System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving and storing evaluation criteria points for each subject or lesson; means for inputting voice data collected from students during a lesson into a voice recognition system and converting the voice data into text data; means for scoring the converted text data based on the evaluation criteria; means for storing the scored results and speech content in a database; and means for notifying teachers of specific suggestions generated and suggestions for future growth.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Currently, elementary and junior high schools face the challenge of fairly and efficiently evaluating students' comments and behavior during class. Because teachers conduct subjective evaluations, the evaluations lack fairness, and the evaluation process itself is time-consuming and labor-intensive. Furthermore, teacher shortages and busy schedules make it even more difficult to properly conduct such evaluations. Therefore, there is a need for a method to easily create specific lesson plans to effectively support student growth. [Means for solving the problem]

[0005] The present invention is a system that includes a means for receiving and storing evaluation criteria points for each subject and lesson, a means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, a means for scoring the converted text data based on the evaluation criteria, a means for storing the scored results and the content of the students' comments in a database, and a means for notifying the teacher of the generated specific lesson plans and suggestions for future growth. The system also includes a means for inputting the evaluation criteria points into an AI model for learning, a means for using a generation AI to analyze the converted text data, a means for correlating the scoring results of the comments for each specific student and recording them in a database, and a means for using the generation AI to create individual lesson plans and advice based on the collected score data and the content of the students' comments, thereby achieving fair and efficient student evaluation and the creation of specific lesson plans.

[0006] "Evaluation criteria points" are the criteria points used to evaluate students' comments and attitudes in a class or a specific subject.

[0007] A "voice recognition system" is a system for converting collected voice data into text data.

[0008] "Scoring" is the process of converting the converted text data into numerical values ​​based on evaluation criteria and evaluating them as scores.

[0009] "Database" refers to an information management system for storing and managing evaluation results and related data.

[0010] "Generative AI" is an artificial intelligence technology that analyzes students' comments and attitudes and automatically generates lesson plans and advice based on pre-learned data.

[0011] A "lesson plan" is a specific proposal for future teaching content and direction, created based on the evaluation results of each student's comments and attitude.

[0012] A "subject" refers to a specific academic field in school education, and is classified as Japanese, mathematics, science, etc.

[0013] A "class" is a part of educational activities in which a teacher teaches students content related to a subject.

[0014] "Text data" refers to data obtained by converting voice data into character information.

[0015] "Student" refers to a learner who attends an educational institution and takes classes.

[0016] "Faculty" refers to teachers and lecturers who teach at educational institutions. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention is a system that combines a speech recognition system and a generative AI to automatically evaluate students' comments and behaviors during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0039] System Configuration

[0040] The system is broadly composed of the following elements:

[0041] 1. Server

[0042] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0043] 3. Web interface accessed by users (teachers)

[0044] Entering and learning evaluation criteria points

[0045] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "coherence of speech," "activeness," etc. These criteria are sent to the server via the terminal and stored in a database.

[0046] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate students' comments in class based on these criteria.

[0047] Collection of audio data during class

[0048] During class, students are placed on their desks with microphones attached to their devices, which capture what they say in real time. The collected audio data is then sent to a server.

[0049] Text conversion using a speech recognition system

[0050] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[0051] Scoring

[0052] The server uses a generative AI to analyze the converted text data and assigns a score based on pre-trained evaluation criteria. Each comment is assigned a score such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0053] Scores and speech records

[0054] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0055] Lesson plan generation and notification

[0056] The server analyzes the collected score data and the content of the comments, and uses the generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions more consistently."

[0057] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[0058] Specific examples

[0059] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0060] This allows teachers to efficiently and fairly evaluate students' comments and attitudes, and with the support of generative AI, they can also strengthen individual instruction.

[0061] The processing flow will be explained below.

[0062] Step 1:

[0063] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," and "activeness."

[0064] Step 2:

[0065] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[0066] Step 3:

[0067] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[0068] Step 4:

[0069] When class begins, a terminal (such as a device with a microphone installed on a student's desk) records what the student says in real time and collects audio data.

[0070] Step 5:

[0071] The collected voice data is sent from the device to a server, which then inputs the received voice data into a voice recognition system and converts the voice data into text data.

[0072] Step 6:

[0073] The server inputs the converted text data into the AI ​​generator, which then scores each comment based on the evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0074] Step 7:

[0075] The server stores the scored results in a database along with the comment content and student ID, thereby recording which student made what comment and how much they were evaluated.

[0076] Step 8:

[0077] Based on the collected score data and the content of the comments, the server uses generative AI to create lesson plans and suggestions for future development for each student. For example, in response to a student's comments that lack consistency, the server generates advice such as, "It would be effective to practice expressing your opinions more consistently."

[0078] Step 9:

[0079] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[0080] Step 10:

[0081] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[0082] Example 1

[0083] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] Conventional educational systems lack the means to evaluate students' comments and behaviors in real time during class and generate lesson plans and suggestions for growth. This makes it difficult for teachers to provide appropriate feedback to each student, and it is difficult to promote student growth efficiently.

[0085] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0086] In this invention, the server includes a means for receiving and storing evaluation criteria points for each subject and lesson, a means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, a means for scoring the converted text data based on the evaluation criteria, a means for saving the scored results and the speech content in a database, a means for notifying teachers of the generated specific lesson plans and suggestions for future growth, a terminal for collecting voice data, and a means for transmitting the data to the server. This allows teachers to efficiently and fairly evaluate students' comments and attitudes and to strengthen individual instruction with the support of the generative AI model.

[0087] "Evaluation criteria points for each subject and class" are the standards used to evaluate students' comments and attitudes, and are evaluation criteria set according to the content of the subject or class.

[0088] A "voice recognition system" is a system for converting voice data collected during classes into text data, and is a technology that analyzes voice and outputs it as text information.

[0089] "Text data" refers to text information of voice data converted using a voice recognition system.

[0090] The "means for scoring based on evaluation criteria" is a method for evaluating the converted text data based on evaluation criteria set in advance and assigning a score.

[0091] "Generative AI" is an artificial intelligence technology that generates appropriate responses and results for specific inputs based on pre-learned data.

[0092] The "database" is an information system for storing and managing information such as evaluation results and scored comments.

[0093] A "lesson plan" is a plan created using generative AI that includes specific teaching guidelines and advice to promote each student's future growth.

[0094] "Suggestions for future growth" are specific advice and improvement plans for each student on how to promote their growth in the future.

[0095] A "terminal" is a device equipped with a microphone that is installed to collect student comments during class and is responsible for transmitting audio data to a server.

[0096] MODE FOR CARRYING OUT THE INVENTION

[0097] This invention is a system that combines a speech recognition system and a generative AI model to automatically evaluate students' comments and behavior during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0098] System Configuration

[0099] The system is broadly composed of the following elements:

[0100] 1. Server

[0101] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0102] 3. Web interface accessed by users (teachers)

[0103] Entering and learning evaluation criteria points

[0104] Users (teachers) input evaluation criteria points for each subject and class through a dedicated web interface. These evaluation criteria points include "responsiveness to lesson content," "coherence of speech," and "activeness." These evaluation criteria points are sent to the server via the terminal and stored in a database. The server then inputs the received evaluation criteria points into the generative AI model, and the AI ​​learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments during class based on these criteria.

[0105] Collection of audio data during class

[0106] During class, students are placed on desks equipped with microphones that capture their speech in real time. The collected audio data is temporarily stored on the devices and then sent to a server.

[0107] Text conversion using a speech recognition system

[0108] The server inputs the received voice data into a speech recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data as, "I think the language in this poem is beautiful."

[0109] Scoring

[0110] The server uses a generative AI (e.g., GPT-4 model) to analyze the converted text data and assign scores based on pre-trained evaluation criteria. Each comment is assigned a score, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0111] Scores and speech records

[0112] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0113] Lesson plan generation and notification

[0114] The server analyzes the collected score data and comments, and uses a generative AI to create lesson plans and suggestions for development for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective for you to practice expressing your opinions more consistently." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[0115] Specific examples

[0116] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these scores are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0117] Example prompts

[0118] Below are some example prompts to input to a generative AI model:

[0119] In your Japanese class, set evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student's attitude positive?". Use a generative AI model to analyze students' comments and generate scores and advice.

[0120] Through these steps, the system enables teachers to efficiently and fairly evaluate students' comments and attitudes, and enhance individual instruction with the assistance of generative AI models.

[0121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0122] Step 1: Enter and save your evaluation criteria points

[0123] The user (teacher) enters the evaluation criteria points for each subject and class (e.g., "responsiveness to class content," "coherence of speech," "proactiveness") through a dedicated web interface. Specifically, the teacher selects the evaluation criteria on the web interface and clicks the "Submit" button.

[0124] Input: Teacher-entered criterion points

[0125] Data processing: Organize the entered evaluation criteria points according to the format

[0126] Output: Data with organized evaluation criteria points

[0127] Specific operation: When the teacher enters the evaluation criteria and presses the submit button, the data is sent to the server and stored in the database.

[0128] Step 2: Collecting audio data

[0129] During class, the terminal (a device with a microphone installed on the student's desk) captures what the student says in real time and temporarily stores it. The collected audio data is then sent from the terminal to a server.

[0130] Input: Audio data captured during class

[0131] Data processing: After temporary storage, organize according to the format

[0132] Output: Organized audio data

[0133] Specific operation: When a student speaks, the device captures the audio and sends the audio data to the server.

[0134] Step 3: Convert audio data to text

[0135] The server inputs the received voice data into a voice recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data.

[0136] Input: Audio data sent to the server

[0137] Data processing: Input the voice data into a voice recognition system and convert it into text data.

[0138] Output: Generated text data

[0139] Specific operation: Sends voice data to a voice recognition system and receives textual results.

[0140] Step 4: Analyze and score the text data

[0141] The server inputs the generated text data into a generative AI model (e.g., GPT-4) for analysis. The generative AI then scores each utterance based on pre-trained evaluation criteria.

[0142] Input: Generated text data

[0143] Data processing: Analyze text data and assign scores based on evaluation criteria

[0144] Output: Scored evaluation results

[0145] Specific operation: Text data is input into the generative AI model, and a score is obtained as the analysis result.

[0146] Step 5: Record your scores and comments

[0147] The server stores the scored results in a database, which includes the student ID, the content of the comment, and the score for each criterion.

[0148] Input: Scored evaluation results

[0149] Data processing: Organize and store the scored results in a database

[0150] Output: Saved evaluation results and comments

[0151] Specific operation: After generating the scoring results, they are immediately written to the database.

[0152] Step 6: Generate and notify lesson plans

[0153] The server inputs the collected score data and comments into a generative AI model to create lesson plans and development suggestions for each student. The generated lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the evaluation results and lesson plans through a web interface.

[0154] Input: Score data and comments

[0155] Data processing: Input into generative AI models to generate lesson plans and development suggestions

[0156] Output: Generated lesson plans and suggestions for development

[0157] Specific operation: When a teacher logs in, a notification of a new lesson plan will be displayed and the contents can be viewed on the details page.

[0158] Through these steps, the system can efficiently evaluate students' comments and behavior during class, and with the support of the generative AI model, provide individual instruction efficiently and effectively.

[0159] (Application example 1)

[0160] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0161] Conventional educational systems have struggled to efficiently and fairly evaluate students' comments and behavior during class. Furthermore, there was no way to automatically generate individual lesson plans or suggestions for improvement, placing a heavy burden on teachers. Furthermore, there was no system in place to evaluate passenger conversations in autonomous vehicles and make suggestions for improving driving conditions, making improving the riding experience a challenge.

[0162] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0163] In this invention, the server includes: means for receiving and storing evaluation criteria points for each subject or lesson; means for inputting student voice data into a speech recognition system and converting it into text data; means for scoring the converted text data based on the evaluation criteria; means for storing the scored results and the speech content in a database; means for notifying the teacher of the generated lesson plan and suggestions for future development; means for evaluating in-car speech data and analyzing the quality of conversation and the degree of agreement between topics; and means for collecting passenger feedback and making suggestions for improving driving conditions. This makes it possible to efficiently and fairly evaluate students' comments and behavior during class and automatically generate individual lesson plans and suggestions for development. It also evaluates passenger conversations and makes suggestions for improving driving conditions in an autonomous vehicle, improving the riding experience.

[0164] "Evaluation criteria points" refer to the evaluation criteria set for each class or subject, and are indicators used to score students' comments and attitudes.

[0165] A "speech recognition system" refers to a technology that converts voice data into text data, and is a system used to accurately transcribe collected speech.

[0166] "Text data" refers to character string data converted by a speech recognition system and is used to record students' utterances as text.

[0167] "Scoring" refers to the process of numerically evaluating the converted text data based on evaluation criteria points.

[0168] "Database" refers to the information system for storing and managing the scored results and comments.

[0169] "Generative AI" refers to artificial intelligence that generates and analyzes text based on large datasets, and is used to automate evaluations and recommendations.

[0170] A "lesson plan" refers to a specific educational strategy to support the individual growth of each student, and is automatically generated by generative AI.

[0171] "Feedback" refers to the opinions and reactions that a system collects from users, and is information used to evaluate and improve the system.

[0172] "Driving situation improvement proposals" refer to specific proposals for improving driving quality based on data collected within an autonomous vehicle.

[0173] "Interface" refers to the means by which users access and operate the system, and is used by teachers to check assessment results and lesson plans.

[0174] "Conversation quality" refers to an indicator that evaluates how meaningful and constructive the conversation is.

[0175] "Topic alignment" refers to an indicator that evaluates how closely a conversation matches the set theme or purpose.

[0176] This invention is a system that can be used in both the fields of educational evaluation and autonomous driving. Specifically, it manages evaluation criteria points for each subject and lesson, automatically evaluates students' comments and behavior during class, and uses generative AI to provide teachers with lesson plans and suggestions for growth. It is also a system that evaluates passenger conversations and makes suggestions for improving driving conditions inside autonomous vehicles.

