[0426] "Emotion analysis" is the process of determining a user's emotional state from their words and actions, enabling the system to respond in accordance with the user's emotions.

[0427] "Text response" refers to character data generated by natural language processing, which is then converted into audio data as feedback to the user.

[0428] "Audio data" refers to information in audio format generated from text responses using speech synthesis technology, and is provided to the user audibly.

[0429] "Learning materials" are educational content selected based on the user's learning goals and interests, and are used during the learning process.

[0430] "Learning progress" refers to the progress a user has made in learning content, and this data is recorded and analyzed.

[0431] A "database" is a system for storing and efficiently managing information, and is used to store learning progress and user information.

[0432] A "machine learning model" is a computer program that learns on its own based on given data, and contributes to improving the performance of a system.

[0433] "Reinforcement learning" is a process of learning the optimal action through trial and error, and as a form of machine learning model, it is used to improve user interaction.

[0434] A "progress report" is a document that summarizes a user's learning progress and emotional changes, and is provided to parents to help them understand their child's learning situation.

[0435] Modes for carrying out the invention

[0436] This educational support system is designed to make user interaction more effective and is achieved by combining technologies such as speech recognition, natural language processing, sentiment analysis, and speech synthesis.

[0437] Hardware and software configuration:

[0438] 1. Terminal:

[0439] It is a device that allows users to interact using voice and text, and is equipped with a microphone and speaker.

[0440] A standard computer or smart device is used to capture and play back audio data.

[0441] Standard software used includes applications for voice recording and communication applications for data transmission and reception.

[0442] 2. Server:

[0443] At the heart of the system, it handles data processing, AI model training, and profile generation.

[0444] Google Cloud Speech-to-Text is used for speech recognition, converting the user's voice into text data.

[0445] Amazon Comprehend is used for sentiment analysis, analyzing user emotions from text data.

[0446] A large-scale language model (e.g., OpenAI GPT) is used to generate appropriate text responses.

[0447] To convert text responses into speech, use a speech synthesis engine such as Amazon Polly.

[0448] Specific example:

[0449] For example, if a user speaks in a depressed tone, saying "Today was a bad day," the server's emotion analysis engine recognizes the negative emotion. In response to this emotion, the server generates an encouraging response as positive feedback, such as "It's okay. Shall we reflect on your day and talk about what happened?" This response is delivered to the user in real time from the device via speech synthesis.

[0450] Example of a prompt:

[0451] "When a user is expressing negative emotions, what kind of encouraging message should be generated?"

[0452] "Please suggest a method for selecting the most suitable learning content based on the user's age and interests."

[0453] This system aims to support the user's learning process and enhance educational effectiveness through emotionally resonant interactions.

[0454] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0455] Step 1:

[0456] The terminal displays a registration screen to the user. The user enters information such as interests, age, and learning goals. The entered data is sent from the terminal to the server. The terminal converts the user input into digital data and sends data packets to the server using the appropriate communication protocol.

[0457] Step 2:

[0458] The server generates personalized datasets based on the user information it receives. Using user information as input, it analyzes user attribute information through a profile generation algorithm and outputs datasets to enable customized educational support. The server stores this data in its internal database.

[0459] Step 3:

[0460] When a user gives a voice command, the device uses its built-in microphone to capture the audio and sends the audio data to the server. Specific operations are then performed to record the audio data and convert it to a digital audio format.

[0461] Step 4:

[0462] The server uses a speech recognition engine to convert the received audio data into text data. It outputs text data from the audio data as input through the speech recognition process. The server then passes this text data to the next process.

[0463] Step 5:

[0464] The server uses an emotion analysis engine to analyze the emotional state of text. It applies emotion analysis techniques to the input text data and outputs an emotional state such as positive, negative, or neutral. This emotional information is then used to generate the next response.

[0465] Step 6:

[0466] The server uses a natural language processing engine to combine text data with sentiment analysis results to generate responses. Based on the input text and its emotional state, it outputs an appropriate text response. The generated response is then formatted using a generative AI model to ensure contextual accuracy.

[0467] Step 7:

[0468] The server passes the generated text response to the speech synthesis engine, which converts it into audio data. It converts the input text response into an audio file and outputs the specific audio to be delivered to the user. The server then sends this audio data to the terminal.

[0469] Step 8:

[0470] The terminal plays audio data received from the server and provides a response to the user. Audio playback is performed using a speaker, providing feedback to the user as auditory information rather than visual.

[0471] Step 9:

[0472] When a user requests learning content, the server selects the most suitable learning materials based on the user profile and sentiment analysis data. It selects appropriate learning materials from the user information and sentiment data as input and provides them to the terminal as output.

[0473] Step 10:

[0474] The terminal displays the supplied learning content to the user and records the learning progress. It presents learning materials via the display device and saves the user's learning behavior as a log. The server collects the log data and uses it as data for reinforcement learning.

[0475] (Application Example 2)

[0476] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0477] In recent years, educational support systems have been required to incorporate technologies that enhance the effectiveness of children's learning activities. However, conventional systems lack the ability to interact with individual users while considering their emotions. Therefore, a key challenge is to create a system that can appropriately respond to the emotions children exhibit during learning and improve the quality of their learning.

[0478] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0479] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for converting voice input into text data and performing natural language processing on the text, and means for acquiring visual input and analyzing the user's emotions. This enables appropriate learning support and feedback in response to the emotions expressed by the child user.

[0480] "User information" refers to basic personal data obtained from users, and is used for profile generation and the provision of personalized services.

[0481] A "personalized profile" is a collection of information customized based on each user's interests, age, learning goals, etc., and is used to optimize the system's response and learning content.

[0482] "Voice input" refers to information that a user provides to a system via voice, which is then converted into text data using speech recognition technology.

[0483] "Natural language processing" is a technology that enables computers to understand and process human language, and it is a process for analyzing user intent and generating appropriate responses.

[0484] "Visual input" refers to acquiring visual information such as a user's facial expressions using devices like cameras, and this data is useful for analyzing the user's emotions.

[0485] "Means of analyzing emotions" refers to technologies that determine a user's emotional state based on their voice and visual information and provide appropriate feedback.

[0486] "Learning content" refers to educational materials and information provided by the system to achieve specific educational goals, and is selected appropriately according to the user's learning progress.

[0487] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and generates optimal responses and suggestions based on user behavior and emotions.

[0488] This invention is a system that provides educational support while taking the user's emotions into consideration. The system includes smart glasses used by the user, which are equipped with a microphone and a camera. When the user, a child, speaks through the glasses, the voice input is collected by the microphone in the glasses, and visual information is captured by the camera. This data is transmitted from the terminal to a server.

[0489] The server converts the provided audio data into text data using Google Speech Recognition. Visual information is processed using OpenCV to analyze the user's facial expressions and estimate their emotions. Then, based on the text data and emotion information, a natural language processing process using Hugging Face Transformers is performed to generate an appropriate response according to the user's emotions. The generated response is converted into speech using gTTS (Google Text-to-Speech) and fed back to the user through the device.

[0490] For example, if a user speaks in a depressed tone saying, "Today was a bad day," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." This allows for emotionally responsive support for children, improving the quality of their learning.

[0491] The following are examples of prompts to input into a generative AI model:

[0492] Child's comment: "Today was unlucky."

[0493] Detected emotion: Negative

[0494] Required response generation: Words of encouragement and support

[0495] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0496] Step 1:

[0497] The user speaks to the system via voice through smart glasses. This input is collected by a microphone built into the smart glasses. Specifically, the smart glasses capture the voice data and save it as a digital audio file. The input is the user's spoken words, and the output is digital audio data.

[0498] Step 2:

[0499] The device sends this audio data to the server. The server uses Google Speech Recognition to convert the audio data into text format. It analyzes the audio waveform data and converts it into a string of words using a speech recognition algorithm. The input is audio data, and the output is text data.

[0500] Step 3:

[0501] Visual information is collected by a camera, and the terminal sends the video data to a server. The server processes this video data using OpenCV and analyzes the user's facial expressions. Specifically, an image processing algorithm extracts facial feature points and determines the emotional state. The input is video data, and the output is emotion data.