[0177] System Configuration

[0178] The system consists of the following elements:

[0179] 1. Server: The central hardware that processes, stores, and analyzes data.

[0180] 2. Devices: Devices with microphones installed on students' desks, microphones in self-driving cars, etc.

[0181] 3. Users: School teachers and passengers in self-driving cars.

[0182] Program Overview

[0183] 1. Collection of audio data

[0184] A microphone installed on the device captures what students say in class and conversations inside the self-driving car in real time, and sends the audio data to a server.

[0185] 2. Speech Recognition and Text Conversion

[0186] The server uses a speech recognition system (for example, the SpeechRecognition library) to convert the received voice data into text data, which records what students say and what passengers say as text.

[0187] 3. Scoring and Evaluation

[0188] The server analyzes the converted text data and scores it based on pre-trained evaluation criteria. This process uses a generative AI model (e.g., GPT-3.5-turbo) to evaluate the student's comments based on criteria such as "responsiveness to the lesson content," "coherence of speech," and "activeness." Additionally, in the self-driving car, the server evaluates the quality of the conversation and the degree of topic agreement.

[0189] 4. Saving to the database

[0190] The scored results and comments are stored in a database. The database includes student IDs, comments, and evaluation scores based on each evaluation criterion. This allows for the management of evaluation history.

[0191] 5. Generating lesson plans and suggestions

[0192] The server uses AI to create individual lesson plans and development suggestions for each student based on the collected score data and speech content. For example, it can generate advice such as, "It would be effective to practice expressing your opinions consistently." Then, within the self-driving car, it generates driving improvement suggestions based on feedback on the driving situation.

[0193] 6. Notification of Results

[0194] The final lesson plans and driving improvement suggestions are notified to users (teachers and passengers), who can check the evaluation results and suggestions via a dedicated web interface or smartphone app.

[0195] Specific examples

[0196] For example, if a student says, "I think the language in this poem is beautiful" during a Japanese class, the device's microphone captures the speech and sends it to the server. The server then uses a speech recognition system to convert this speech into text data: "I think the language in this poem is beautiful." The generative AI model then scores this text as "Readiness to respond to lesson content: 5 points," "Consistency of speech: 4 points," and "Positiveness: 3 points," and stores the results in a database.

[0197] Prompt Sentence Examples

[0198] "Analyze the following text and rate the quality and consistency of the conversation: 'The scenery is beautiful. Let's take a photo at the next rest stop.'"

[0199] In this way, the present invention provides a specific form for efficiently collecting, analyzing, evaluating, and proposing improvements to data in both educational evaluation and automated driving.

[0200] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0201] Step 1: Collecting audio data

[0202] The device captures students' speech and conversations inside the self-driving car in real time. A microphone installed on the device picks up the audio and sends the audio data to the server. The input is "audio data" and the output is "audio data sent to the server."

[0203] Step 2: Speech recognition and text conversion

[0204] The server uses the SpeechRecognition library to convert the received voice data into text data. The voice data is input into the voice recognition system and the converted text data is obtained. The input is "voice data" and the output is "text data." Specifically, the voice recognition system analyzes the voice sample and generates a corresponding string of characters.

[0205] Step 3: Scoring based on the evaluation criteria

[0206] The server then scores the converted text data based on pre-trained evaluation criteria. Using a generative AI model (e.g., GPT-3.5-turbo), it assigns a score to the given text based on criteria such as "responsiveness to lesson content," "coherence of speech," and "activeness." The input is "text data" and "evaluation criteria points," and the output is "scored data."

[0207] Step 4: Saving to the Database

[0208] The server saves the scoring results and the comments together in a database. The input is the "score data" and "text data," and the output is the "saved data." Specifically, a new record is added to the database, and the record contains the student ID, comments, and evaluation score.

[0209] Step 5: Generate lesson plans and suggestions

[0210] The server uses generative AI to create individual lesson plans and suggestions for development based on the collected score data and utterances. The inputs are "score data" and "text data," and the output is a "generated lesson plan" and "proposal." Specifically, the server provides prompts to the AI ​​model to generate appropriate advice. For example, "Analyze the following text and generate a lesson plan: What the student uttered."

[0211] Step 6: Notification of results

[0212] The user (teacher) checks the generated lesson plans and proposals through a dedicated web interface or smartphone app. The server sends the evaluation results and proposals to the notification system. The inputs are the "generated lesson plans" and "proposals," and the output is the "notified results." Specifically, the results for each student are displayed on the dashboard of the web interface.

[0213] These steps will enable educational assessment and a conversational assessment and feedback system within an autonomous vehicle.

[0214] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0215] This invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and even emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0216] System Configuration

[0217] The system is broadly composed of the following elements:

[0218] 1. Server

[0219] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0220] 3. Web interface accessed by users (teachers)

[0221] 4. Emotion Engine

[0222] Entering and learning evaluation criteria points

[0223] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the terminal and stored in a database.

[0224] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these criteria. Through this learning process, the AI ​​model can evaluate students' comments and emotions during class based on these criteria.

[0225] Collection of speech and emotion data during class

[0226] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device uses an emotion engine to collect emotional data from the student's voice and facial expressions.

[0227] Speech recognition system for text conversion and emotion recognition

[0228] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[0229] At the same time, the emotion engine analyzes emotional information from voice and students' facial expressions, generating data such as "positive emotions are expressed when speaking."

[0230] Scoring

[0231] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each comment and the emotion it conveys is given a score, such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points."

[0232] Recording scores, comments, and emotions

[0233] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0234] Lesson plan generation and notification

[0235] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend that you provide instruction to increase self-esteem in order to increase positive comments."

[0236] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[0237] Specific examples

[0238] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0239] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[0240] The processing flow will be explained below.

[0241] Step 1:

[0242] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," "activeness," "emotional responsiveness," "emotional consistency," and "emotional activeness."

[0243] Step 2:

[0244] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[0245] Step 3:

[0246] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[0247] Step 4:

[0248] When the lesson begins, the device (such as a microphone-equipped device installed on the student's desk) records what the student says in real time and collects audio data. At the same time, the device uses a camera to capture the student's facial expressions and collect emotional information in real time.

[0249] Step 5:

[0250] The device sends the collected voice data and emotion data to a server, which inputs the voice data into a voice recognition system and converts it into text data.

[0251] Step 6:

[0252] The emotion engine analyzes the emotion data sent to the server and generates emotional information about the student's speech. For example, it recognizes emotions such as "positive" or "negative" from the tone of voice and facial expressions.

[0253] Step 7:

[0254] The server inputs the converted text data and the recognized emotional information into the AI ​​generator, which then scores each utterance based on the evaluation criteria. For example, the evaluation scores could be "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "emotional responsiveness: 4 points," "emotional consistency: 5 points," and "emotional activeness: 3 points."

[0255] Step 8:

[0256] The server stores the scored results in a database along with the comments and student IDs, thereby recording which students made what comments and how they were evaluated.

[0257] Step 9:

[0258] The server uses generative AI to create lesson plans and suggestions for future growth for each student based on the collected score data, comment content, and emotional data. For example, if the comments lack consistency or the emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing consistent opinions. We recommend providing instruction to increase self-esteem in order to increase positive comments."

[0259] Step 10:

[0260] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[0261] Step 11:

[0262] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[0263] Example 2

[0264] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0265] Conventional classroom evaluation systems evaluate students based on subjectivity, which creates issues with the fairness and efficiency of evaluation. In addition, it is difficult to evaluate students based on their emotions and create individual lesson plans, making it difficult to support students' overall growth.

[0266] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for collecting and analyzing students' emotional data, means for scoring the converted text data and emotional data based on the evaluation criteria, means for saving the scored results, the content of the comments, and the emotional data in a database, and means for notifying the teacher of the generated specific lesson plan and suggestions for future growth. This makes it possible to fairly and efficiently evaluate students' comments, attitudes, and even emotions during lessons and provide individual lesson plans.

[0267] "Criteria points" are specific criteria established to assess a student's performance for a particular subject or lesson.

[0268] A "voice recognition system" is a technical system that analyzes input voice data and converts it into corresponding text data.

[0269] "Emotional data" is data that indicates the emotional state of students analyzed from their comments and facial expressions, and includes types of emotions such as positive and negative.

[0270] "Generative AI" refers to artificial intelligence technology that automatically generates evaluations and suggestions based on collected data.

[0271] "Scoring" is the process of numerically evaluating students' comments, attitudes, emotions, etc. based on evaluation criteria.

[0272] A "database" is a system for systematically storing and managing collected data.

[0273] A "lesson plan" is a document that outlines specific plans and policies for teachers to effectively teach students.

[0274] "Notifying" refers to the action of informing teachers of the generated lesson plans and evaluation results.

[0275] The present invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the present invention are described below.

[0276] System Configuration

[0277] The system is broadly composed of the following elements:

[0278] 1. Server

[0279] 2. Terminal (device with microphone installed on student's desk)

[0280] 3. Web interface accessed by users (teachers)

[0281] 4. Emotion engine (e.g., Microsoft Azure Emotion API)

[0282] Entering and learning evaluation criteria points

[0283] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the device and stored in a database. The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments and emotions during class based on these criteria.

[0284] Collection of speech and emotion data during class

[0285] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device also uses an emotion engine to collect emotional data from the student's voice and facial expressions. This is done using a microphone device and a facial recognition camera.

[0286] Speech recognition system for text conversion and emotion recognition

[0287] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful." At the same time, the emotion engine analyzes emotional information from the voice and the student's facial expressions, generating data such as "positive emotion is expressed in the statement."

[0288] Scoring

[0289] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each utterance and its associated emotion are given scores such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points." The prompt is as follows:

[0290] "Based on the student's comments and emotional data, please calculate this student's grade for the lesson. The evaluation criteria are as follows: responsiveness to lesson content, coherence of speech, assertiveness, emotional responsiveness, emotional coherence, and emotional assertiveness."

[0291] Recording scores, comments, and emotions

[0292] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0293] Lesson plan generation and notification

[0294] The server analyzes the collected score data, comment content, and emotional data, and uses a generative AI to create lesson plans and development suggestions for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend providing instruction to increase self-esteem in order to increase positive comments." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[0295] Specific examples

[0296] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0297] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[0298] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0299] Step 1: Enter and save your evaluation criteria points

[0300] Users (teachers) access a dedicated web interface and enter evaluation criteria points for each subject and class.

[0301] Specific actions: Open a web browser, enter your username and password in the login form to log in, enter your evaluation criteria points in the input form, and click the submit button.

[0302] Input: Subject, lesson, evaluation criteria points (e.g., "Readiness to follow lesson content: 5", "Consistency of speech: 4", "Proactiveness: 3")

[0303] Data processing: Send the evaluation criteria points to the server.

[0304] Output: The server stores the metric points in the database.

[0305] Step 2: Learning the evaluation criteria points

[0306] The server inputs the received evaluation criteria points into the AI ​​model, which then learns from them.

[0307] How it works: The server retrieves the evaluation criteria points from the database and passes the data to the AI ​​model, which then performs the learning process based on the evaluation criteria.

[0308] Input: Evaluation criteria point data

[0309] Data calculation: The AI ​​model analyzes the evaluation criteria points and learns.

[0310] Output: The AI ​​model obtains the ability to make an evaluation based on the evaluation criteria.

[0311] Step 3: Collecting audio and emotion data during class

[0312] The device captures students' comments in real time during class, collecting audio data, and simultaneously collects students' emotional data using an emotion engine.

[0313] How it works: The device's built-in microphone records speech in real time, and the device uses an emotion engine to analyze emotions from the voice data.

[0314] Input: Student speech and facial expression data

[0315] Data processing: Record audio data and perform sentiment analysis.

[0316] Output: Voice data and emotion data are obtained.

[0317] Step 4: Text conversion by speech recognition system

[0318] The server inputs the received voice data into a voice recognition system and converts it into text data.

[0319] Specific operation: The server sends the audio file to the speech recognition API and receives text data as the API response.

[0320] Input: Audio data file

[0321] Data processing: The voice recognition system analyzes the voice data and converts it into text data.

[0322] Output: Text data (e.g., "I think the language in this poem is beautiful.")

[0323] Step 5: Analyze the sentiment data

[0324] The server receives the emotion data analyzed by the emotion engine.

[0325] Specific operation: The emotion engine analyzes voice and facial expression data and sends the emotion data to the server.

[0326] Input: Voice data, facial expression data

[0327] Data processing: Emotion analysis using an emotion engine

[0328] Output: Emotion data (e.g. "positive")

[0329] Step 6: Scoring

[0330] The server uses generative AI to score the converted text data and emotional data based on evaluation criteria.

[0331] Specific operation: The server sends text data and emotion data to the generative AI model and obtains an evaluation score from the model.

[0332] Input: Text data, emotion data, evaluation criteria points

[0333] Data calculation: The generating AI performs the scoring.

[0334] Output: Score evaluation (e.g., "Readiness to content: 5 points," "Spoken consistency: 4 points," "Positiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," "Emotional positiveness: 3 points")

[0335] Step 7: Record your scores, comments, and feelings

[0336] The server stores the scored results in a database.

[0337] Specific behavior: The server executes an SQL query to insert the evaluation result into the database.

[0338] Input: Scored evaluation, student ID, speech content, emotion data

[0339] Data processing: Processing to save data in the database.

[0340] Output: Recording to database completed

[0341] Step 8: Generate and notify lesson plans

[0342] The server analyzes the collected score data, speech content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth, which are then notified to the user.

[0343] Specific operations: The server analyzes the score data, sends prompts to the generative AI model, obtains the generated lesson plans and suggestions, and sends a request to notify the user of the obtained lesson plans and suggestions.

[0344] Input: Score data, speech content, emotion data

[0345] Data calculation: Generative AI generates lesson plans and suggestions.

[0346] Output: Lesson plans, suggestions, and user notifications

[0347] Through these processing steps, the present invention realizes a system that can effectively and fairly evaluate students' comments and emotions and provide optimal teaching plans.

[0348] (Application example 2)

[0349] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0350] Traditional methods for evaluating students' comments and behavior during class rely on the teacher's subjective judgment and lack objectivity and efficiency. Furthermore, evaluations of store staff's customer service are often subjective, making it difficult to provide appropriate guidance for improving staff skills and training. There is a need to solve these problems and provide fair and efficient evaluations and lesson plans.

[0351] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for scoring the converted text data based on the evaluation criteria, means for saving the scored results and the content of the comments in a database, means for notifying the teacher of the generated specific lesson plans and suggestions for future development, means for applying the voice recognition system to the customer service of employees in a physical store and converting the comments made during customer service into text data, means for analyzing the emotions of employees and customers using an emotion recognition system, means for evaluating the customer service response based on the content of the comments and emotions, and means for generating and notifying points for improvement and training suggestions from the evaluation results. This enables fair and efficient evaluation and the generation and notification of specific lesson plans and training suggestions.

[0352] "Evaluation criteria points for each subject and class" are the evaluation criteria points set for each subject and class, and refer to specific evaluation items and standards.

[0353] "Audio data" refers to collected audio information from students and employees, including speech during classes and in physical stores.

[0354] A "voice recognition system" is a system that converts collected voice data into text data.

[0355] "Text data" is character information converted from voice data by a voice recognition system.

[0356] "Evaluation criteria" refers to the items and indicators that serve as the basis for scoring based on text data.

[0357] "Scoring" is the process of quantitatively evaluating text data based on evaluation criteria and expressing it as a number.

[0358] A "database" is a data storage system for centrally storing and managing information such as scored results and comment content.

[0359] "Generative AI" refers to an artificial intelligence model that generates specific advice, such as lesson plans and training suggestions, based on collected data.

[0360] An "emotion recognition system" is a system that analyzes emotions from voice and facial expressions and generates the results as data.

[0361] "Customer service" refers to the contact and service provided by employees to customers in physical stores.

[0362] "Training proposals" are proposals based on the evaluation results for specific improvements and practice methods to improve and develop the skills of employees or students.

[0363] This invention is a system that combines a voice recognition system, generative AI, and an emotion recognition system to efficiently and fairly evaluate the comments and attitudes of students and employees during classes or in physical stores, and provide specific lesson plans and training suggestions.

[0364] System Configuration

[0365] The system roughly consists of the following elements:

[0366] 1. Server

[0367] 2. Terminals (such as devices with microphones installed on students' or employees' desks or in physical stores)

[0368] 3. Web interface accessed by users (teachers and store managers)

[0369] 4. Emotion Recognition System

[0370] Entering and learning evaluation criteria points

[0371] Users (teachers and store managers) use a dedicated web interface to input evaluation criteria points for each lesson or customer service. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "proactiveness," as well as "emotional responsiveness," "emotional consistency," and "emotional proactiveness." These criteria are sent to the server via the terminal and stored in a database.

[0372] The server inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate the comments and emotions of students in class or employees in a physical store based on these criteria.

[0373] Collecting speech and emotion data during classes and in-store

[0374] During classes or in brick-and-mortar stores, the devices capture students' and employees' speech in real time. The collected voice data is sent from the devices to a server. At the same time, the devices also use an emotion recognition system to collect emotional data from voice and facial expressions.

[0375] Speech recognition system for text conversion and emotion recognition

[0376] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," or an employee says, "Hello, customer. Please take your time and enjoy your shopping," the voice is converted into text data. At the same time, the emotion recognition system analyzes emotional information from the voice and facial expressions and generates data such as, "Positive emotions are expressed in the statement."

[0377] Scoring

[0378] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "activeness: 3 points" for each comment and its associated emotion.

[0379] Recording scores, comments, and emotions

[0380] The scoring results are stored in a database on the server. This includes student and employee IDs, comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students and employees made what comments and how they were evaluated.

[0381] Generation and notification of lesson plans and training proposals

[0382] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans for each student and employee training suggestions. For example, if a student's comments lack consistency, the generative AI might advise, "It would be effective to practice expressing your opinions consistently." If an employee's emotions are negative, the AI ​​might suggest, "We recommend training to increase self-esteem in order to increase positive expressions."

[0383] The created lesson plans and suggestions are notified to the user (teacher or store manager), who can then check the evaluation results and suggestions for the relevant student or employee through a web interface.

[0384] Specific examples

[0385] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech consistent?", "Is the student's attitude positive?", and also "Emotional readiness," "Emotional consistency," and "Emotional positivity." When a student speaks during class, a device equipped with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the comment is recognized as "positive."

[0386] In a physical store, if a store clerk says, "Hello, customer. Please take your time and enjoy your shopping," a microphone-equipped device captures the voice and sends it to a server. The voice data is converted into text, and the emotion expressed at the time of the statement is recognized as "positive."

[0387] The generative AI assigns a score based on this data, for example, "Responsiveness to customer service: 5 points," "Consistency of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and teachers and store managers can use the evaluation results and suggestions to plan the next response or training.

[0388] Prompt Sentence Examples

[0389] Customer service rating: Hello, customer. We hope you enjoy your shopping experience. Sentiment rating: [{'label': 'POSITIVE', 'score': 0.99}] Generate improvement suggestions.

[0390] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0391] Step 1:

[0392] Users input evaluation criteria points for each lesson or customer service through a web interface. The input evaluation criteria points include "responsiveness to lesson content," "coherence of speech," "proactiveness," etc. These evaluation criteria points are sent to the server via the terminal and stored in a database. The input data is in text format, and the output is the evaluation criteria points stored in the database.

[0393] Step 2:

[0394] The server inputs the received evaluation criteria points into the AI ​​model, and the AI ​​learns these criteria. Through this learning process, the AI ​​model acquires the ability to score statements made in class or at a physical store based on the evaluation criteria. The input data are the evaluation criteria points, and the output data is the trained AI model.

[0395] Step 3:

[0396] During classes or in physical stores, devices use microphones to capture speech from students or employees in real time. The audio data is sent from the devices to a server. The collected audio data is the input data, and the output data is the audio file sent to the server.

[0397] Step 4:

[0398] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. Specifically, a voice recognition system (such as Google Speech Recognition API) is used. The input data is an audio file, and the output data is text data.

[0399] Step 5:

[0400] The device uses an emotion recognition system to analyze emotional data from voice and facial expressions. Using an emotion recognition system (e.g., Hugging Face emotion analysis model), the device receives voice recognition results and voice and facial expression data as input data and generates analyzed emotional data as output data.

[0401] Step 6:

[0402] The server uses generative AI to score the converted text data and emotion data based on the evaluation criteria. The generative AI model (e.g., GPT-2) performs the process of assigning scores based on the input data. The input data is the converted text data and emotion data, and the output data is the scored data based on the evaluation criteria.

[0403] Step 7:

[0404] The scoring results are stored in a database on the server. This includes the specific student or employee ID, the content of the comments, emotional data, and scores based on each evaluation criterion. The input data is the scoring results, and the output data is the evaluation information stored in the database.

[0405] Step 8:

[0406] The server analyzes the collected score data, comment content, and emotion data, and uses a generative AI to create lesson plans for each student or training suggestions for employees. For example, if a student's comments lack consistency, the generative AI generates advice such as, "It would be effective to practice expressing your opinions consistently." Specific operations involve inputting a prompt into the generative AI model and using the output suggestion. The input data are score data, comment content, and emotion data, and the output data are the generated lesson plans or training suggestions.

[0407] Step 9:

[0408] Finally, the created lesson plans and suggestions are notified to the user. The user can check the evaluation results and suggestions for the relevant students or employees through a web interface. The input data are the generated lesson plans and training suggestions, and the output data are the notified lesson plans and suggestions.

[0409] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0410] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0411] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0412] [Second embodiment]

[0413] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0414] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0415] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0416] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0417] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0418] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0419] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0420] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0421] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0422] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0423] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0424] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0425] This invention is a system that combines a speech recognition system and a generative AI to automatically evaluate students' comments and behaviors during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0426] System Configuration

[0427] The system is broadly composed of the following elements:

[0428] 1. Server

[0429] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0430] 3. Web interface accessed by users (teachers)

[0431] Entering and learning evaluation criteria points

[0432] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "coherence of speech," "activeness," etc. These criteria are sent to the server via the terminal and stored in a database.

[0433] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate students' comments in class based on these criteria.

[0434] Collection of audio data during class

[0435] During class, students are placed on their desks with microphones attached to their devices, which capture what they say in real time. The collected audio data is then sent to a server.

[0436] Text conversion using a speech recognition system

[0437] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[0438] Scoring

[0439] The server uses a generative AI to analyze the converted text data and assigns a score based on pre-trained evaluation criteria. Each comment is assigned a score such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0440] Scores and speech records

[0441] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0442] Lesson plan generation and notification

[0443] The server analyzes the collected score data and the content of the comments, and uses the generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions more consistently."

[0444] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[0445] Specific examples

[0446] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0447] This allows teachers to efficiently and fairly evaluate students' comments and attitudes, and with the support of generative AI, they can also strengthen individual instruction.

[0448] The processing flow will be explained below.

[0449] Step 1:

[0450] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," and "activeness."

[0451] Step 2:

[0452] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[0453] Step 3:

[0454] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[0455] Step 4:

[0456] When class begins, a terminal (such as a device with a microphone installed on a student's desk) records what the student says in real time and collects audio data.

[0457] Step 5:

[0458] The collected voice data is sent from the device to a server, which then inputs the received voice data into a voice recognition system and converts the voice data into text data.

[0459] Step 6:

[0460] The server inputs the converted text data into the AI ​​generator, which then scores each comment based on the evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0461] Step 7:

[0462] The server stores the scored results in a database along with the comment content and student ID, thereby recording which student made what comment and how much they were evaluated.

[0463] Step 8:

[0464] Based on the collected score data and the content of the comments, the server uses generative AI to create lesson plans and suggestions for future development for each student. For example, in response to a student's comments that lack consistency, the server generates advice such as, "It would be effective to practice expressing your opinions more consistently."

[0465] Step 9:

[0466] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[0467] Step 10:

[0468] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[0469] Example 1

[0470] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0471] Conventional educational systems lack the means to evaluate students' comments and behaviors in real time during class and generate lesson plans and suggestions for growth. This makes it difficult for teachers to provide appropriate feedback to each student, and it is difficult to promote student growth efficiently.

[0472] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0473] In this invention, the server includes a means for receiving and storing evaluation criteria points for each subject and lesson, a means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, a means for scoring the converted text data based on the evaluation criteria, a means for saving the scored results and the speech content in a database, a means for notifying teachers of the generated specific lesson plans and suggestions for future growth, a terminal for collecting voice data, and a means for transmitting the data to the server. This allows teachers to efficiently and fairly evaluate students' comments and attitudes and to strengthen individual instruction with the support of the generative AI model.

[0474] "Evaluation criteria points for each subject and class" are the standards used to evaluate students' comments and attitudes, and are evaluation criteria set according to the content of the subject or class.

[0475] A "voice recognition system" is a system for converting voice data collected during classes into text data, and is a technology that analyzes voice and outputs it as text information.

[0476] "Text data" refers to text information of voice data converted using a voice recognition system.

[0477] The "means for scoring based on evaluation criteria" is a method for evaluating the converted text data based on evaluation criteria set in advance and assigning a score.

[0478] "Generative AI" is an artificial intelligence technology that generates appropriate responses and results for specific inputs based on pre-learned data.

[0479] The "database" is an information system for storing and managing information such as evaluation results and scored comments.

[0480] A "lesson plan" is a plan created using generative AI that includes specific teaching guidelines and advice to promote each student's future growth.

[0481] "Suggestions for future growth" are specific advice and improvement plans for each student on how to promote their growth in the future.

[0482] A "terminal" is a device equipped with a microphone that is installed to collect student comments during class and is responsible for transmitting audio data to a server.

[0483] MODE FOR CARRYING OUT THE INVENTION

[0484] This invention is a system that combines a speech recognition system and a generative AI model to automatically evaluate students' comments and behavior during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0485] System Configuration

[0486] The system is broadly composed of the following elements:

[0487] 1. Server

[0488] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0489] 3. Web interface accessed by users (teachers)

[0490] Entering and learning evaluation criteria points

[0491] Users (teachers) input evaluation criteria points for each subject and class through a dedicated web interface. These evaluation criteria points include "responsiveness to lesson content," "coherence of speech," and "activeness." These evaluation criteria points are sent to the server via the terminal and stored in a database. The server then inputs the received evaluation criteria points into the generative AI model, and the AI ​​learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments during class based on these criteria.

[0492] Collection of audio data during class

[0493] During class, students are placed on desks equipped with microphones that capture their speech in real time. The collected audio data is temporarily stored on the devices and then sent to a server.

[0494] Text conversion using a speech recognition system

[0495] The server inputs the received voice data into a speech recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data as, "I think the language in this poem is beautiful."

[0496] Scoring

[0497] The server uses a generative AI (e.g., GPT-4 model) to analyze the converted text data and assign scores based on pre-trained evaluation criteria. Each comment is assigned a score, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0498] Scores and speech records

[0499] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0500] Lesson plan generation and notification

[0501] The server analyzes the collected score data and comments, and uses a generative AI to create lesson plans and suggestions for development for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective for you to practice expressing your opinions more consistently." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[0502] Specific examples

[0503] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these scores are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0504] Example prompts

[0505] Below are some example prompts to input to a generative AI model:

[0506] In your Japanese class, set evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student's attitude positive?". Use a generative AI model to analyze students' comments and generate scores and advice.

[0507] Through these steps, the system enables teachers to efficiently and fairly evaluate students' comments and attitudes, and enhance individual instruction with the assistance of generative AI models.

[0508] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0509] Step 1: Enter and save your evaluation criteria points

[0510] The user (teacher) enters the evaluation criteria points for each subject and class (e.g., "responsiveness to class content," "coherence of speech," "proactiveness") through a dedicated web interface. Specifically, the teacher selects the evaluation criteria on the web interface and clicks the "Submit" button.