[0502] Step 4:

[0503] The server performs natural language processing based on the acquired text and sentiment data. It uses Hugging Face Transformers to generate appropriate responses corresponding to emotions. The generative AI model analyzes text time-series data and sentiment characteristics, processing the data to output appropriate language responses. The input is text data and sentiment data, and the output is a text-based response.

[0504] Step 5:

[0505] The server converts the generated text response into audio data using gTTS. By converting natural language text into synthesized speech, the output is made easier for the user to understand. The input is a text-based response, and the output is audio data.

[0506] Step 6:

[0507] The terminal receives audio data transmitted from the server and plays it back to the user. The audio is played through the smart glasses' speaker, and the user receives the system's response audibly. The input is audio data, and the output is audio output to the user.

[0508] This series of processes allows users to receive emotionally responsive interactions.

[0509] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0510] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0511] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0512] [Third Embodiment]

[0513] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0514] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0515] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0516] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0517] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0518] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0519] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0520] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0521] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0522] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0523] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0524] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0525] The system of this invention is designed to provide interactive education and communication support to users, namely children and their guardians. This system combines speech recognition, natural language processing, large-scale language models, speech synthesis, and reinforcement learning.

[0526] When a user accesses the system, the terminal requests initial registration and prompts them to enter basic information about their child. This information is sent to the server, which creates an individualized user profile. The profile includes information such as grade level, hobbies, and goals, and all interactions are personalized based on this profile.

[0527] When a child speaks to the AI ​​agent, the device receives the audio and sends the audio data to the server. The server uses a speech recognition engine to convert the audio into text and generates an appropriate response using a large-scale language model. By referring to individualized profile information, it can create a response that is optimal for the child.

[0528] The generated text response is converted into audio data by a speech synthesis engine and played back by the device. This audio response allows the child to continue the conversation with the AI ​​agent. For example, if the child asks, "What should I learn today?", the server will suggest learning content tailored to the child's interests and learning goals.

[0529] Furthermore, learning content is filtered by the server and sent to the device in a difficulty-controlled format. As the user progresses through the learning process using this content, the system records progress information and provides problems tailored to their abilities as needed. This recorded data is periodically analyzed and used for reinforcement learning to improve the AI ​​agent's responses and the learning content.

[0530] Parents can monitor their child's learning progress and interaction history through a dashboard and receive reports from the server as needed, thereby enhancing educational support at home. In this way, the system complements parent-child communication and supports the healthy growth and learning of children.

[0531] The following describes the processing flow.

[0532] Step 1:

[0533] The user accesses the system, and a registration screen appears on their device. The user enters necessary information such as the child's name, age, hobbies, and learning goals.

[0534] Step 2:

[0535] The terminal sends the entered user information to the server. The server receives this information, stores it in a database, and then generates an individualized profile.

[0536] Step 3:

[0537] When a user speaks to the AI ​​agent, the device records the audio and sends the audio data to the server.

[0538] Step 4:

[0539] The server uses a speech recognition engine to convert speech data into text. The converted text is then passed to a large-scale language model to generate an appropriate text response.

[0540] Step 5:

[0541] The server passes the generated text response to the speech synthesis engine, which then generates the audio data. The generated audio data is then sent to the terminal.

[0542] Step 6:

[0543] The device plays back the received audio data and provides a response to the user. The user can then continue conversing with the AI ​​agent.

[0544] Step 7:

[0545] When a user requests access to learning content, the device sends that request to the server.

[0546] Step 8:

[0547] The server selects appropriate learning content based on the child's profile information, adjusts the difficulty level, and sends it to the device.

[0548] Step 9:

[0549] The device displays the received learning content and allows the user to progress through the activities. It also records learning progress data.

[0550] Step 10:

[0551] The server periodically receives interaction data from the terminals and updates the artificial intelligence model by applying reinforcement learning algorithms.

[0552] Step 11:

[0553] The server generates progress reports for parents and displays them on their devices. By reviewing these reports, parents can understand their child's learning progress.

[0554] (Example 1)

[0555] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0556] In today's family environment, there are limited means to efficiently and effectively provide education for children and support for parents. Furthermore, it is difficult to provide education tailored to each child's learning style and progress, leading to a heavy burden on parents. There is also a need for systems that effectively utilize AI technology to improve the quality of learning and deepen parent-child communication.

[0557] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0558] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information; means for converting voice input into text data and performing natural language processing on the text; means for converting the generated text response into voice data and providing it to the user; means for selecting educational content and evaluating the user's learning progress; and means for utilizing recorded data to optimize a machine learning model. This makes it possible to realize a system that provides individualized education and complements support for parents and communication between parents and children.

[0559] "User information" refers to information about a user, including their personal attributes, interests, and learning goals.

[0560] A "personalized profile" is a collection of data that holds attributes and settings specific to each user, based on user information.

[0561] "Voice input" refers to instructions or questions that a user makes to their device using their voice.

[0562] "Text data" refers to string data converted from voice input using speech recognition technology.

[0563] "Natural language processing" is a technology that analyzes text data, understands its intent and meaning, and generates responses.

[0564] A "text response" is character information that represents a reply or reaction generated by natural language processing.

[0565] "Audio data" refers to acoustic signals obtained by converting text responses using speech synthesis technology.

[0566] "Educational content" refers to a collection of learning materials and exercises provided for the purpose of user learning.

[0567] "Learning status" refers to the state of a user's learning progress and level of understanding.

[0568] A "machine learning model" is a software algorithm or system that improves its accuracy by self-improving based on data.

[0569] "Optimization" is the process of adjusting a system or algorithm to achieve a specific goal and extract the maximum effect.

[0570] "Recorded data" refers to information stored as user activity or system operation history.

[0571] This invention is designed to provide users with a personalized educational experience. Specific examples of this system are described below.

[0572] Users first access the system through a device. Available devices include personal computers, smartphones, and tablets. The device displays an initial registration screen where users enter their information. This information includes the child's name, grade level, hobbies, and goals. Once the information is entered and submitted, the device sends it to the server using a secure communication protocol. The server then uses this information to create a user profile in its database.

[0573] Next, when the user speaks to the device, the device receives voice input using its built-in or external microphone. The device sends this voice data to a server. The server uses speech recognition software (e.g., a speech recognition engine) to convert the voice data into text data. The text data is then input into a generative AI model (e.g., a large-scale language model) that performs natural language processing to generate an appropriate text response. This response is personalized by referencing the user profile.

[0574] The generated text response is converted into audio data by speech synthesis software (e.g., a speech synthesis engine) on the server. This audio data is then sent back to the terminal and played back through the terminal's speaker, conveying the message to the user.

[0575] In providing educational content, the server selects the most suitable learning materials based on the user's learning history and profile information. These materials are sent to the device, and the user can use them to progress with their learning. Parents can view progress reports generated by the server and support their child's learning as needed.

[0576] For example, if a user asks the device, "What should I study today?", the server will respond with a voice suggestion based on the user's interests and goals, such as, "Let's learn about space today."

[0577] An example of a prompt might be, "Please suggest some space topics that would interest an 8-year-old child."

[0578] In this way, we can support children's learning by providing beneficial and personalized educational support to users.

[0579] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0580] Step 1:

[0581] The user starts up the terminal and accesses the system. The terminal displays a user information input screen, where the user enters the child's name, grade level, hobbies, and goals. This input information is sent from the user to the terminal and then transmitted from the terminal to the server using a secure protocol. The server receives this information and creates an individualized user profile in its database.

[0582] Step 2:

[0583] When a user speaks into the device, the device receives voice input through its built-in microphone. The received voice is converted into a digital signal and stored in a buffer. This voice data is then sent from the device to the server. The server uses a speech recognition engine to convert the voice data into text data. This converted text data becomes the input for the next process.

[0584] Step 3:

[0585] The server passes the text data obtained through speech recognition to a generative AI model. Here, the AI ​​model creates prompts based on the user's profile information and generates the optimal text response using a large-scale language model. The generated text becomes the output of this process.

[0586] Step 4:

[0587] The text response is input into the server's speech synthesis engine and converted into audio data. This audio data is then sent from the server to the terminal. The terminal plays the received audio data through its speaker, conveying the response to the user.

[0588] Step 5:

[0589] The server analyzes the user's learning history and profile to select the most suitable educational content. The selected content is sent to the device, and the user progresses through their learning using the displayed content. Progress is fed back from the device to the server and recorded.