[0511] Input: Teacher-entered criterion points

[0512] Data processing: Organize the entered evaluation criteria points according to the format

[0513] Output: Data with organized evaluation criteria points

[0514] Specific operation: When the teacher enters the evaluation criteria and presses the submit button, the data is sent to the server and stored in the database.

[0515] Step 2: Collecting audio data

[0516] During class, the terminal (a device with a microphone installed on the student's desk) captures what the student says in real time and temporarily stores it. The collected audio data is then sent from the terminal to a server.

[0517] Input: Audio data captured during class

[0518] Data processing: After temporary storage, organize according to the format

[0519] Output: Organized audio data

[0520] Specific operation: When a student speaks, the device captures the audio and sends the audio data to the server.

[0521] Step 3: Convert audio data to text

[0522] The server inputs the received voice data into a voice recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data.

[0523] Input: Audio data sent to the server

[0524] Data processing: Input the voice data into a voice recognition system and convert it into text data.

[0525] Output: Generated text data

[0526] Specific operation: Sends voice data to a voice recognition system and receives textual results.

[0527] Step 4: Analyze and score the text data

[0528] The server inputs the generated text data into a generative AI model (e.g., GPT-4) for analysis. The generative AI then scores each utterance based on pre-trained evaluation criteria.

[0529] Input: Generated text data

[0530] Data processing: Analyze text data and assign scores based on evaluation criteria

[0531] Output: Scored evaluation results

[0532] Specific operation: Text data is input into the generative AI model, and a score is obtained as the analysis result.

[0533] Step 5: Record your scores and comments

[0534] The server stores the scored results in a database, which includes the student ID, the content of the comment, and the score for each criterion.

[0535] Input: Scored evaluation results

[0536] Data processing: Organize and store the scored results in a database

[0537] Output: Saved evaluation results and comments

[0538] Specific operation: After generating the scoring results, they are immediately written to the database.

[0539] Step 6: Generate and notify lesson plans

[0540] The server inputs the collected score data and comments into a generative AI model to create lesson plans and development suggestions for each student. The generated lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the evaluation results and lesson plans through a web interface.

[0541] Input: Score data and comments

[0542] Data processing: Input into generative AI models to generate lesson plans and development suggestions

[0543] Output: Generated lesson plans and suggestions for development

[0544] Specific operation: When a teacher logs in, a notification of a new lesson plan will be displayed and the contents can be viewed on the details page.

[0545] Through these steps, the system can efficiently evaluate students' comments and behavior during class, and with the support of the generative AI model, provide individual instruction efficiently and effectively.

[0546] (Application example 1)

[0547] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0548] Conventional educational systems have struggled to efficiently and fairly evaluate students' comments and behavior during class. Furthermore, there was no way to automatically generate individual lesson plans or suggestions for improvement, placing a heavy burden on teachers. Furthermore, there was no system in place to evaluate passenger conversations in autonomous vehicles and make suggestions for improving driving conditions, making improving the riding experience a challenge.

[0549] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0550] In this invention, the server includes: means for receiving and storing evaluation criteria points for each subject or lesson; means for inputting student voice data into a speech recognition system and converting it into text data; means for scoring the converted text data based on the evaluation criteria; means for storing the scored results and the speech content in a database; means for notifying the teacher of the generated lesson plan and suggestions for future development; means for evaluating in-car speech data and analyzing the quality of conversation and the degree of agreement between topics; and means for collecting passenger feedback and making suggestions for improving driving conditions. This makes it possible to efficiently and fairly evaluate students' comments and behavior during class and automatically generate individual lesson plans and suggestions for development. It also evaluates passenger conversations and makes suggestions for improving driving conditions in an autonomous vehicle, improving the riding experience.

[0551] "Evaluation criteria points" refer to the evaluation criteria set for each class or subject, and are indicators used to score students' comments and attitudes.

[0552] A "speech recognition system" refers to a technology that converts voice data into text data, and is a system used to accurately transcribe collected speech.

[0553] "Text data" refers to character string data converted by a speech recognition system and is used to record students' utterances as text.

[0554] "Scoring" refers to the process of numerically evaluating the converted text data based on evaluation criteria points.

[0555] "Database" refers to the information system for storing and managing the scored results and comments.

[0556] "Generative AI" refers to artificial intelligence that generates and analyzes text based on large datasets, and is used to automate evaluations and recommendations.

[0557] A "lesson plan" refers to a specific educational strategy to support the individual growth of each student, and is automatically generated by generative AI.

[0558] "Feedback" refers to the opinions and reactions that a system collects from users, and is information used to evaluate and improve the system.

[0559] "Driving situation improvement proposals" refer to specific proposals for improving driving quality based on data collected within an autonomous vehicle.

[0560] "Interface" refers to the means by which users access and operate the system, and is used by teachers to check assessment results and lesson plans.

[0561] "Conversation quality" refers to an indicator that evaluates how meaningful and constructive the conversation is.

[0562] "Topic alignment" refers to an indicator that evaluates how closely a conversation matches the set theme or purpose.

[0563] This invention is a system that can be used in both the fields of educational evaluation and autonomous driving. Specifically, it manages evaluation criteria points for each subject and lesson, automatically evaluates students' comments and behavior during class, and uses generative AI to provide teachers with lesson plans and suggestions for growth. It is also a system that evaluates passenger conversations and makes suggestions for improving driving conditions inside autonomous vehicles.

[0564] System Configuration

[0565] The system consists of the following elements:

[0566] 1. Server: The central hardware that processes, stores, and analyzes data.

[0567] 2. Devices: Devices with microphones installed on students' desks, microphones in self-driving cars, etc.

[0568] 3. Users: School teachers and passengers in self-driving cars.

[0569] Program Overview

[0570] 1. Collection of audio data

[0571] A microphone installed on the device captures what students say in class and conversations inside the self-driving car in real time, and sends the audio data to a server.

[0572] 2. Speech Recognition and Text Conversion

[0573] The server uses a speech recognition system (for example, the SpeechRecognition library) to convert the received voice data into text data, which records what students say and what passengers say as text.

[0574] 3. Scoring and Evaluation

[0575] The server analyzes the converted text data and scores it based on pre-trained evaluation criteria. This process uses a generative AI model (e.g., GPT-3.5-turbo) to evaluate the student's comments based on criteria such as "responsiveness to the lesson content," "coherence of speech," and "activeness." Additionally, in the self-driving car, the server evaluates the quality of the conversation and the degree of topic agreement.

[0576] 4. Saving to the database

[0577] The scored results and comments are stored in a database. The database includes student IDs, comments, and evaluation scores based on each evaluation criterion. This allows for the management of evaluation history.

[0578] 5. Generating lesson plans and suggestions

[0579] The server uses AI to create individual lesson plans and development suggestions for each student based on the collected score data and speech content. For example, it can generate advice such as, "It would be effective to practice expressing your opinions consistently." Then, within the self-driving car, it generates driving improvement suggestions based on feedback on the driving situation.

[0580] 6. Notification of Results

[0581] The final lesson plans and driving improvement suggestions are notified to users (teachers and passengers), who can check the evaluation results and suggestions via a dedicated web interface or smartphone app.

[0582] Specific examples

[0583] For example, if a student says, "I think the language in this poem is beautiful" during a Japanese class, the device's microphone captures the speech and sends it to the server. The server then uses a speech recognition system to convert this speech into text data: "I think the language in this poem is beautiful." The generative AI model then scores this text as "Readiness to respond to lesson content: 5 points," "Consistency of speech: 4 points," and "Positiveness: 3 points," and stores the results in a database.

[0584] Prompt Sentence Examples

[0585] "Analyze the following text and rate the quality and consistency of the conversation: 'The scenery is beautiful. Let's take a photo at the next rest stop.'"

[0586] In this way, the present invention provides a specific form for efficiently collecting, analyzing, evaluating, and proposing improvements to data in both educational evaluation and automated driving.

[0587] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0588] Step 1: Collecting audio data

[0589] The device captures students' speech and conversations inside the self-driving car in real time. A microphone installed on the device picks up the audio and sends the audio data to the server. The input is "audio data" and the output is "audio data sent to the server."

[0590] Step 2: Speech recognition and text conversion

[0591] The server uses the SpeechRecognition library to convert the received voice data into text data. The voice data is input into the voice recognition system and the converted text data is obtained. The input is "voice data" and the output is "text data." Specifically, the voice recognition system analyzes the voice sample and generates a corresponding string of characters.

[0592] Step 3: Scoring based on the evaluation criteria

[0593] The server then scores the converted text data based on pre-trained evaluation criteria. Using a generative AI model (e.g., GPT-3.5-turbo), it assigns a score to the given text based on criteria such as "responsiveness to lesson content," "coherence of speech," and "activeness." The input is "text data" and "evaluation criteria points," and the output is "scored data."

[0594] Step 4: Saving to the Database

[0595] The server saves the scoring results and the comments together in a database. The input is the "score data" and "text data," and the output is the "saved data." Specifically, a new record is added to the database, and the record contains the student ID, comments, and evaluation score.

[0596] Step 5: Generate lesson plans and suggestions

[0597] The server uses generative AI to create individual lesson plans and suggestions for development based on the collected score data and utterances. The inputs are "score data" and "text data," and the output is a "generated lesson plan" and "proposal." Specifically, the server provides prompts to the AI ​​model to generate appropriate advice. For example, "Analyze the following text and generate a lesson plan: What the student uttered."

[0598] Step 6: Notification of results

[0599] The user (teacher) checks the generated lesson plans and proposals through a dedicated web interface or smartphone app. The server sends the evaluation results and proposals to the notification system. The inputs are the "generated lesson plans" and "proposals," and the output is the "notified results." Specifically, the results for each student are displayed on the dashboard of the web interface.

[0600] These steps will enable educational assessment and a conversational assessment and feedback system within an autonomous vehicle.

[0601] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0602] This invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and even emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0603] System Configuration

[0604] The system is broadly composed of the following elements:

[0605] 1. Server

[0606] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0607] 3. Web interface accessed by users (teachers)

[0608] 4. Emotion Engine

[0609] Entering and learning evaluation criteria points

[0610] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the terminal and stored in a database.

[0611] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these criteria. Through this learning process, the AI ​​model can evaluate students' comments and emotions during class based on these criteria.

[0612] Collection of speech and emotion data during class

[0613] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device uses an emotion engine to collect emotional data from the student's voice and facial expressions.

[0614] Speech recognition system for text conversion and emotion recognition

[0615] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[0616] At the same time, the emotion engine analyzes emotional information from voice and students' facial expressions, generating data such as "positive emotions are expressed when speaking."

[0617] Scoring

[0618] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each comment and the emotion it conveys is given a score, such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points."

[0619] Recording scores, comments, and emotions

[0620] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0621] Lesson plan generation and notification

[0622] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend that you provide instruction to increase self-esteem in order to increase positive comments."

[0623] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[0624] Specific examples

[0625] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0626] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[0627] The processing flow will be explained below.

[0628] Step 1:

[0629] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," "activeness," "emotional responsiveness," "emotional consistency," and "emotional activeness."

[0630] Step 2:

[0631] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[0632] Step 3:

[0633] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[0634] Step 4:

[0635] When the lesson begins, the device (such as a microphone-equipped device installed on the student's desk) records what the student says in real time and collects audio data. At the same time, the device uses a camera to capture the student's facial expressions and collect emotional information in real time.

[0636] Step 5:

[0637] The device sends the collected voice data and emotion data to a server, which inputs the voice data into a voice recognition system and converts it into text data.

[0638] Step 6:

[0639] The emotion engine analyzes the emotion data sent to the server and generates emotional information about the student's speech. For example, it recognizes emotions such as "positive" or "negative" from the tone of voice and facial expressions.

[0640] Step 7:

[0641] The server inputs the converted text data and the recognized emotional information into the AI ​​generator, which then scores each utterance based on the evaluation criteria. For example, the evaluation scores could be "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "emotional responsiveness: 4 points," "emotional consistency: 5 points," and "emotional activeness: 3 points."

[0642] Step 8:

[0643] The server stores the scored results in a database along with the comments and student IDs, thereby recording which students made what comments and how they were evaluated.

[0644] Step 9:

[0645] The server uses generative AI to create lesson plans and suggestions for future growth for each student based on the collected score data, comment content, and emotional data. For example, if the comments lack consistency or the emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing consistent opinions. We recommend providing instruction to increase self-esteem in order to increase positive comments."

[0646] Step 10:

[0647] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[0648] Step 11:

[0649] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[0650] Example 2

[0651] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0652] Conventional classroom evaluation systems evaluate students based on subjectivity, which creates issues with the fairness and efficiency of evaluation. In addition, it is difficult to evaluate students based on their emotions and create individual lesson plans, making it difficult to support students' overall growth.

[0653] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for collecting and analyzing students' emotional data, means for scoring the converted text data and emotional data based on the evaluation criteria, means for saving the scored results, the content of the comments, and the emotional data in a database, and means for notifying the teacher of the generated specific lesson plan and suggestions for future growth. This makes it possible to fairly and efficiently evaluate students' comments, attitudes, and even emotions during lessons and provide individual lesson plans.

[0654] "Criteria points" are specific criteria established to assess a student's performance for a particular subject or lesson.

[0655] A "voice recognition system" is a technical system that analyzes input voice data and converts it into corresponding text data.

[0656] "Emotional data" is data that indicates the emotional state of students analyzed from their comments and facial expressions, and includes types of emotions such as positive and negative.

[0657] "Generative AI" refers to artificial intelligence technology that automatically generates evaluations and suggestions based on collected data.

[0658] "Scoring" is the process of numerically evaluating students' comments, attitudes, emotions, etc. based on evaluation criteria.

[0659] A "database" is a system for systematically storing and managing collected data.

[0660] A "lesson plan" is a document that outlines specific plans and policies for teachers to effectively teach students.

[0661] "Notifying" refers to the action of informing teachers of the generated lesson plans and evaluation results.

[0662] The present invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the present invention are described below.

[0663] System Configuration

[0664] The system is broadly composed of the following elements:

[0665] 1. Server

[0666] 2. Terminal (device with microphone installed on student's desk)

[0667] 3. Web interface accessed by users (teachers)

[0668] 4. Emotion engine (e.g., Microsoft Azure Emotion API)

[0669] Entering and learning evaluation criteria points

[0670] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the device and stored in a database. The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments and emotions during class based on these criteria.

[0671] Collection of speech and emotion data during class

[0672] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device also uses an emotion engine to collect emotional data from the student's voice and facial expressions. This is done using a microphone device and a facial recognition camera.

[0673] Speech recognition system for text conversion and emotion recognition

[0674] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful." At the same time, the emotion engine analyzes emotional information from the voice and the student's facial expressions, generating data such as "positive emotion is expressed in the statement."

[0675] Scoring

[0676] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each utterance and its associated emotion are given scores such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points." The prompt is as follows:

[0677] "Based on the student's comments and emotional data, please calculate this student's grade for the lesson. The evaluation criteria are as follows: responsiveness to lesson content, coherence of speech, assertiveness, emotional responsiveness, emotional coherence, and emotional assertiveness."

[0678] Recording scores, comments, and emotions

[0679] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0680] Lesson plan generation and notification

[0681] The server analyzes the collected score data, comment content, and emotional data, and uses a generative AI to create lesson plans and development suggestions for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend providing instruction to increase self-esteem in order to increase positive comments." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[0682] Specific examples

[0683] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0684] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[0685] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0686] Step 1: Enter and save your evaluation criteria points

[0687] Users (teachers) access a dedicated web interface and enter evaluation criteria points for each subject and class.

[0688] Specific actions: Open a web browser, enter your username and password in the login form to log in, enter your evaluation criteria points in the input form, and click the submit button.

[0689] Input: Subject, lesson, evaluation criteria points (e.g., "Readiness to follow lesson content: 5", "Consistency of speech: 4", "Proactiveness: 3")

[0690] Data processing: Send the evaluation criteria points to the server.

[0691] Output: The server stores the metric points in the database.

[0692] Step 2: Learning the evaluation criteria points

[0693] The server inputs the received evaluation criteria points into the AI ​​model, which then learns from them.

[0694] How it works: The server retrieves the evaluation criteria points from the database and passes the data to the AI ​​model, which then performs the learning process based on the evaluation criteria.

[0695] Input: Evaluation criteria point data

[0696] Data calculation: The AI ​​model analyzes the evaluation criteria points and learns.

[0697] Output: The AI ​​model obtains the ability to make an evaluation based on the evaluation criteria.

[0698] Step 3: Collecting audio and emotion data during class

[0699] The device captures students' comments in real time during class, collecting audio data, and simultaneously collects students' emotional data using an emotion engine.

[0700] How it works: The device's built-in microphone records speech in real time, and the device uses an emotion engine to analyze emotions from the voice data.

[0701] Input: Student speech and facial expression data

[0702] Data processing: Record audio data and perform sentiment analysis.

[0703] Output: Voice data and emotion data are obtained.

[0704] Step 4: Text conversion by speech recognition system

[0705] The server inputs the received voice data into a voice recognition system and converts it into text data.

[0706] Specific operation: The server sends the audio file to the speech recognition API and receives text data as the API response.

[0707] Input: Audio data file

[0708] Data processing: The voice recognition system analyzes the voice data and converts it into text data.

[0709] Output: Text data (e.g., "I think the language in this poem is beautiful.")

[0710] Step 5: Analyze the sentiment data

[0711] The server receives the emotion data analyzed by the emotion engine.

[0712] Specific operation: The emotion engine analyzes voice and facial expression data and sends the emotion data to the server.

[0713] Input: Voice data, facial expression data

[0714] Data processing: Emotion analysis using an emotion engine

[0715] Output: Emotion data (e.g. "positive")

[0716] Step 6: Scoring

[0717] The server uses generative AI to score the converted text data and emotional data based on evaluation criteria.

[0718] Specific operation: The server sends text data and emotion data to the generative AI model and obtains an evaluation score from the model.

[0719] Input: Text data, emotion data, evaluation criteria points

[0720] Data calculation: The generating AI performs the scoring.

[0721] Output: Score evaluation (e.g., "Readiness to content: 5 points," "Spoken consistency: 4 points," "Positiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," "Emotional positiveness: 3 points")

[0722] Step 7: Record your scores, comments, and feelings

[0723] The server stores the scored results in a database.

[0724] Specific behavior: The server executes an SQL query to insert the evaluation result into the database.

[0725] Input: Scored evaluation, student ID, speech content, emotion data

[0726] Data processing: Processing to save data in the database.

[0727] Output: Recording to database completed

[0728] Step 8: Generate and notify lesson plans

[0729] The server analyzes the collected score data, speech content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth, which are then notified to the user.

[0730] Specific operations: The server analyzes the score data, sends prompts to the generative AI model, obtains the generated lesson plans and suggestions, and sends a request to notify the user of the obtained lesson plans and suggestions.

[0731] Input: Score data, speech content, emotion data

[0732] Data calculation: Generative AI generates lesson plans and suggestions.

[0733] Output: Lesson plans, suggestions, and user notifications

[0734] Through these processing steps, the present invention realizes a system that can effectively and fairly evaluate students' comments and emotions and provide optimal teaching plans.

[0735] (Application example 2)

[0736] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0737] Traditional methods for evaluating students' comments and behavior during class rely on the teacher's subjective judgment and lack objectivity and efficiency. Furthermore, evaluations of store staff's customer service are often subjective, making it difficult to provide appropriate guidance for improving staff skills and training. There is a need to solve these problems and provide fair and efficient evaluations and lesson plans.

[0738] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for scoring the converted text data based on the evaluation criteria, means for saving the scored results and the content of the comments in a database, means for notifying the teacher of the generated specific lesson plans and suggestions for future development, means for applying the voice recognition system to the customer service of employees in a physical store and converting the comments made during customer service into text data, means for analyzing the emotions of employees and customers using an emotion recognition system, means for evaluating the customer service response based on the content of the comments and emotions, and means for generating and notifying points for improvement and training suggestions from the evaluation results. This enables fair and efficient evaluation and the generation and notification of specific lesson plans and training suggestions.

[0739] "Evaluation criteria points for each subject and class" are the evaluation criteria points set for each subject and class, and refer to specific evaluation items and standards.

[0740] "Audio data" refers to collected audio information from students and employees, including speech during classes and in physical stores.

[0741] A "voice recognition system" is a system that converts collected voice data into text data.

[0742] "Text data" is character information converted from voice data by a voice recognition system.

[0743] "Evaluation criteria" refers to the items and indicators that serve as the basis for scoring based on text data.

[0744] "Scoring" is the process of quantitatively evaluating text data based on evaluation criteria and expressing it as a number.

[0745] A "database" is a data storage system for centrally storing and managing information such as scored results and comment content.

[0746] "Generative AI" refers to an artificial intelligence model that generates specific advice, such as lesson plans and training suggestions, based on collected data.

[0747] An "emotion recognition system" is a system that analyzes emotions from voice and facial expressions and generates the results as data.

[0748] "Customer service" refers to the contact and service provided by employees to customers in physical stores.

[0749] "Training proposals" are proposals based on the evaluation results for specific improvements and practice methods to improve and develop the skills of employees or students.

[0750] This invention is a system that combines a voice recognition system, generative AI, and an emotion recognition system to efficiently and fairly evaluate the comments and attitudes of students and employees during classes or in physical stores, and provide specific lesson plans and training suggestions.

[0751] System Configuration

[0752] The system roughly consists of the following elements:

[0753] 1. Server

[0754] 2. Terminals (such as devices with microphones installed on students' or employees' desks or in physical stores)

[0755] 3. Web interface accessed by users (teachers and store managers)

[0756] 4. Emotion Recognition System

[0757] Entering and learning evaluation criteria points

[0758] Users (teachers and store managers) use a dedicated web interface to input evaluation criteria points for each lesson or customer service. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "proactiveness," as well as "emotional responsiveness," "emotional consistency," and "emotional proactiveness." These criteria are sent to the server via the terminal and stored in a database.

[0759] The server inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate the comments and emotions of students in class or employees in a physical store based on these criteria.

[0760] Collecting speech and emotion data during classes and in-store

[0761] During classes or in brick-and-mortar stores, the devices capture students' and employees' speech in real time. The collected voice data is sent from the devices to a server. At the same time, the devices also use an emotion recognition system to collect emotional data from voice and facial expressions.

[0762] Speech recognition system for text conversion and emotion recognition

[0763] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," or an employee says, "Hello, customer. Please take your time and enjoy your shopping," the voice is converted into text data. At the same time, the emotion recognition system analyzes emotional information from the voice and facial expressions and generates data such as, "Positive emotions are expressed in the statement."

[0764] Scoring

[0765] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "activeness: 3 points" for each comment and its associated emotion.

[0766] Recording scores, comments, and emotions

[0767] The scoring results are stored in a database on the server. This includes student and employee IDs, comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students and employees made what comments and how they were evaluated.

[0768] Generation and notification of lesson plans and training proposals

[0769] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans for each student and employee training suggestions. For example, if a student's comments lack consistency, the generative AI might advise, "It would be effective to practice expressing your opinions consistently." If an employee's emotions are negative, the AI ​​might suggest, "We recommend training to increase self-esteem in order to increase positive expressions."

[0770] The created lesson plans and suggestions are notified to the user (teacher or store manager), who can then check the evaluation results and suggestions for the relevant student or employee through a web interface.

[0771] Specific examples

[0772] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech consistent?", "Is the student's attitude positive?", and also "Emotional readiness," "Emotional consistency," and "Emotional positivity." When a student speaks during class, a device equipped with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the comment is recognized as "positive."

[0773] In a physical store, if a store clerk says, "Hello, customer. Please take your time and enjoy your shopping," a microphone-equipped device captures the voice and sends it to a server. The voice data is converted into text, and the emotion expressed at the time of the statement is recognized as "positive."

[0774] The generative AI assigns a score based on this data, for example, "Responsiveness to customer service: 5 points," "Consistency of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and teachers and store managers can use the evaluation results and suggestions to plan the next response or training.

[0775] Prompt Sentence Examples

[0776] Customer service rating: Hello, customer. We hope you enjoy your shopping experience. Sentiment rating: [{'label': 'POSITIVE', 'score': 0.99}] Generate improvement suggestions.

[0777] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0778] Step 1:

[0779] Users input evaluation criteria points for each lesson or customer service through a web interface. The input evaluation criteria points include "responsiveness to lesson content," "coherence of speech," "proactiveness," etc. These evaluation criteria points are sent to the server via the terminal and stored in a database. The input data is in text format, and the output is the evaluation criteria points stored in the database.

[0780] Step 2:

[0781] The server inputs the received evaluation criteria points into the AI ​​model, and the AI ​​learns these criteria. Through this learning process, the AI ​​model acquires the ability to score statements made in class or at a physical store based on the evaluation criteria. The input data are the evaluation criteria points, and the output data is the trained AI model.

[0782] Step 3:

[0783] During classes or in physical stores, devices use microphones to capture speech from students or employees in real time. The audio data is sent from the devices to a server. The collected audio data is the input data, and the output data is the audio file sent to the server.

[0784] Step 4:

[0785] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. Specifically, a voice recognition system (such as Google Speech Recognition API) is used. The input data is an audio file, and the output data is text data.

[0786] Step 5:

[0787] The device uses an emotion recognition system to analyze emotional data from voice and facial expressions. Using an emotion recognition system (e.g., Hugging Face emotion analysis model), the device receives voice recognition results and voice and facial expression data as input data and generates analyzed emotional data as output data.

[0788] Step 6:

[0789] The server uses generative AI to score the converted text data and emotion data based on the evaluation criteria. The generative AI model (e.g., GPT-2) performs the process of assigning scores based on the input data. The input data is the converted text data and emotion data, and the output data is the scored data based on the evaluation criteria.

[0790] Step 7:

[0791] The scoring results are stored in a database on the server. This includes the specific student or employee ID, the content of the comments, emotional data, and scores based on each evaluation criterion. The input data is the scoring results, and the output data is the evaluation information stored in the database.

[0792] Step 8:

[0793] The server analyzes the collected score data, comment content, and emotion data, and uses a generative AI to create lesson plans for each student or training suggestions for employees. For example, if a student's comments lack consistency, the generative AI generates advice such as, "It would be effective to practice expressing your opinions consistently." Specific operations involve inputting a prompt into the generative AI model and using the output suggestion. The input data are score data, comment content, and emotion data, and the output data are the generated lesson plans or training suggestions.

[0794] Step 9:

[0795] Finally, the created lesson plans and suggestions are notified to the user. The user can check the evaluation results and suggestions for the relevant students or employees through a web interface. The input data are the generated lesson plans and training suggestions, and the output data are the notified lesson plans and suggestions.

[0796] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0797] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0798] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0799] [Third embodiment]

[0800] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0801] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0802] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0803] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0804] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0805] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0806] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0807] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0808] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0809] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0810] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0811] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0812] This invention is a system that combines a speech recognition system and a generative AI to automatically evaluate students' comments and behaviors during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0813] System Configuration

[0814] The system is broadly composed of the following elements:

[0815] 1. Server

[0816] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0817] 3. Web interface accessed by users (teachers)

[0818] Entering and learning evaluation criteria points

[0819] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "coherence of speech," "activeness," etc. These criteria are sent to the server via the terminal and stored in a database.

[0820] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate students' comments in class based on these criteria.

[0821] Collection of audio data during class

[0822] During class, students are placed on their desks with microphones attached to their devices, which capture what they say in real time. The collected audio data is then sent to a server.

[0823] Text conversion using a speech recognition system

[0824] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[0825] Scoring

[0826] The server uses a generative AI to analyze the converted text data and assigns a score based on pre-trained evaluation criteria. Each comment is assigned a score such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0827] Scores and speech records

[0828] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0829] Lesson plan generation and notification

[0830] The server analyzes the collected score data and the content of the comments, and uses the generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions more consistently."