[0590] Step 6:

[0591] The server analyzes recorded training and interaction data and optimizes the machine learning model based on reinforcement learning. This aims to improve the accuracy of future responses and the quality of educational content.

[0592] (Application Example 1)

[0593] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0594] In educational support systems for children, there is a need for the provision of individualized educational content and the promotion of learning through visual experiences. However, conventional systems have challenges in providing sufficient personalization to meet user interests and in offering interactive experiences that integrate visual and auditory senses. Furthermore, there is a lack of means for parents to accurately understand their child's learning progress and provide appropriate support.

[0595] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0596] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for recognizing objects from visual information and providing information, and means for selecting educational content based on dialogue and presenting it to the child. This enables the provision of individualized educational content and the effective promotion of learning through visual experiences. Furthermore, it creates a system that allows parents to easily grasp progress and support education.

[0597] "User information" refers to basic data about system users and is used to generate individualized profiles.

[0598] A "personalized profile" is a data structure generated based on user information, and serves as a foundation for providing information and services optimized for each user.

[0599] "Voice input" refers to voice data from the user, which is then converted into text data for natural language processing.

[0600] "Natural language processing" is a technology that analyzes input text data and understands its meaning in context.

[0601] "Generated text response" refers to the content of the response to the user obtained as a result of processing by a large-scale language model.

[0602] "Audio data" refers to a data format that enables text responses to be output as speech through speech synthesis.

[0603] "Visual information" refers to image and video data acquired by a system through cameras and sensors, and is used for object recognition.

[0604] "Means for recognizing objects and providing information" refers to technologies that analyze visual information and provide users with information about identified objects.

[0605] "Learning content" refers to educational materials and activities provided by the system for the purpose of educating users.

[0606] "A means of selecting and presenting educational content to children based on dialogue" refers to a function that selects and provides appropriate learning content through communication with users.

[0607] A "progress report" is a report summarizing the user's learning and activity status, and is provided to parents / guardians.

[0608] A "knowledge processing system" is a system that uses artificial intelligence to analyze information and generate responses to users or select content.

[0609] This invention relates to a system for providing interactive education and communication support to child users and their guardians. The system generates an individualized profile based on user information and interacts with the child by combining speech recognition, natural language processing, large-scale language models, and speech synthesis.

[0610] The server receives voice input from the user and converts the speech to text using a speech recognition engine. The text is then analyzed using natural language processing, and an appropriate response is generated using a large-scale language model (e.g., GPT-3). In this process, personalized responses are created by referencing profile information. The generated response text is converted into audio data by a speech synthesis engine and played back on the user's device. Specifically, Amazon Polly and Google Speech-to-Text are used for this purpose.

[0611] The device also acquires visual information through visual sensors and performs object recognition. It generates information about the recognized objects and provides it to the child in an appropriate format. This process uses, for example, a camera and image recognition software.

[0612] As users interact with the system, educational content is automatically selected, and appropriate learning materials are provided. This content selection and delivery is optimized based on the user's learning history and interests. Learning progress is recorded on the server, and progress reports are generated for parents, allowing them to support their child's education based on up-to-date information.

[0613] For example, if a child wearing smart glasses points to a tree in a park and asks, "What kind of tree is this?", the system can analyze the visual information and provide an explanation such as, "This tree is a cherry tree."

[0614] An example of a prompt for a generative AI model is: "A child in the park asked what kind of tree it is. How would you answer through smart glasses?"

[0615] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0616] Step 1:

[0617] The user provides voice input to the device. The device captures this voice data and prepares to send it to the server. At this stage, the input is the user's voice, and the output is data in the form of an audio file.

[0618] Step 2:

[0619] The server receives audio data and converts it into text data using a speech recognition engine. This process analyzes the features of the audio waveform and generates the corresponding text. The input is audio data, and the output is text data.

[0620] Step 3:

[0621] The server performs natural language processing on the generated text data to analyze the user's intent. The input is text data, and the output is context-based analysis results, which are used in the next processing step.

[0622] Step 4:

[0623] Based on the analysis results, the server uses a generative AI model to generate appropriate response text. This process leverages a large-scale language model to create personalized content tailored to the user profile. The input is the analysis results from the previous stage, and the output is the response text.

[0624] Step 5:

[0625] The server uses a speech synthesis engine to convert the response text into audio data. At this stage, the text information is processed into audio information and made playable. The input is the response text, and the output is data in audio file format.

[0626] Step 6:

[0627] The terminal receives native audio from the server and plays it back to the user. At this stage, the user receives an audio response, enabling two-way communication. The input is audio data, and the output is an audio response to the user.

[0628] Step 7:

[0629] When the server performs object recognition using visual data, it analyzes the video data obtained from sensors, identifies the target object, and then provides that information to the user. The input is video data, and the output is information about the recognized object.

[0630] Step 8:

[0631] The server records the user's learning progress and generates progress reports for parents as needed. It also adjusts the knowledge processing system based on the recorded progress to suggest the most suitable learning content for the user. Input is user behavior data, and output is reports and adjusted system parameters.

[0632] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0633] This invention is an educational support system that combines an emotion engine to improve interaction with child users. This system consists of the following elements: user information registration, speech recognition, natural language processing, large-scale language modeling, speech synthesis, emotion recognition, and reinforcement learning.

[0634] When a user first accesses the system, a registration screen appears on their device, prompting them to enter user information. This information is sent to the server, which generates a personalized profile. This profile reflects the user's interests, age, and learning goals, and is used to guide their interaction with the system.

[0635] When a user speaks to the AI ​​agent, the device records the voice and sends the audio data to the server. The server first uses a speech recognition engine to convert the voice into text, and then uses an emotion engine to analyze the emotional state. This emotion analysis information is referenced during the natural language processing stage to generate an appropriate response that takes the child's emotions into account. The generated text response is then converted back into audio data by a speech synthesis engine and sent to the device to be provided to the user.

[0636] For example, if a user speaks in a dejected voice saying, "Today was just bad," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." Because the response changes according to the emotion, it enables more meaningful conversations for children.

[0637] Furthermore, when learning content is requested, the server selects the most suitable learning material based on profile information and emotional state and sends it to the device. As the user progresses through the learning process and the device records its progress, the server stores the child's emotions and learning progress in a database, which is then used for reinforcement learning of the artificial intelligence model.

[0638] Finally, parents can review a progress report that includes their child's learning status and emotional tendencies. This report is generated by the server and presented in an easy-to-understand visual format. This system aims to improve educational effectiveness and support parents by understanding children's emotions and providing appropriate support.

[0639] The following describes the processing flow.

[0640] Step 1:

[0641] The user starts up their device, and the registration screen is displayed. The user enters information such as their name, age, hobbies, and learning goals, and then registers.

[0642] Step 2:

[0643] The terminal sends the entered user information to the server. The server generates an individualized profile based on the received information and stores it in a database.

[0644] Step 3:

[0645] When a user speaks to the AI ​​agent, the device records the audio in real time and sends the audio data to a server.

[0646] Step 4:

[0647] The server processes the audio data through a speech recognition engine and converts it into text. The converted text data is then passed to an emotion engine to analyze the user's emotional state.

[0648] Step 5:

[0649] Based on the emotional information analyzed by the emotion engine, the server uses a natural language processing engine to generate personalized text responses that correspond to the user's emotional state.

[0650] Step 6:

[0651] The generated text response is converted into audio data by a speech synthesis engine and sent to the terminal. The terminal then plays this audio data back to the user to provide the response.

[0652] Step 7:

[0653] When a user requests learning content, the device sends the request to the server. The server refers to profile information and sentiment data, selects appropriate learning content, and sends it.

[0654] Step 8:

[0655] The device displays learning content received from the server and prompts the user to engage in activities. The user's learning progress is recorded by the device.

[0656] Step 9:

[0657] The server periodically receives interaction data and emotion data from the terminals and updates the artificial intelligence model using reinforcement learning algorithms.

[0658] Step 10:

[0659] The server generates and displays reports on the device for parents, including information on the child's learning progress and emotional tendencies. Based on this information, parents can gain a detailed understanding of their child's situation.