[0831] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[0832] Specific examples

[0833] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0834] This allows teachers to efficiently and fairly evaluate students' comments and attitudes, and with the support of generative AI, they can also strengthen individual instruction.

[0835] The processing flow will be explained below.

[0836] Step 1:

[0837] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," and "activeness."

[0838] Step 2:

[0839] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[0840] Step 3:

[0841] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[0842] Step 4:

[0843] When class begins, a terminal (such as a device with a microphone installed on a student's desk) records what the student says in real time and collects audio data.

[0844] Step 5:

[0845] The collected voice data is sent from the device to a server, which then inputs the received voice data into a voice recognition system and converts the voice data into text data.

[0846] Step 6:

[0847] The server inputs the converted text data into the AI ​​generator, which then scores each comment based on the evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0848] Step 7:

[0849] The server stores the scored results in a database along with the comment content and student ID, thereby recording which student made what comment and how much they were evaluated.

[0850] Step 8:

[0851] Based on the collected score data and the content of the comments, the server uses generative AI to create lesson plans and suggestions for future development for each student. For example, in response to a student's comments that lack consistency, the server generates advice such as, "It would be effective to practice expressing your opinions more consistently."

[0852] Step 9:

[0853] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[0854] Step 10:

[0855] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[0856] Example 1

[0857] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0858] Conventional educational systems lack the means to evaluate students' comments and behaviors in real time during class and generate lesson plans and suggestions for growth. This makes it difficult for teachers to provide appropriate feedback to each student, and it is difficult to promote student growth efficiently.

[0859] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0860] In this invention, the server includes a means for receiving and storing evaluation criteria points for each subject and lesson, a means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, a means for scoring the converted text data based on the evaluation criteria, a means for saving the scored results and the speech content in a database, a means for notifying teachers of the generated specific lesson plans and suggestions for future growth, a terminal for collecting voice data, and a means for transmitting the data to the server. This allows teachers to efficiently and fairly evaluate students' comments and attitudes and to strengthen individual instruction with the support of the generative AI model.

[0861] "Evaluation criteria points for each subject and class" are the standards used to evaluate students' comments and attitudes, and are evaluation criteria set according to the content of the subject or class.

[0862] A "voice recognition system" is a system for converting voice data collected during classes into text data, and is a technology that analyzes voice and outputs it as text information.

[0863] "Text data" refers to text information of voice data converted using a voice recognition system.

[0864] The "means for scoring based on evaluation criteria" is a method for evaluating the converted text data based on evaluation criteria set in advance and assigning a score.

[0865] "Generative AI" is an artificial intelligence technology that generates appropriate responses and results for specific inputs based on pre-learned data.

[0866] The "database" is an information system for storing and managing information such as evaluation results and scored comments.

[0867] A "lesson plan" is a plan created using generative AI that includes specific teaching guidelines and advice to promote each student's future growth.

[0868] "Suggestions for future growth" are specific advice and improvement plans for each student on how to promote their growth in the future.

[0869] A "terminal" is a device equipped with a microphone that is installed to collect student comments during class and is responsible for transmitting audio data to a server.

[0870] MODE FOR CARRYING OUT THE INVENTION

[0871] This invention is a system that combines a speech recognition system and a generative AI model to automatically evaluate students' comments and behavior during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0872] System Configuration

[0873] The system is broadly composed of the following elements:

[0874] 1. Server

[0875] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0876] 3. Web interface accessed by users (teachers)

[0877] Entering and learning evaluation criteria points

[0878] Users (teachers) input evaluation criteria points for each subject and class through a dedicated web interface. These evaluation criteria points include "responsiveness to lesson content," "coherence of speech," and "activeness." These evaluation criteria points are sent to the server via the terminal and stored in a database. The server then inputs the received evaluation criteria points into the generative AI model, and the AI ​​learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments during class based on these criteria.

[0879] Collection of audio data during class

[0880] During class, students are placed on desks equipped with microphones that capture their speech in real time. The collected audio data is temporarily stored on the devices and then sent to a server.

[0881] Text conversion using a speech recognition system

[0882] The server inputs the received voice data into a speech recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data as, "I think the language in this poem is beautiful."

[0883] Scoring

[0884] The server uses a generative AI (e.g., GPT-4 model) to analyze the converted text data and assign scores based on pre-trained evaluation criteria. Each comment is assigned a score, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[0885] Scores and speech records

[0886] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[0887] Lesson plan generation and notification

[0888] The server analyzes the collected score data and comments, and uses a generative AI to create lesson plans and suggestions for development for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective for you to practice expressing your opinions more consistently." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[0889] Specific examples

[0890] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these scores are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[0891] Example prompts

[0892] Below are some example prompts to input to a generative AI model:

[0893] In your Japanese class, set evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student's attitude positive?". Use a generative AI model to analyze students' comments and generate scores and advice.

[0894] Through these steps, the system enables teachers to efficiently and fairly evaluate students' comments and attitudes, and enhance individual instruction with the assistance of generative AI models.

[0895] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0896] Step 1: Enter and save your evaluation criteria points

[0897] The user (teacher) enters the evaluation criteria points for each subject and class (e.g., "responsiveness to class content," "coherence of speech," "proactiveness") through a dedicated web interface. Specifically, the teacher selects the evaluation criteria on the web interface and clicks the "Submit" button.

[0898] Input: Teacher-entered criterion points

[0899] Data processing: Organize the entered evaluation criteria points according to the format

[0900] Output: Data with organized evaluation criteria points

[0901] Specific operation: When the teacher enters the evaluation criteria and presses the submit button, the data is sent to the server and stored in the database.

[0902] Step 2: Collecting audio data

[0903] During class, the terminal (a device with a microphone installed on the student's desk) captures what the student says in real time and temporarily stores it. The collected audio data is then sent from the terminal to a server.

[0904] Input: Audio data captured during class

[0905] Data processing: After temporary storage, organize according to the format

[0906] Output: Organized audio data

[0907] Specific operation: When a student speaks, the device captures the audio and sends the audio data to the server.

[0908] Step 3: Convert audio data to text

[0909] The server inputs the received voice data into a voice recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data.

[0910] Input: Audio data sent to the server

[0911] Data processing: Input the voice data into a voice recognition system and convert it into text data.

[0912] Output: Generated text data

[0913] Specific operation: Sends voice data to a voice recognition system and receives textual results.

[0914] Step 4: Analyze and score the text data

[0915] The server inputs the generated text data into a generative AI model (e.g., GPT-4) for analysis. The generative AI then scores each utterance based on pre-trained evaluation criteria.

[0916] Input: Generated text data

[0917] Data processing: Analyze text data and assign scores based on evaluation criteria

[0918] Output: Scored evaluation results

[0919] Specific operation: Text data is input into the generative AI model, and a score is obtained as the analysis result.

[0920] Step 5: Record your scores and comments

[0921] The server stores the scored results in a database, which includes the student ID, the content of the comment, and the score for each criterion.

[0922] Input: Scored evaluation results

[0923] Data processing: Organize and store the scored results in a database

[0924] Output: Saved evaluation results and comments

[0925] Specific operation: After generating the scoring results, they are immediately written to the database.

[0926] Step 6: Generate and notify lesson plans

[0927] The server inputs the collected score data and comments into a generative AI model to create lesson plans and development suggestions for each student. The generated lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the evaluation results and lesson plans through a web interface.

[0928] Input: Score data and comments

[0929] Data processing: Input into generative AI models to generate lesson plans and development suggestions

[0930] Output: Generated lesson plans and suggestions for development

[0931] Specific operation: When a teacher logs in, a notification of a new lesson plan will be displayed and the contents can be viewed on the details page.

[0932] Through these steps, the system can efficiently evaluate students' comments and behavior during class, and with the support of the generative AI model, provide individual instruction efficiently and effectively.

[0933] (Application example 1)

[0934] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0935] Conventional educational systems have struggled to efficiently and fairly evaluate students' comments and behavior during class. Furthermore, there was no way to automatically generate individual lesson plans or suggestions for improvement, placing a heavy burden on teachers. Furthermore, there was no system in place to evaluate passenger conversations in autonomous vehicles and make suggestions for improving driving conditions, making improving the riding experience a challenge.

[0936] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0937] In this invention, the server includes: means for receiving and storing evaluation criteria points for each subject or lesson; means for inputting student voice data into a speech recognition system and converting it into text data; means for scoring the converted text data based on the evaluation criteria; means for storing the scored results and the speech content in a database; means for notifying the teacher of the generated lesson plan and suggestions for future development; means for evaluating in-car speech data and analyzing the quality of conversation and the degree of agreement between topics; and means for collecting passenger feedback and making suggestions for improving driving conditions. This makes it possible to efficiently and fairly evaluate students' comments and behavior during class and automatically generate individual lesson plans and suggestions for development. It also evaluates passenger conversations and makes suggestions for improving driving conditions in an autonomous vehicle, improving the riding experience.

[0938] "Evaluation criteria points" refer to the evaluation criteria set for each class or subject, and are indicators used to score students' comments and attitudes.

[0939] A "speech recognition system" refers to a technology that converts voice data into text data, and is a system used to accurately transcribe collected speech.

[0940] "Text data" refers to character string data converted by a speech recognition system and is used to record students' utterances as text.

[0941] "Scoring" refers to the process of numerically evaluating the converted text data based on evaluation criteria points.

[0942] "Database" refers to the information system for storing and managing the scored results and comments.

[0943] "Generative AI" refers to artificial intelligence that generates and analyzes text based on large datasets, and is used to automate evaluations and recommendations.

[0944] A "lesson plan" refers to a specific educational strategy to support the individual growth of each student, and is automatically generated by generative AI.

[0945] "Feedback" refers to the opinions and reactions that a system collects from users, and is information used to evaluate and improve the system.

[0946] "Driving situation improvement proposals" refer to specific proposals for improving driving quality based on data collected within an autonomous vehicle.

[0947] "Interface" refers to the means by which users access and operate the system, and is used by teachers to check assessment results and lesson plans.

[0948] "Conversation quality" refers to an indicator that evaluates how meaningful and constructive the conversation is.

[0949] "Topic alignment" refers to an indicator that evaluates how closely a conversation matches the set theme or purpose.

[0950] This invention is a system that can be used in both the fields of educational evaluation and autonomous driving. Specifically, it manages evaluation criteria points for each subject and lesson, automatically evaluates students' comments and behavior during class, and uses generative AI to provide teachers with lesson plans and suggestions for growth. It is also a system that evaluates passenger conversations and makes suggestions for improving driving conditions inside autonomous vehicles.

[0951] System Configuration

[0952] The system consists of the following elements:

[0953] 1. Server: The central hardware that processes, stores, and analyzes data.

[0954] 2. Devices: Devices with microphones installed on students' desks, microphones in self-driving cars, etc.

[0955] 3. Users: School teachers and passengers in self-driving cars.

[0956] Program Overview

[0957] 1. Collection of audio data

[0958] A microphone installed on the device captures what students say in class and conversations inside the self-driving car in real time, and sends the audio data to a server.

[0959] 2. Speech Recognition and Text Conversion

[0960] The server uses a speech recognition system (for example, the SpeechRecognition library) to convert the received voice data into text data, which records what students say and what passengers say as text.

[0961] 3. Scoring and Evaluation

[0962] The server analyzes the converted text data and scores it based on pre-trained evaluation criteria. This process uses a generative AI model (e.g., GPT-3.5-turbo) to evaluate the student's comments based on criteria such as "responsiveness to the lesson content," "coherence of speech," and "activeness." Additionally, in the self-driving car, the server evaluates the quality of the conversation and the degree of topic agreement.

[0963] 4. Saving to the database

[0964] The scored results and comments are stored in a database. The database includes student IDs, comments, and evaluation scores based on each evaluation criterion. This allows for the management of evaluation history.

[0965] 5. Generating lesson plans and suggestions

[0966] The server uses AI to create individual lesson plans and development suggestions for each student based on the collected score data and speech content. For example, it can generate advice such as, "It would be effective to practice expressing your opinions consistently." Then, within the self-driving car, it generates driving improvement suggestions based on feedback on the driving situation.

[0967] 6. Notification of Results

[0968] The final lesson plans and driving improvement suggestions are notified to users (teachers and passengers), who can check the evaluation results and suggestions via a dedicated web interface or smartphone app.

[0969] Specific examples

[0970] For example, if a student says, "I think the language in this poem is beautiful" during a Japanese class, the device's microphone captures the speech and sends it to the server. The server then uses a speech recognition system to convert this speech into text data: "I think the language in this poem is beautiful." The generative AI model then scores this text as "Readiness to respond to lesson content: 5 points," "Consistency of speech: 4 points," and "Positiveness: 3 points," and stores the results in a database.

[0971] Prompt Sentence Examples

[0972] "Analyze the following text and rate the quality and consistency of the conversation: 'The scenery is beautiful. Let's take a photo at the next rest stop.'"

[0973] In this way, the present invention provides a specific form for efficiently collecting, analyzing, evaluating, and proposing improvements to data in both educational evaluation and automated driving.

[0974] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0975] Step 1: Collecting audio data

[0976] The device captures students' speech and conversations inside the self-driving car in real time. A microphone installed on the device picks up the audio and sends the audio data to the server. The input is "audio data" and the output is "audio data sent to the server."

[0977] Step 2: Speech recognition and text conversion

[0978] The server uses the SpeechRecognition library to convert the received voice data into text data. The voice data is input into the voice recognition system and the converted text data is obtained. The input is "voice data" and the output is "text data." Specifically, the voice recognition system analyzes the voice sample and generates a corresponding string of characters.

[0979] Step 3: Scoring based on the evaluation criteria

[0980] The server then scores the converted text data based on pre-trained evaluation criteria. Using a generative AI model (e.g., GPT-3.5-turbo), it assigns a score to the given text based on criteria such as "responsiveness to lesson content," "coherence of speech," and "activeness." The input is "text data" and "evaluation criteria points," and the output is "scored data."