[0660] (Example 2)

[0661] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0662] Conventional educational support systems often fail to adequately consider user emotions during interaction, making it difficult to maximize learning effectiveness. Furthermore, there is a lack of easy ways for parents to understand their child's learning progress and emotional changes. Therefore, there is a need for flexible responses that respond to user emotions and appropriate management of progress.

[0663] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0664] In this invention, the server includes means for receiving user information and generating personalized datasets based on that information, means for converting voice input into text data and performing natural language processing on the text data, and means for determining the user's emotional state through sentiment analysis. This makes it possible to provide interactions that are tailored to the user's individual characteristics and emotions, thereby enhancing the effectiveness of learning. In addition, parents can understand their child's learning progress and emotional changes in real time through progress reports.

[0665] "User information" refers to data about individual users, including attribute information such as interests, age, and learning goals.

[0666] A "personalized dataset" is a collection of information optimized for each individual user, based on the user information collected.

[0667] "Voice input" is a method by which users transmit information to a system through voice data, which is then converted into a digital format using speech recognition technology.

[0668] "Character data" refers to linguistic information in text format converted from speech input, and is subject to analysis using natural language processing.

[0669] "Natural language processing" is a technology that enables computers to understand and process human language, and it is used to analyze the meaning of text data and generate responses.

[0670] "Emotion analysis" is the process of determining a user's emotional state from their words and actions, enabling the system to respond in accordance with the user's emotions.

[0671] "Text response" refers to character data generated by natural language processing, which is then converted into audio data as feedback to the user.

[0672] "Audio data" refers to information in audio format generated from text responses using speech synthesis technology, and is provided to the user audibly.

[0673] "Learning materials" are educational content selected based on the user's learning goals and interests, and are used during the learning process.

[0674] "Learning progress" refers to the progress a user has made in learning content, and this data is recorded and analyzed.

[0675] A "database" is a system for storing and efficiently managing information, and is used to store learning progress and user information.

[0676] A "machine learning model" is a computer program that learns on its own based on given data, and contributes to improving the performance of a system.

[0677] "Reinforcement learning" is a process of learning the optimal action through trial and error, and as a form of machine learning model, it is used to improve user interaction.

[0678] A "progress report" is a document that summarizes a user's learning progress and emotional changes, and is provided to parents to help them understand their child's learning situation.

[0679] Modes for carrying out the invention

[0680] This educational support system is designed to make user interaction more effective and is achieved by combining technologies such as speech recognition, natural language processing, sentiment analysis, and speech synthesis.

[0681] Hardware and software configuration:

[0682] 1. Terminal:

[0683] It is a device that allows users to interact using voice and text, and is equipped with a microphone and speaker.

[0684] A standard computer or smart device is used to capture and play back audio data.

[0685] Standard software used includes applications for voice recording and communication applications for data transmission and reception.

[0686] 2. Server:

[0687] At the heart of the system, it handles data processing, AI model training, and profile generation.

[0688] Google Cloud Speech-to-Text is used for speech recognition, converting the user's voice into text data.

[0689] Amazon Comprehend is used for sentiment analysis, analyzing user emotions from text data.

[0690] A large-scale language model (e.g., OpenAI GPT) is used to generate appropriate text responses.

[0691] To convert text responses into speech, use a speech synthesis engine such as Amazon Polly.

[0692] Specific example:

[0693] For example, if a user speaks in a depressed tone, saying "Today was a bad day," the server's emotion analysis engine recognizes the negative emotion. In response to this emotion, the server generates an encouraging response as positive feedback, such as "It's okay. Shall we reflect on your day and talk about what happened?" This response is delivered to the user in real time from the device via speech synthesis.

[0694] Example of a prompt:

[0695] "When a user is expressing negative emotions, what kind of encouraging message should be generated?"

[0696] "Please suggest a method for selecting the most suitable learning content based on the user's age and interests."

[0697] This system aims to support the user's learning process and enhance educational effectiveness through emotionally resonant interactions.

[0698] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0699] Step 1:

[0700] The terminal displays a registration screen to the user. The user enters information such as interests, age, and learning goals. The entered data is sent from the terminal to the server. The terminal converts the user input into digital data and sends data packets to the server using the appropriate communication protocol.

[0701] Step 2:

[0702] The server generates personalized datasets based on the user information it receives. Using user information as input, it analyzes user attribute information through a profile generation algorithm and outputs datasets to enable customized educational support. The server stores this data in its internal database.

[0703] Step 3:

[0704] When a user gives a voice command, the device uses its built-in microphone to capture the audio and sends the audio data to the server. Specific operations are then performed to record the audio data and convert it to a digital audio format.

[0705] Step 4:

[0706] The server uses a speech recognition engine to convert the received audio data into text data. It outputs text data from the audio data as input through the speech recognition process. The server then passes this text data to the next process.

[0707] Step 5:

[0708] The server uses an emotion analysis engine to analyze the emotional state of text. It applies emotion analysis techniques to the input text data and outputs an emotional state such as positive, negative, or neutral. This emotional information is then used to generate the next response.

[0709] Step 6:

[0710] The server uses a natural language processing engine to combine text data with sentiment analysis results to generate responses. Based on the input text and its emotional state, it outputs an appropriate text response. The generated response is then formatted using a generative AI model to ensure contextual accuracy.

[0711] Step 7:

[0712] The server passes the generated text response to the speech synthesis engine, which converts it into audio data. It converts the input text response into an audio file and outputs the specific audio to be delivered to the user. The server then sends this audio data to the terminal.

[0713] Step 8:

[0714] The terminal plays audio data received from the server and provides a response to the user. Audio playback is performed using a speaker, providing feedback to the user as auditory information rather than visual.

[0715] Step 9:

[0716] When a user requests learning content, the server selects the most suitable learning materials based on the user profile and sentiment analysis data. It selects appropriate learning materials from the user information and sentiment data as input and provides them to the terminal as output.

[0717] Step 10:

[0718] The terminal displays the supplied learning content to the user and records the learning progress. It presents learning materials via the display device and saves the user's learning behavior as a log. The server collects the log data and uses it as data for reinforcement learning.

[0719] (Application Example 2)

[0720] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0721] In recent years, educational support systems have been required to incorporate technologies that enhance the effectiveness of children's learning activities. However, conventional systems lack the ability to interact with individual users while considering their emotions. Therefore, a key challenge is to create a system that can appropriately respond to the emotions children exhibit during learning and improve the quality of their learning.

[0722] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0723] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for converting voice input into text data and performing natural language processing on the text, and means for acquiring visual input and analyzing the user's emotions. This enables appropriate learning support and feedback in response to the emotions expressed by the child user.

[0724] "User information" refers to basic personal data obtained from users, and is used for profile generation and the provision of personalized services.

[0725] A "personalized profile" is a collection of information customized based on each user's interests, age, learning goals, etc., and is used to optimize the system's response and learning content.

[0726] "Voice input" refers to information that a user provides to a system via voice, which is then converted into text data using speech recognition technology.

[0727] "Natural language processing" is a technology that enables computers to understand and process human language, and it is a process for analyzing user intent and generating appropriate responses.

[0728] "Visual input" refers to acquiring visual information such as a user's facial expressions using devices like cameras, and this data is useful for analyzing the user's emotions.

[0729] "Means of analyzing emotions" refers to technologies that determine a user's emotional state based on their voice and visual information and provide appropriate feedback.

[0730] "Learning content" refers to educational materials and information provided by the system to achieve specific educational goals, and is selected appropriately according to the user's learning progress.

[0731] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and generates optimal responses and suggestions based on user behavior and emotions.

[0732] This invention is a system that provides educational support while taking the user's emotions into consideration. The system includes smart glasses used by the user, which are equipped with a microphone and a camera. When the user, a child, speaks through the glasses, the voice input is collected by the microphone in the glasses, and visual information is captured by the camera. This data is transmitted from the terminal to a server.

[0733] The server converts the provided audio data into text data using Google Speech Recognition. Visual information is processed using OpenCV to analyze the user's facial expressions and estimate their emotions. Then, based on the text data and emotion information, a natural language processing process using Hugging Face Transformers is performed to generate an appropriate response according to the user's emotions. The generated response is converted into speech using gTTS (Google Text-to-Speech) and fed back to the user through the device.

[0734] For example, if a user speaks in a depressed tone saying, "Today was a bad day," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." This allows for emotionally responsive support for children, improving the quality of their learning.

[0735] The following are examples of prompts to input into a generative AI model:

[0736] Child's comment: "Today was unlucky."

[0737] Detected emotion: Negative

[0738] Required response generation: Words of encouragement and support

[0739] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0740] Step 1:

[0741] The user speaks to the system via voice through smart glasses. This input is collected by a microphone built into the smart glasses. Specifically, the smart glasses capture the voice data and save it as a digital audio file. The input is the user's spoken words, and the output is digital audio data.

[0742] Step 2:

[0743] The device sends this audio data to the server. The server uses Google Speech Recognition to convert the audio data into text format. It analyzes the audio waveform data and converts it into a string of words using a speech recognition algorithm. The input is audio data, and the output is text data.

[0744] Step 3:

[0745] Visual information is collected by a camera, and the terminal sends the video data to a server. The server processes this video data using OpenCV and analyzes the user's facial expressions. Specifically, an image processing algorithm extracts facial feature points and determines the emotional state. The input is video data, and the output is emotion data.

[0746] Step 4:

[0747] The server performs natural language processing based on the acquired text and sentiment data. It uses Hugging Face Transformers to generate appropriate responses corresponding to emotions. The generative AI model analyzes text time-series data and sentiment characteristics, processing the data to output appropriate language responses. The input is text data and sentiment data, and the output is a text-based response.

[0748] Step 5:

[0749] The server converts the generated text response into audio data using gTTS. By converting natural language text into synthesized speech, the output is made easier for the user to understand. The input is a text-based response, and the output is audio data.

[0750] Step 6:

[0751] The terminal receives audio data transmitted from the server and plays it back to the user. The audio is played through the smart glasses' speaker, and the user receives the system's response audibly. The input is audio data, and the output is audio output to the user.

[0752] This series of processes allows users to receive emotionally responsive interactions.

[0753] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0754] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0755] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0756] [Fourth Embodiment]

[0757] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0758] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0759] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0760] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0761] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0762] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0763] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0764] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0765] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0766] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0767] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0768] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0769] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0770] The system of this invention is designed to provide interactive education and communication support to users, namely children and their guardians. This system combines speech recognition, natural language processing, large-scale language models, speech synthesis, and reinforcement learning.

[0771] When a user accesses the system, the terminal requests initial registration and prompts them to enter basic information about their child. This information is sent to the server, which creates an individualized user profile. The profile includes information such as grade level, hobbies, and goals, and all interactions are personalized based on this profile.

[0772] When a child speaks to the AI ​​agent, the device receives the audio and sends the audio data to the server. The server uses a speech recognition engine to convert the audio into text and generates an appropriate response using a large-scale language model. By referring to individualized profile information, it can create a response that is optimal for the child.

[0773] The generated text response is converted into audio data by a speech synthesis engine and played back by the device. This audio response allows the child to continue the conversation with the AI ​​agent. For example, if the child asks, "What should I learn today?", the server will suggest learning content tailored to the child's interests and learning goals.

[0774] Furthermore, learning content is filtered by the server and sent to the device in a difficulty-controlled format. As the user progresses through the learning process using this content, the system records progress information and provides problems tailored to their abilities as needed. This recorded data is periodically analyzed and used for reinforcement learning to improve the AI ​​agent's responses and the learning content.

[0775] Parents can monitor their child's learning progress and interaction history through a dashboard and receive reports from the server as needed, thereby enhancing educational support at home. In this way, the system complements parent-child communication and supports the healthy growth and learning of children.

[0776] The following describes the processing flow.

[0777] Step 1:

[0778] The user accesses the system, and a registration screen appears on their device. The user enters necessary information such as the child's name, age, hobbies, and learning goals.

[0779] Step 2:

[0780] The terminal sends the entered user information to the server. The server receives this information, stores it in a database, and then generates an individualized profile.

[0781] Step 3:

[0782] When a user speaks to the AI ​​agent, the device records the audio and sends the audio data to the server.

[0783] Step 4:

[0784] The server uses a speech recognition engine to convert speech data into text. The converted text is then passed to a large-scale language model to generate an appropriate text response.

[0785] Step 5:

[0786] The server passes the generated text response to the speech synthesis engine, which then generates the audio data. The generated audio data is then sent to the terminal.

[0787] Step 6:

[0788] The device plays back the received audio data and provides a response to the user. The user can then continue conversing with the AI ​​agent.

[0789] Step 7:

[0790] When a user requests access to learning content, the device sends that request to the server.

[0791] Step 8:

[0792] The server selects appropriate learning content based on the child's profile information, adjusts the difficulty level, and sends it to the device.

[0793] Step 9:

[0794] The device displays the received learning content and allows the user to progress through the activities. It also records learning progress data.

[0795] Step 10:

[0796] The server periodically receives interaction data from the terminals and updates the artificial intelligence model by applying reinforcement learning algorithms.

[0797] Step 11:

[0798] The server generates progress reports for parents and displays them on their devices. By reviewing these reports, parents can understand their child's learning progress.

[0799] (Example 1)

[0800] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0801] In today's family environment, there are limited means to efficiently and effectively provide education for children and support for parents. Furthermore, it is difficult to provide education tailored to each child's learning style and progress, leading to a heavy burden on parents. There is also a need for systems that effectively utilize AI technology to improve the quality of learning and deepen parent-child communication.

[0802] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0803] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information; means for converting voice input into text data and performing natural language processing on the text; means for converting the generated text response into voice data and providing it to the user; means for selecting educational content and evaluating the user's learning progress; and means for utilizing recorded data to optimize a machine learning model. This makes it possible to realize a system that provides individualized education and complements support for parents and communication between parents and children.

[0804] "User information" refers to information about a user, including their personal attributes, interests, and learning goals.

[0805] A "personalized profile" is a collection of data that holds attributes and settings specific to each user, based on user information.

[0806] "Voice input" refers to instructions or questions that a user makes to their device using their voice.

[0807] "Text data" refers to string data converted from voice input using speech recognition technology.

[0808] "Natural language processing" is a technology that analyzes text data, understands its intent and meaning, and generates responses.

[0809] A "text response" is character information that represents a reply or reaction generated by natural language processing.

[0810] "Audio data" refers to acoustic signals obtained by converting text responses using speech synthesis technology.

[0811] "Educational content" refers to a collection of learning materials and exercises provided for the purpose of user learning.

[0812] "Learning status" refers to the state of a user's learning progress and level of understanding.

[0813] A "machine learning model" is a software algorithm or system that improves its accuracy by self-improving based on data.

[0814] "Optimization" is the process of adjusting a system or algorithm to achieve a specific goal and extract the maximum effect.

[0815] "Recorded data" refers to information stored as user activity or system operation history.

[0816] This invention is designed to provide users with a personalized educational experience. Specific examples of this system are described below.

[0817] Users first access the system through a device. Available devices include personal computers, smartphones, and tablets. The device displays an initial registration screen where users enter their information. This information includes the child's name, grade level, hobbies, and goals. Once the information is entered and submitted, the device sends it to the server using a secure communication protocol. The server then uses this information to create a user profile in its database.

[0818] Next, when the user speaks to the device, the device receives voice input using its built-in or external microphone. The device sends this voice data to a server. The server uses speech recognition software (e.g., a speech recognition engine) to convert the voice data into text data. The text data is then input into a generative AI model (e.g., a large-scale language model) that performs natural language processing to generate an appropriate text response. This response is personalized by referencing the user profile.

[0819] The generated text response is converted into audio data by speech synthesis software (e.g., a speech synthesis engine) on the server. This audio data is then sent back to the terminal and played back through the terminal's speaker, conveying the message to the user.

[0820] In providing educational content, the server selects the most suitable learning materials based on the user's learning history and profile information. These materials are sent to the device, and the user can use them to progress with their learning. Parents can view progress reports generated by the server and support their child's learning as needed.

[0821] For example, if a user asks the device, "What should I study today?", the server will respond with a voice suggestion based on the user's interests and goals, such as, "Let's learn about space today."

[0822] An example of a prompt might be, "Please suggest some space topics that would interest an 8-year-old child."

[0823] In this way, we can support children's learning by providing beneficial and personalized educational support to users.