[0981] Step 4: Saving to the Database

[0982] The server saves the scoring results and the comments together in a database. The input is the "score data" and "text data," and the output is the "saved data." Specifically, a new record is added to the database, and the record contains the student ID, comments, and evaluation score.

[0983] Step 5: Generate lesson plans and suggestions

[0984] The server uses generative AI to create individual lesson plans and suggestions for development based on the collected score data and utterances. The inputs are "score data" and "text data," and the output is a "generated lesson plan" and "proposal." Specifically, the server provides prompts to the AI ​​model to generate appropriate advice. For example, "Analyze the following text and generate a lesson plan: What the student uttered."

[0985] Step 6: Notification of results

[0986] The user (teacher) checks the generated lesson plans and proposals through a dedicated web interface or smartphone app. The server sends the evaluation results and proposals to the notification system. The inputs are the "generated lesson plans" and "proposals," and the output is the "notified results." Specifically, the results for each student are displayed on the dashboard of the web interface.

[0987] These steps will enable educational assessment and a conversational assessment and feedback system within an autonomous vehicle.

[0988] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0989] This invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and even emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[0990] System Configuration

[0991] The system is broadly composed of the following elements:

[0992] 1. Server

[0993] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[0994] 3. Web interface accessed by users (teachers)

[0995] 4. Emotion Engine

[0996] Entering and learning evaluation criteria points

[0997] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the terminal and stored in a database.

[0998] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these criteria. Through this learning process, the AI ​​model can evaluate students' comments and emotions during class based on these criteria.

[0999] Collection of speech and emotion data during class

[1000] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device uses an emotion engine to collect emotional data from the student's voice and facial expressions.

[1001] Speech recognition system for text conversion and emotion recognition

[1002] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[1003] At the same time, the emotion engine analyzes emotional information from voice and students' facial expressions, generating data such as "positive emotions are expressed when speaking."

[1004] Scoring

[1005] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each comment and the emotion it conveys is given a score, such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points."

[1006] Recording scores, comments, and emotions

[1007] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[1008] Lesson plan generation and notification

[1009] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend that you provide instruction to increase self-esteem in order to increase positive comments."

[1010] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[1011] Specific examples

[1012] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[1013] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[1014] The processing flow will be explained below.

[1015] Step 1:

[1016] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," "activeness," "emotional responsiveness," "emotional consistency," and "emotional activeness."

[1017] Step 2:

[1018] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[1019] Step 3:

[1020] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[1021] Step 4:

[1022] When the lesson begins, the device (such as a microphone-equipped device installed on the student's desk) records what the student says in real time and collects audio data. At the same time, the device uses a camera to capture the student's facial expressions and collect emotional information in real time.

[1023] Step 5:

[1024] The device sends the collected voice data and emotion data to a server, which inputs the voice data into a voice recognition system and converts it into text data.

[1025] Step 6:

[1026] The emotion engine analyzes the emotion data sent to the server and generates emotional information about the student's speech. For example, it recognizes emotions such as "positive" or "negative" from the tone of voice and facial expressions.

[1027] Step 7:

[1028] The server inputs the converted text data and the recognized emotional information into the AI ​​generator, which then scores each utterance based on the evaluation criteria. For example, the evaluation scores could be "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "emotional responsiveness: 4 points," "emotional consistency: 5 points," and "emotional activeness: 3 points."

[1029] Step 8:

[1030] The server stores the scored results in a database along with the comments and student IDs, thereby recording which students made what comments and how they were evaluated.

[1031] Step 9:

[1032] The server uses generative AI to create lesson plans and suggestions for future growth for each student based on the collected score data, comment content, and emotional data. For example, if the comments lack consistency or the emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing consistent opinions. We recommend providing instruction to increase self-esteem in order to increase positive comments."

[1033] Step 10:

[1034] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[1035] Step 11:

[1036] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[1037] Example 2

[1038] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1039] Conventional classroom evaluation systems evaluate students based on subjectivity, which creates issues with the fairness and efficiency of evaluation. In addition, it is difficult to evaluate students based on their emotions and create individual lesson plans, making it difficult to support students' overall growth.

[1040] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for collecting and analyzing students' emotional data, means for scoring the converted text data and emotional data based on the evaluation criteria, means for saving the scored results, the content of the comments, and the emotional data in a database, and means for notifying the teacher of the generated specific lesson plan and suggestions for future growth. This makes it possible to fairly and efficiently evaluate students' comments, attitudes, and even emotions during lessons and provide individual lesson plans.

[1041] "Criteria points" are specific criteria established to assess a student's performance for a particular subject or lesson.

[1042] A "voice recognition system" is a technical system that analyzes input voice data and converts it into corresponding text data.

[1043] "Emotional data" is data that indicates the emotional state of students analyzed from their comments and facial expressions, and includes types of emotions such as positive and negative.

[1044] "Generative AI" refers to artificial intelligence technology that automatically generates evaluations and suggestions based on collected data.

[1045] "Scoring" is the process of numerically evaluating students' comments, attitudes, emotions, etc. based on evaluation criteria.

[1046] A "database" is a system for systematically storing and managing collected data.

[1047] A "lesson plan" is a document that outlines specific plans and policies for teachers to effectively teach students.

[1048] "Notifying" refers to the action of informing teachers of the generated lesson plans and evaluation results.

[1049] The present invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the present invention are described below.

[1050] System Configuration

[1051] The system is broadly composed of the following elements:

[1052] 1. Server

[1053] 2. Terminal (device with microphone installed on student's desk)

[1054] 3. Web interface accessed by users (teachers)

[1055] 4. Emotion engine (e.g., Microsoft Azure Emotion API)

[1056] Entering and learning evaluation criteria points

[1057] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the device and stored in a database. The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments and emotions during class based on these criteria.

[1058] Collection of speech and emotion data during class

[1059] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device also uses an emotion engine to collect emotional data from the student's voice and facial expressions. This is done using a microphone device and a facial recognition camera.

[1060] Speech recognition system for text conversion and emotion recognition

[1061] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful." At the same time, the emotion engine analyzes emotional information from the voice and the student's facial expressions, generating data such as "positive emotion is expressed in the statement."

[1062] Scoring

[1063] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each utterance and its associated emotion are given scores such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points." The prompt is as follows:

[1064] "Based on the student's comments and emotional data, please calculate this student's grade for the lesson. The evaluation criteria are as follows: responsiveness to lesson content, coherence of speech, assertiveness, emotional responsiveness, emotional coherence, and emotional assertiveness."

[1065] Recording scores, comments, and emotions

[1066] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[1067] Lesson plan generation and notification

[1068] The server analyzes the collected score data, comment content, and emotional data, and uses a generative AI to create lesson plans and development suggestions for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend providing instruction to increase self-esteem in order to increase positive comments." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[1069] Specific examples

[1070] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[1071] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[1072] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1073] Step 1: Enter and save your evaluation criteria points

[1074] Users (teachers) access a dedicated web interface and enter evaluation criteria points for each subject and class.

[1075] Specific actions: Open a web browser, enter your username and password in the login form to log in, enter your evaluation criteria points in the input form, and click the submit button.

[1076] Input: Subject, lesson, evaluation criteria points (e.g., "Readiness to follow lesson content: 5", "Consistency of speech: 4", "Proactiveness: 3")

[1077] Data processing: Send the evaluation criteria points to the server.

[1078] Output: The server stores the metric points in the database.

[1079] Step 2: Learning the evaluation criteria points

[1080] The server inputs the received evaluation criteria points into the AI ​​model, which then learns from them.

[1081] How it works: The server retrieves the evaluation criteria points from the database and passes the data to the AI ​​model, which then performs the learning process based on the evaluation criteria.

[1082] Input: Evaluation criteria point data

[1083] Data calculation: The AI ​​model analyzes the evaluation criteria points and learns.

[1084] Output: The AI ​​model obtains the ability to make an evaluation based on the evaluation criteria.

[1085] Step 3: Collecting audio and emotion data during class

[1086] The device captures students' comments in real time during class, collecting audio data, and simultaneously collects students' emotional data using an emotion engine.

[1087] How it works: The device's built-in microphone records speech in real time, and the device uses an emotion engine to analyze emotions from the voice data.

[1088] Input: Student speech and facial expression data

[1089] Data processing: Record audio data and perform sentiment analysis.

[1090] Output: Voice data and emotion data are obtained.

[1091] Step 4: Text conversion by speech recognition system

[1092] The server inputs the received voice data into a voice recognition system and converts it into text data.

[1093] Specific operation: The server sends the audio file to the speech recognition API and receives text data as the API response.

[1094] Input: Audio data file

[1095] Data processing: The voice recognition system analyzes the voice data and converts it into text data.

[1096] Output: Text data (e.g., "I think the language in this poem is beautiful.")

[1097] Step 5: Analyze the sentiment data

[1098] The server receives the emotion data analyzed by the emotion engine.

[1099] Specific operation: The emotion engine analyzes voice and facial expression data and sends the emotion data to the server.

[1100] Input: Voice data, facial expression data

[1101] Data processing: Emotion analysis using an emotion engine

[1102] Output: Emotion data (e.g. "positive")

[1103] Step 6: Scoring

[1104] The server uses generative AI to score the converted text data and emotional data based on evaluation criteria.

[1105] Specific operation: The server sends text data and emotion data to the generative AI model and obtains an evaluation score from the model.

[1106] Input: Text data, emotion data, evaluation criteria points

[1107] Data calculation: The generating AI performs the scoring.

[1108] Output: Score evaluation (e.g., "Readiness to content: 5 points," "Spoken consistency: 4 points," "Positiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," "Emotional positiveness: 3 points")

[1109] Step 7: Record your scores, comments, and feelings

[1110] The server stores the scored results in a database.

[1111] Specific behavior: The server executes an SQL query to insert the evaluation result into the database.

[1112] Input: Scored evaluation, student ID, speech content, emotion data

[1113] Data processing: Processing to save data in the database.

[1114] Output: Recording to database completed

[1115] Step 8: Generate and notify lesson plans

[1116] The server analyzes the collected score data, speech content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth, which are then notified to the user.

[1117] Specific operations: The server analyzes the score data, sends prompts to the generative AI model, obtains the generated lesson plans and suggestions, and sends a request to notify the user of the obtained lesson plans and suggestions.

[1118] Input: Score data, speech content, emotion data

[1119] Data calculation: Generative AI generates lesson plans and suggestions.

[1120] Output: Lesson plans, suggestions, and user notifications

[1121] Through these processing steps, the present invention realizes a system that can effectively and fairly evaluate students' comments and emotions and provide optimal teaching plans.

[1122] (Application example 2)

[1123] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1124] Traditional methods for evaluating students' comments and behavior during class rely on the teacher's subjective judgment and lack objectivity and efficiency. Furthermore, evaluations of store staff's customer service are often subjective, making it difficult to provide appropriate guidance for improving staff skills and training. There is a need to solve these problems and provide fair and efficient evaluations and lesson plans.

[1125] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for scoring the converted text data based on the evaluation criteria, means for saving the scored results and the content of the comments in a database, means for notifying the teacher of the generated specific lesson plans and suggestions for future development, means for applying the voice recognition system to the customer service of employees in a physical store and converting the comments made during customer service into text data, means for analyzing the emotions of employees and customers using an emotion recognition system, means for evaluating the customer service response based on the content of the comments and emotions, and means for generating and notifying points for improvement and training suggestions from the evaluation results. This enables fair and efficient evaluation and the generation and notification of specific lesson plans and training suggestions.

[1126] "Evaluation criteria points for each subject and class" are the evaluation criteria points set for each subject and class, and refer to specific evaluation items and standards.

[1127] "Audio data" refers to collected audio information from students and employees, including speech during classes and in physical stores.

[1128] A "voice recognition system" is a system that converts collected voice data into text data.

[1129] "Text data" is character information converted from voice data by a voice recognition system.

[1130] "Evaluation criteria" refers to the items and indicators that serve as the basis for scoring based on text data.

[1131] "Scoring" is the process of quantitatively evaluating text data based on evaluation criteria and expressing it as a number.

[1132] A "database" is a data storage system for centrally storing and managing information such as scored results and comment content.

[1133] "Generative AI" refers to an artificial intelligence model that generates specific advice, such as lesson plans and training suggestions, based on collected data.

[1134] An "emotion recognition system" is a system that analyzes emotions from voice and facial expressions and generates the results as data.

[1135] "Customer service" refers to the contact and service provided by employees to customers in physical stores.

[1136] "Training proposals" are proposals based on the evaluation results for specific improvements and practice methods to improve and develop the skills of employees or students.

[1137] This invention is a system that combines a voice recognition system, generative AI, and an emotion recognition system to efficiently and fairly evaluate the comments and attitudes of students and employees during classes or in physical stores, and provide specific lesson plans and training suggestions.

[1138] System Configuration

[1139] The system roughly consists of the following elements:

[1140] 1. Server

[1141] 2. Terminals (such as devices with microphones installed on students' or employees' desks or in physical stores)

[1142] 3. Web interface accessed by users (teachers and store managers)

[1143] 4. Emotion Recognition System

[1144] Entering and learning evaluation criteria points

[1145] Users (teachers and store managers) use a dedicated web interface to input evaluation criteria points for each lesson or customer service. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "proactiveness," as well as "emotional responsiveness," "emotional consistency," and "emotional proactiveness." These criteria are sent to the server via the terminal and stored in a database.

[1146] The server inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate the comments and emotions of students in class or employees in a physical store based on these criteria.

[1147] Collecting speech and emotion data during classes and in-store

[1148] During classes or in brick-and-mortar stores, the devices capture students' and employees' speech in real time. The collected voice data is sent from the devices to a server. At the same time, the devices also use an emotion recognition system to collect emotional data from voice and facial expressions.