[0824] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0825] Step 1:

[0826] The user starts up the terminal and accesses the system. The terminal displays a user information input screen, where the user enters the child's name, grade level, hobbies, and goals. This input information is sent from the user to the terminal and then transmitted from the terminal to the server using a secure protocol. The server receives this information and creates an individualized user profile in its database.

[0827] Step 2:

[0828] When a user speaks into the device, the device receives voice input through its built-in microphone. The received voice is converted into a digital signal and stored in a buffer. This voice data is then sent from the device to the server. The server uses a speech recognition engine to convert the voice data into text data. This converted text data becomes the input for the next process.

[0829] Step 3:

[0830] The server passes the text data obtained through speech recognition to a generative AI model. Here, the AI ​​model creates prompts based on the user's profile information and generates the optimal text response using a large-scale language model. The generated text becomes the output of this process.

[0831] Step 4:

[0832] The text response is input into the server's speech synthesis engine and converted into audio data. This audio data is then sent from the server to the terminal. The terminal plays the received audio data through its speaker, conveying the response to the user.

[0833] Step 5:

[0834] The server analyzes the user's learning history and profile to select the most suitable educational content. The selected content is sent to the device, and the user progresses through their learning using the displayed content. Progress is fed back from the device to the server and recorded.

[0835] Step 6:

[0836] The server analyzes recorded training and interaction data and optimizes the machine learning model based on reinforcement learning. This aims to improve the accuracy of future responses and the quality of educational content.

[0837] (Application Example 1)

[0838] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0839] In educational support systems for children, there is a need for the provision of individualized educational content and the promotion of learning through visual experiences. However, conventional systems have challenges in providing sufficient personalization to meet user interests and in offering interactive experiences that integrate visual and auditory senses. Furthermore, there is a lack of means for parents to accurately understand their child's learning progress and provide appropriate support.

[0840] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0841] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for recognizing objects from visual information and providing information, and means for selecting educational content based on dialogue and presenting it to the child. This enables the provision of individualized educational content and the effective promotion of learning through visual experiences. Furthermore, it creates a system that allows parents to easily grasp progress and support education.

[0842] "User information" refers to basic data about system users and is used to generate individualized profiles.

[0843] A "personalized profile" is a data structure generated based on user information, and serves as a foundation for providing information and services optimized for each user.

[0844] "Voice input" refers to voice data from the user, which is then converted into text data for natural language processing.

[0845] "Natural language processing" is a technology that analyzes input text data and understands its meaning in context.

[0846] "Generated text response" refers to the content of the response to the user obtained as a result of processing by a large-scale language model.

[0847] "Audio data" refers to a data format that enables text responses to be output as speech through speech synthesis.

[0848] "Visual information" refers to image and video data acquired by a system through cameras and sensors, and is used for object recognition.

[0849] "Means for recognizing objects and providing information" refers to technologies that analyze visual information and provide users with information about identified objects.

[0850] "Learning content" refers to educational materials and activities provided by the system for the purpose of educating users.

[0851] "A means of selecting and presenting educational content to children based on dialogue" refers to a function that selects and provides appropriate learning content through communication with users.

[0852] A "progress report" is a report summarizing the user's learning and activity status, and is provided to parents / guardians.

[0853] A "knowledge processing system" is a system that uses artificial intelligence to analyze information and generate responses to users or select content.

[0854] This invention relates to a system for providing interactive education and communication support to child users and their guardians. The system generates an individualized profile based on user information and interacts with the child by combining speech recognition, natural language processing, large-scale language models, and speech synthesis.

[0855] The server receives voice input from the user and converts the speech to text using a speech recognition engine. The text is then analyzed using natural language processing, and an appropriate response is generated using a large-scale language model (e.g., GPT-3). In this process, personalized responses are created by referencing profile information. The generated response text is converted into audio data by a speech synthesis engine and played back on the user's device. Specifically, Amazon Polly and Google Speech-to-Text are used for this purpose.

[0856] The device also acquires visual information through visual sensors and performs object recognition. It generates information about the recognized objects and provides it to the child in an appropriate format. This process uses, for example, a camera and image recognition software.

[0857] As users interact with the system, educational content is automatically selected, and appropriate learning materials are provided. This content selection and delivery is optimized based on the user's learning history and interests. Learning progress is recorded on the server, and progress reports are generated for parents, allowing them to support their child's education based on up-to-date information.

[0858] For example, if a child wearing smart glasses points to a tree in a park and asks, "What kind of tree is this?", the system can analyze the visual information and provide an explanation such as, "This tree is a cherry tree."

[0859] An example of a prompt for a generative AI model is: "A child in the park asked what kind of tree it is. How would you answer through smart glasses?"

[0860] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0861] Step 1:

[0862] The user provides voice input to the device. The device captures this voice data and prepares to send it to the server. At this stage, the input is the user's voice, and the output is data in the form of an audio file.

[0863] Step 2:

[0864] The server receives audio data and converts it into text data using a speech recognition engine. This process analyzes the features of the audio waveform and generates the corresponding text. The input is audio data, and the output is text data.

[0865] Step 3:

[0866] The server performs natural language processing on the generated text data to analyze the user's intent. The input is text data, and the output is context-based analysis results, which are used in the next processing step.

[0867] Step 4:

[0868] Based on the analysis results, the server uses a generative AI model to generate appropriate response text. This process leverages a large-scale language model to create personalized content tailored to the user profile. The input is the analysis results from the previous stage, and the output is the response text.

[0869] Step 5:

[0870] The server uses a speech synthesis engine to convert the response text into audio data. At this stage, the text information is processed into audio information and made playable. The input is the response text, and the output is data in audio file format.

[0871] Step 6:

[0872] The terminal receives native audio from the server and plays it back to the user. At this stage, the user receives an audio response, enabling two-way communication. The input is audio data, and the output is an audio response to the user.

[0873] Step 7:

[0874] When the server performs object recognition using visual data, it analyzes the video data obtained from sensors, identifies the target object, and then provides that information to the user. The input is video data, and the output is information about the recognized object.

[0875] Step 8:

[0876] The server records the user's learning progress and generates progress reports for parents as needed. It also adjusts the knowledge processing system based on the recorded progress to suggest the most suitable learning content for the user. Input is user behavior data, and output is reports and adjusted system parameters.

[0877] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0878] This invention is an educational support system that combines an emotion engine to improve interaction with child users. This system consists of the following elements: user information registration, speech recognition, natural language processing, large-scale language modeling, speech synthesis, emotion recognition, and reinforcement learning.

[0879] When a user first accesses the system, a registration screen appears on their device, prompting them to enter user information. This information is sent to the server, which generates a personalized profile. This profile reflects the user's interests, age, and learning goals, and is used to guide their interaction with the system.

[0880] When a user speaks to the AI ​​agent, the device records the voice and sends the audio data to the server. The server first uses a speech recognition engine to convert the voice into text, and then uses an emotion engine to analyze the emotional state. This emotion analysis information is referenced during the natural language processing stage to generate an appropriate response that takes the child's emotions into account. The generated text response is then converted back into audio data by a speech synthesis engine and sent to the device to be provided to the user.

[0881] For example, if a user speaks in a dejected voice saying, "Today was just bad," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." Because the response changes according to the emotion, it enables more meaningful conversations for children.

[0882] Furthermore, when learning content is requested, the server selects the most suitable learning material based on profile information and emotional state and sends it to the device. As the user progresses through the learning process and the device records its progress, the server stores the child's emotions and learning progress in a database, which is then used for reinforcement learning of the artificial intelligence model.

[0883] Finally, parents can review a progress report that includes their child's learning status and emotional tendencies. This report is generated by the server and presented in an easy-to-understand visual format. This system aims to improve educational effectiveness and support parents by understanding children's emotions and providing appropriate support.

[0884] The following describes the processing flow.

[0885] Step 1:

[0886] The user starts up their device, and the registration screen is displayed. The user enters information such as their name, age, hobbies, and learning goals, and then registers.

[0887] Step 2:

[0888] The terminal sends the entered user information to the server. The server generates an individualized profile based on the received information and stores it in a database.

[0889] Step 3:

[0890] When a user speaks to the AI ​​agent, the device records the audio in real time and sends the audio data to a server.

[0891] Step 4:

[0892] The server processes the audio data through a speech recognition engine and converts it into text. The converted text data is then passed to an emotion engine to analyze the user's emotional state.

[0893] Step 5:

[0894] Based on the emotional information analyzed by the emotion engine, the server uses a natural language processing engine to generate personalized text responses that correspond to the user's emotional state.

[0895] Step 6:

[0896] The generated text response is converted into audio data by a speech synthesis engine and sent to the terminal. The terminal then plays this audio data back to the user to provide the response.

[0897] Step 7:

[0898] When a user requests learning content, the device sends the request to the server. The server refers to profile information and sentiment data, selects appropriate learning content, and sends it.

[0899] Step 8:

[0900] The device displays learning content received from the server and prompts the user to engage in activities. The user's learning progress is recorded by the device.

[0901] Step 9:

[0902] The server periodically receives interaction data and emotion data from the terminals and updates the artificial intelligence model using reinforcement learning algorithms.

[0903] Step 10:

[0904] The server generates and displays reports on the device for parents, including information on the child's learning progress and emotional tendencies. Based on this information, parents can gain a detailed understanding of their child's situation.

[0905] (Example 2)

[0906] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0907] Conventional educational support systems often fail to adequately consider user emotions during interaction, making it difficult to maximize learning effectiveness. Furthermore, there is a lack of easy ways for parents to understand their child's learning progress and emotional changes. Therefore, there is a need for flexible responses that respond to user emotions and appropriate management of progress.

[0908] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0909] In this invention, the server includes means for receiving user information and generating personalized datasets based on that information, means for converting voice input into text data and performing natural language processing on the text data, and means for determining the user's emotional state through sentiment analysis. This makes it possible to provide interactions that are tailored to the user's individual characteristics and emotions, thereby enhancing the effectiveness of learning. In addition, parents can understand their child's learning progress and emotional changes in real time through progress reports.

[0910] "User information" refers to data about individual users, including attribute information such as interests, age, and learning goals.

[0911] A "personalized dataset" is a collection of information optimized for each individual user, based on the user information collected.

[0912] "Voice input" is a method by which users transmit information to a system through voice data, which is then converted into a digital format using speech recognition technology.

[0913] "Character data" refers to linguistic information in text format converted from speech input, and is subject to analysis using natural language processing.

[0914] "Natural language processing" is a technology that enables computers to understand and process human language, and it is used to analyze the meaning of text data and generate responses.

[0915] "Emotion analysis" is the process of determining a user's emotional state from their words and actions, enabling the system to respond in accordance with the user's emotions.

[0916] "Text response" refers to character data generated by natural language processing, which is then converted into audio data as feedback to the user.

[0917] "Audio data" refers to information in audio format generated from text responses using speech synthesis technology, and is provided to the user audibly.

[0918] "Learning materials" are educational content selected based on the user's learning goals and interests, and are used during the learning process.

[0919] "Learning progress" refers to the progress a user has made in learning content, and this data is recorded and analyzed.

[0920] A "database" is a system for storing and efficiently managing information, and is used to store learning progress and user information.

[0921] A "machine learning model" is a computer program that learns on its own based on given data, and contributes to improving the performance of a system.

[0922] "Reinforcement learning" is a process of learning the optimal action through trial and error, and as a form of machine learning model, it is used to improve user interaction.

[0923] A "progress report" is a document that summarizes a user's learning progress and emotional changes, and is provided to parents to help them understand their child's learning situation.

[0924] Modes for carrying out the invention

[0925] This educational support system is designed to make user interaction more effective and is achieved by combining technologies such as speech recognition, natural language processing, sentiment analysis, and speech synthesis.

[0926] Hardware and software configuration:

[0927] 1. Terminal:

[0928] It is a device that allows users to interact using voice and text, and is equipped with a microphone and speaker.

[0929] A standard computer or smart device is used to capture and play back audio data.

[0930] Standard software used includes applications for voice recording and communication applications for data transmission and reception.

[0931] 2. Server:

[0932] At the heart of the system, it handles data processing, AI model training, and profile generation.

[0933] Google Cloud Speech-to-Text is used for speech recognition, converting the user's voice into text data.

[0934] Amazon Comprehend is used for sentiment analysis, analyzing user emotions from text data.

[0935] A large-scale language model (e.g., OpenAI GPT) is used to generate appropriate text responses.

[0936] To convert text responses into speech, use a speech synthesis engine such as Amazon Polly.

[0937] Specific example:

[0938] For example, if a user speaks in a depressed tone, saying "Today was a bad day," the server's emotion analysis engine recognizes the negative emotion. In response to this emotion, the server generates an encouraging response as positive feedback, such as "It's okay. Shall we reflect on your day and talk about what happened?" This response is delivered to the user in real time from the device via speech synthesis.

[0939] Example of a prompt:

[0940] "When a user is expressing negative emotions, what kind of encouraging message should be generated?"

[0941] "Please suggest a method for selecting the most suitable learning content based on the user's age and interests."

[0942] This system aims to support the user's learning process and enhance educational effectiveness through emotionally resonant interactions.

[0943] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0944] Step 1:

[0945] The terminal displays a registration screen to the user. The user enters information such as interests, age, and learning goals. The entered data is sent from the terminal to the server. The terminal converts the user input into digital data and sends data packets to the server using the appropriate communication protocol.

[0946] Step 2:

[0947] The server generates personalized datasets based on the user information it receives. Using user information as input, it analyzes user attribute information through a profile generation algorithm and outputs datasets to enable customized educational support. The server stores this data in its internal database.

[0948] Step 3:

[0949] When a user gives a voice command, the device uses its built-in microphone to capture the audio and sends the audio data to the server. Specific operations are then performed to record the audio data and convert it to a digital audio format.

[0950] Step 4:

[0951] The server uses a speech recognition engine to convert the received audio data into text data. It outputs text data from the audio data as input through the speech recognition process. The server then passes this text data to the next process.

[0952] Step 5:

[0953] The server uses an emotion analysis engine to analyze the emotional state of text. It applies emotion analysis techniques to the input text data and outputs an emotional state such as positive, negative, or neutral. This emotional information is then used to generate the next response.

[0954] Step 6:

[0955] The server uses a natural language processing engine to combine text data with sentiment analysis results to generate responses. Based on the input text and its emotional state, it outputs an appropriate text response. The generated response is then formatted using a generative AI model to ensure contextual accuracy.

[0956] Step 7:

[0957] The server passes the generated text response to the speech synthesis engine, which converts it into audio data. It converts the input text response into an audio file and outputs the specific audio to be delivered to the user. The server then sends this audio data to the terminal.

[0958] Step 8:

[0959] The terminal plays audio data received from the server and provides a response to the user. Audio playback is performed using a speaker, providing feedback to the user as auditory information rather than visual.

[0960] Step 9:

[0961] When a user requests learning content, the server selects the most suitable learning materials based on the user profile and sentiment analysis data. It selects appropriate learning materials from the user information and sentiment data as input and provides them to the terminal as output.

[0962] Step 10:

[0963] The terminal displays the supplied learning content to the user and records the learning progress. It presents learning materials via the display device and saves the user's learning behavior as a log. The server collects the log data and uses it as data for reinforcement learning.

[0964] (Application Example 2)

[0965] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0966] In recent years, educational support systems have been required to incorporate technologies that enhance the effectiveness of children's learning activities. However, conventional systems lack the ability to interact with individual users while considering their emotions. Therefore, a key challenge is to create a system that can appropriately respond to the emotions children exhibit during learning and improve the quality of their learning.

[0967] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0968] In this invention, the server includes means for receiving user information and generating an individualized profile based on that information, means for converting voice input into text data and performing natural language processing on the text, and means for acquiring visual input and analyzing the user's emotions. This enables appropriate learning support and feedback in response to the emotions expressed by the child user.

[0969] "User information" refers to basic personal data obtained from users, and is used for profile generation and the provision of personalized services.

[0970] A "personalized profile" is a collection of information customized based on each user's interests, age, learning goals, etc., and is used to optimize the system's response and learning content.

[0971] "Voice input" refers to information that a user provides to a system via voice, which is then converted into text data using speech recognition technology.

[0972] "Natural language processing" is a technology that enables computers to understand and process human language, and it is a process for analyzing user intent and generating appropriate responses.

[0973] "Visual input" refers to acquiring visual information such as a user's facial expressions using devices like cameras, and this data is useful for analyzing the user's emotions.

[0974] "Means of analyzing emotions" refers to technologies that determine a user's emotional state based on their voice and visual information and provide appropriate feedback.

[0975] "Learning content" refers to educational materials and information provided by the system to achieve specific educational goals, and is selected appropriately according to the user's learning progress.

[0976] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and generates optimal responses and suggestions based on user behavior and emotions.

[0977] This invention is a system that provides educational support while taking the user's emotions into consideration. The system includes smart glasses used by the user, which are equipped with a microphone and a camera. When the user, a child, speaks through the glasses, the voice input is collected by the microphone in the glasses, and visual information is captured by the camera. This data is transmitted from the terminal to a server.

[0978] The server converts the provided audio data into text data using Google Speech Recognition. Visual information is processed using OpenCV to analyze the user's facial expressions and estimate their emotions. Then, based on the text data and emotion information, a natural language processing process using Hugging Face Transformers is performed to generate an appropriate response according to the user's emotions. The generated response is converted into speech using gTTS (Google Text-to-Speech) and fed back to the user through the device.

[0979] For example, if a user speaks in a depressed tone saying, "Today was a bad day," the emotion engine detects the negative emotion, and the server generates an encouraging response such as, "It's okay. Let's reflect on your day and talk about what happened." This allows for emotionally responsive support for children, improving the quality of their learning.

[0980] The following are examples of prompts to input into a generative AI model:

[0981] Child's comment: "Today was unlucky."

[0982] Detected emotion: Negative

[0983] Required response generation: Words of encouragement and support

[0984] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0985] Step 1:

[0986] The user speaks to the system via voice through smart glasses. This input is collected by a microphone built into the smart glasses. Specifically, the smart glasses capture the voice data and save it as a digital audio file. The input is the user's spoken words, and the output is digital audio data.

[0987] Step 2:

[0988] The device sends this audio data to the server. The server uses Google Speech Recognition to convert the audio data into text format. It analyzes the audio waveform data and converts it into a string of words using a speech recognition algorithm. The input is audio data, and the output is text data.

[0989] Step 3:

[0990] Visual information is collected by a camera, and the terminal sends the video data to a server. The server processes this video data using OpenCV and analyzes the user's facial expressions. Specifically, an image processing algorithm extracts facial feature points and determines the emotional state. The input is video data, and the output is emotion data.

[0991] Step 4:

[0992] The server performs natural language processing based on the acquired text and sentiment data. It uses Hugging Face Transformers to generate appropriate responses corresponding to emotions. The generative AI model analyzes text time-series data and sentiment characteristics, processing the data to output appropriate language responses. The input is text data and sentiment data, and the output is a text-based response.

[0993] Step 5:

[0994] The server converts the generated text response into audio data using gTTS. By converting natural language text into synthesized speech, the output is made easier for the user to understand. The input is a text-based response, and the output is audio data.

[0995] Step 6:

[0996] The terminal receives audio data transmitted from the server and plays it back to the user. The audio is played through the smart glasses' speaker, and the user receives the system's response audibly. The input is audio data, and the output is audio output to the user.

[0997] This series of processes allows users to receive emotionally responsive interactions.

[0998] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0999] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1000] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1001] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1002] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1003] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1004] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1005] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1006] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1007] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1008] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1009] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1010] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1011] 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.

[1012] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1013] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1014] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1015] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1016] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1017] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1018] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1019] The following is further disclosed regarding the embodiments described above.

[1020] (Claim 1)

[1021] A means for receiving user information and generating an individualized profile based on that information,

[1022] A means for converting voice input into text data and performing natural language processing on the text,

[1023] A means of converting the generated text response into audio data and providing it to the user,

[1024] A means of generating learning content and recording the user's learning progress,

[1025] A means of adjusting an artificial intelligence model using recorded data,

[1026] A system that includes this.

[1027] (Claim 2)

[1028] The system according to claim 1, characterized by comprising means for generating and displaying progress reports for parents.

[1029] (Claim 3)

[1030] The system according to claim 1, characterized by comprising means for periodically sending user interaction data in batches and using it for reinforcement learning of an artificial intelligence model.

[1031] "Example 1"

[1032] (Claim 1)

[1033] A means for receiving user information and generating an individualized profile based on that information,

[1034] A means for converting voice input into text data and performing natural language processing on the text,

[1035] A means of converting the generated text response into audio data and providing it to the user,

[1036] A means of selecting educational content and evaluating the user's learning progress,

[1037] A means of optimizing a machine learning model using recorded data,

[1038] A means to complement communication between children and their guardians,

[1039] A system that includes this.

[1040] (Claim 2)

[1041] The system according to claim 1, characterized in that it includes means for generating and visualizing progress reports for parents.

[1042] (Claim 3)

[1043] The system according to claim 1, characterized in that it includes means for periodically aggregating user interaction data and using it for reinforcement learning of a machine learning model.

[1044] "Application Example 1"

[1045] (Claim 1)

[1046] A means for receiving user information and generating an individualized profile based on that information,

[1047] A means for converting voice input into text data and performing natural language processing on the text,

[1048] A means of converting the generated text response into audio data and providing it to the user,

[1049] A means of generating learning content and recording the user's learning progress,

[1050] Means for adjusting a knowledge processing system using recorded data,

[1051] A means of recognizing objects from visual information and providing information,

[1052] A means of selecting and presenting educational content to children based on dialogue,

[1053] A system that includes this.

[1054] (Claim 2)

[1055] The system according to claim 1, characterized by comprising means for generating and displaying progress reports for parents.

[1056] (Claim 3)

[1057] The system according to claim 1, characterized in that it includes means for periodically sending user interaction data in batches and using it for reinforcement learning of a knowledge processing system.

[1058] "Example 2 of combining an emotion engine"

[1059] (Claim 1)

[1060] A means for receiving user information and generating an individualized dataset based on that information,

[1061] A means for converting voice input into text data and performing natural language processing on the text data,

[1062] A means of determining a user's emotional state through emotion analysis,

[1063] A means of generating text responses corresponding to the determined emotions, converting them into audio data, and providing them to the user,

[1064] A means of selecting learning materials, recording the user's learning progress, and saving it in a database,

[1065] A method for tuning machine learning models using recorded data and utilizing them for reinforcement learning,

[1066] A system that includes this.

[1067] (Claim 2)

[1068] The system according to claim 1, comprising means for generating and visually displaying progress reports for parents.

[1069] (Claim 3)

[1070] The system according to claim 1, comprising means for periodically aggregating and transmitting user interaction data and using it for reinforcement learning of a machine learning model.

[1071] "Application example 2 when combining with an emotional engine"

[1072] (Claim 1)

[1073] A means for receiving user information and generating an individualized profile based on that information,

[1074] A means for converting voice input into text data and performing natural language processing on the text,

[1075] A means of converting the generated text response into audio data and providing it to the user,

[1076] A means of generating learning content and recording the user's learning progress,

[1077] A means of adjusting an artificial intelligence model using recorded data,

[1078] A means of acquiring visual input and analyzing the user's emotions,

[1079] A means of generating an appropriate response based on analyzed emotional information,

[1080] A system that includes this.

[1081] (Claim 2)

[1082] The system according to claim 1, characterized by comprising means for generating and displaying progress reports for parents.

[1083] (Claim 3)

[1084] The system according to claim 1, characterized by comprising means for periodically sending user interaction data in batches and using it for reinforcement learning of an artificial intelligence model. [Explanation of Symbols]

[1085] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for receiving user information and generating an individualized profile based on that information, A means for converting voice input into text data and performing natural language processing on the text, A means of converting the generated text response into audio data and providing it to the user, A means of generating learning content and recording the user's learning progress, Means for adjusting a knowledge processing system using recorded data, A means of recognizing objects from visual information and providing information, A means of selecting and presenting educational content to children based on dialogue, A system that includes this.

2. The system according to claim 1, characterized in that it includes means for generating and displaying progress reports for parents.

3. The system according to claim 1, characterized in that it includes means for periodically sending user interaction data in batches and using it for reinforcement learning of a knowledge processing system.

Images