[1149] Speech recognition system for text conversion and emotion recognition

[1150] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," or an employee says, "Hello, customer. Please take your time and enjoy your shopping," the voice is converted into text data. At the same time, the emotion recognition system analyzes emotional information from the voice and facial expressions and generates data such as, "Positive emotions are expressed in the statement."

[1151] Scoring

[1152] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "activeness: 3 points" for each comment and its associated emotion.

[1153] Recording scores, comments, and emotions

[1154] The scoring results are stored in a database on the server. This includes student and employee IDs, comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students and employees made what comments and how they were evaluated.

[1155] Generation and notification of lesson plans and training proposals

[1156] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans for each student and employee training suggestions. For example, if a student's comments lack consistency, the generative AI might advise, "It would be effective to practice expressing your opinions consistently." If an employee's emotions are negative, the AI ​​might suggest, "We recommend training to increase self-esteem in order to increase positive expressions."

[1157] The created lesson plans and suggestions are notified to the user (teacher or store manager), who can then check the evaluation results and suggestions for the relevant student or employee through a web interface.

[1158] Specific examples

[1159] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech consistent?", "Is the student's attitude positive?", and also "Emotional readiness," "Emotional consistency," and "Emotional positivity." When a student speaks during class, a device equipped with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the comment is recognized as "positive."

[1160] In a physical store, if a store clerk says, "Hello, customer. Please take your time and enjoy your shopping," a microphone-equipped device captures the voice and sends it to a server. The voice data is converted into text, and the emotion expressed at the time of the statement is recognized as "positive."

[1161] The generative AI assigns a score based on this data, for example, "Responsiveness to customer service: 5 points," "Consistency of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and teachers and store managers can use the evaluation results and suggestions to plan the next response or training.

[1162] Prompt Sentence Examples

[1163] Customer service rating: Hello, customer. We hope you enjoy your shopping experience. Sentiment rating: [{'label': 'POSITIVE', 'score': 0.99}] Generate improvement suggestions.

[1164] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1165] Step 1:

[1166] Users input evaluation criteria points for each lesson or customer service through a web interface. The input evaluation criteria points include "responsiveness to lesson content," "coherence of speech," "proactiveness," etc. These evaluation criteria points are sent to the server via the terminal and stored in a database. The input data is in text format, and the output is the evaluation criteria points stored in the database.

[1167] Step 2:

[1168] The server inputs the received evaluation criteria points into the AI ​​model, and the AI ​​learns these criteria. Through this learning process, the AI ​​model acquires the ability to score statements made in class or at a physical store based on the evaluation criteria. The input data are the evaluation criteria points, and the output data is the trained AI model.

[1169] Step 3:

[1170] During classes or in physical stores, devices use microphones to capture speech from students or employees in real time. The audio data is sent from the devices to a server. The collected audio data is the input data, and the output data is the audio file sent to the server.

[1171] Step 4:

[1172] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. Specifically, a voice recognition system (such as Google Speech Recognition API) is used. The input data is an audio file, and the output data is text data.

[1173] Step 5:

[1174] The device uses an emotion recognition system to analyze emotional data from voice and facial expressions. Using an emotion recognition system (e.g., Hugging Face emotion analysis model), the device receives voice recognition results and voice and facial expression data as input data and generates analyzed emotional data as output data.

[1175] Step 6:

[1176] The server uses generative AI to score the converted text data and emotion data based on the evaluation criteria. The generative AI model (e.g., GPT-2) performs the process of assigning scores based on the input data. The input data is the converted text data and emotion data, and the output data is the scored data based on the evaluation criteria.

[1177] Step 7:

[1178] The scoring results are stored in a database on the server. This includes the specific student or employee ID, the content of the comments, emotional data, and scores based on each evaluation criterion. The input data is the scoring results, and the output data is the evaluation information stored in the database.

[1179] Step 8:

[1180] The server analyzes the collected score data, comment content, and emotion data, and uses a generative AI to create lesson plans for each student or training suggestions for employees. For example, if a student's comments lack consistency, the generative AI generates advice such as, "It would be effective to practice expressing your opinions consistently." Specific operations involve inputting a prompt into the generative AI model and using the output suggestion. The input data are score data, comment content, and emotion data, and the output data are the generated lesson plans or training suggestions.

[1181] Step 9:

[1182] Finally, the created lesson plans and suggestions are notified to the user. The user can check the evaluation results and suggestions for the relevant students or employees through a web interface. The input data are the generated lesson plans and training suggestions, and the output data are the notified lesson plans and suggestions.

[1183] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1184] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1185] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1186] [Fourth embodiment]

[1187] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1188] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1189] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1190] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1191] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1192] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1193] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1194] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1195] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1196] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1197] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1198] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1199] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1200] This invention is a system that combines a speech recognition system and a generative AI to automatically evaluate students' comments and behaviors during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[1201] System Configuration

[1202] The system is broadly composed of the following elements:

[1203] 1. Server

[1204] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[1205] 3. Web interface accessed by users (teachers)

[1206] Entering and learning evaluation criteria points

[1207] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "coherence of speech," "activeness," etc. These criteria are sent to the server via the terminal and stored in a database.

[1208] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate students' comments in class based on these criteria.

[1209] Collection of audio data during class

[1210] During class, students are placed on their desks with microphones attached to their devices, which capture what they say in real time. The collected audio data is then sent to a server.

[1211] Text conversion using a speech recognition system

[1212] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[1213] Scoring

[1214] The server uses a generative AI to analyze the converted text data and assigns a score based on pre-trained evaluation criteria. Each comment is assigned a score such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[1215] Scores and speech records

[1216] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[1217] Lesson plan generation and notification

[1218] The server analyzes the collected score data and the content of the comments, and uses the generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions more consistently."

[1219] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[1220] Specific examples

[1221] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[1222] This allows teachers to efficiently and fairly evaluate students' comments and attitudes, and with the support of generative AI, they can also strengthen individual instruction.

[1223] The processing flow will be explained below.

[1224] Step 1:

[1225] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," and "activeness."

[1226] Step 2:

[1227] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[1228] Step 3:

[1229] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[1230] Step 4:

[1231] When class begins, a terminal (such as a device with a microphone installed on a student's desk) records what the student says in real time and collects audio data.

[1232] Step 5:

[1233] The collected voice data is sent from the device to a server, which then inputs the received voice data into a voice recognition system and converts the voice data into text data.

[1234] Step 6:

[1235] The server inputs the converted text data into the AI ​​generator, which then scores each comment based on the evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[1236] Step 7:

[1237] The server stores the scored results in a database along with the comment content and student ID, thereby recording which student made what comment and how much they were evaluated.

[1238] Step 8:

[1239] Based on the collected score data and the content of the comments, the server uses generative AI to create lesson plans and suggestions for future development for each student. For example, in response to a student's comments that lack consistency, the server generates advice such as, "It would be effective to practice expressing your opinions more consistently."

[1240] Step 9:

[1241] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[1242] Step 10:

[1243] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[1244] Example 1

[1245] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1246] Conventional educational systems lack the means to evaluate students' comments and behaviors in real time during class and generate lesson plans and suggestions for growth. This makes it difficult for teachers to provide appropriate feedback to each student, and it is difficult to promote student growth efficiently.

[1247] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1248] In this invention, the server includes a means for receiving and storing evaluation criteria points for each subject and lesson, a means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, a means for scoring the converted text data based on the evaluation criteria, a means for saving the scored results and the speech content in a database, a means for notifying teachers of the generated specific lesson plans and suggestions for future growth, a terminal for collecting voice data, and a means for transmitting the data to the server. This allows teachers to efficiently and fairly evaluate students' comments and attitudes and to strengthen individual instruction with the support of the generative AI model.

[1249] "Evaluation criteria points for each subject and class" are the standards used to evaluate students' comments and attitudes, and are evaluation criteria set according to the content of the subject or class.

[1250] A "voice recognition system" is a system for converting voice data collected during classes into text data, and is a technology that analyzes voice and outputs it as text information.

[1251] "Text data" refers to text information of voice data converted using a voice recognition system.

[1252] The "means for scoring based on evaluation criteria" is a method for evaluating the converted text data based on evaluation criteria set in advance and assigning a score.

[1253] "Generative AI" is an artificial intelligence technology that generates appropriate responses and results for specific inputs based on pre-learned data.

[1254] The "database" is an information system for storing and managing information such as evaluation results and scored comments.

[1255] A "lesson plan" is a plan created using generative AI that includes specific teaching guidelines and advice to promote each student's future growth.

[1256] "Suggestions for future growth" are specific advice and improvement plans for each student on how to promote their growth in the future.

[1257] A "terminal" is a device equipped with a microphone that is installed to collect student comments during class and is responsible for transmitting audio data to a server.

[1258] MODE FOR CARRYING OUT THE INVENTION

[1259] This invention is a system that combines a speech recognition system and a generative AI model to automatically evaluate students' comments and behavior during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[1260] System Configuration

[1261] The system is broadly composed of the following elements:

[1262] 1. Server

[1263] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[1264] 3. Web interface accessed by users (teachers)

[1265] Entering and learning evaluation criteria points

[1266] Users (teachers) input evaluation criteria points for each subject and class through a dedicated web interface. These evaluation criteria points include "responsiveness to lesson content," "coherence of speech," and "activeness." These evaluation criteria points are sent to the server via the terminal and stored in a database. The server then inputs the received evaluation criteria points into the generative AI model, and the AI ​​learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments during class based on these criteria.

[1267] Collection of audio data during class

[1268] During class, students are placed on desks equipped with microphones that capture their speech in real time. The collected audio data is temporarily stored on the devices and then sent to a server.

[1269] Text conversion using a speech recognition system

[1270] The server inputs the received voice data into a speech recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data as, "I think the language in this poem is beautiful."

[1271] Scoring

[1272] The server uses a generative AI (e.g., GPT-4 model) to analyze the converted text data and assign scores based on pre-trained evaluation criteria. Each comment is assigned a score, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "proactiveness: 3 points."

[1273] Scores and speech records

[1274] The scored results are stored in a database on the server, which includes the student ID, the content of the comment, and the score based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[1275] Lesson plan generation and notification

[1276] The server analyzes the collected score data and comments, and uses a generative AI to create lesson plans and suggestions for development for each student. For example, if a student's comments lack consistency, the generative AI will generate advice such as, "It would be effective for you to practice expressing your opinions more consistently." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[1277] Specific examples

[1278] For example, in a Japanese class, a teacher inputs evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student proactive?". When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text such as "I think the language in this poem is beautiful," and the generating AI assigns scores such as "Readiness to respond to lesson content: 5 points," "Coherence of speech: 4 points," and "Proactiveness: 3 points." Finally, these scores are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[1279] Example prompts

[1280] Below are some example prompts to input to a generative AI model:

[1281] In your Japanese class, set evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech coherent?", and "Is the student's attitude positive?". Use a generative AI model to analyze students' comments and generate scores and advice.

[1282] Through these steps, the system enables teachers to efficiently and fairly evaluate students' comments and attitudes, and enhance individual instruction with the assistance of generative AI models.

[1283] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1284] Step 1: Enter and save your evaluation criteria points

[1285] The user (teacher) enters the evaluation criteria points for each subject and class (e.g., "responsiveness to class content," "coherence of speech," "proactiveness") through a dedicated web interface. Specifically, the teacher selects the evaluation criteria on the web interface and clicks the "Submit" button.

[1286] Input: Teacher-entered criterion points

[1287] Data processing: Organize the entered evaluation criteria points according to the format

[1288] Output: Data with organized evaluation criteria points

[1289] Specific operation: When the teacher enters the evaluation criteria and presses the submit button, the data is sent to the server and stored in the database.

[1290] Step 2: Collecting audio data

[1291] During class, the terminal (a device with a microphone installed on the student's desk) captures what the student says in real time and temporarily stores it. The collected audio data is then sent from the terminal to a server.

[1292] Input: Audio data captured during class

[1293] Data processing: After temporary storage, organize according to the format

[1294] Output: Organized audio data

[1295] Specific operation: When a student speaks, the device captures the audio and sends the audio data to the server.

[1296] Step 3: Convert audio data to text

[1297] The server inputs the received voice data into a voice recognition system (e.g., Google Cloud Speech-to-Text API) and converts the voice data into text data.

[1298] Input: Audio data sent to the server

[1299] Data processing: Input the voice data into a voice recognition system and convert it into text data.

[1300] Output: Generated text data

[1301] Specific operation: Sends voice data to a voice recognition system and receives textual results.

[1302] Step 4: Analyze and score the text data

[1303] The server inputs the generated text data into a generative AI model (e.g., GPT-4) for analysis. The generative AI then scores each utterance based on pre-trained evaluation criteria.

[1304] Input: Generated text data

[1305] Data processing: Analyze text data and assign scores based on evaluation criteria

[1306] Output: Scored evaluation results

[1307] Specific operation: Text data is input into the generative AI model, and a score is obtained as the analysis result.

[1308] Step 5: Record your scores and comments

[1309] The server stores the scored results in a database, which includes the student ID, the content of the comment, and the score for each criterion.

[1310] Input: Scored evaluation results

[1311] Data processing: Organize and store the scored results in a database

[1312] Output: Saved evaluation results and comments

[1313] Specific operation: After generating the scoring results, they are immediately written to the database.

[1314] Step 6: Generate and notify lesson plans

[1315] The server inputs the collected score data and comments into a generative AI model to create lesson plans and development suggestions for each student. The generated lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the evaluation results and lesson plans through a web interface.

[1316] Input: Score data and comments

[1317] Data processing: Input into generative AI models to generate lesson plans and development suggestions

[1318] Output: Generated lesson plans and suggestions for development

[1319] Specific operation: When a teacher logs in, a notification of a new lesson plan will be displayed and the contents can be viewed on the details page.

[1320] Through these steps, the system can efficiently evaluate students' comments and behavior during class, and with the support of the generative AI model, provide individual instruction efficiently and effectively.

[1321] (Application example 1)

[1322] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1323] Conventional educational systems have struggled to efficiently and fairly evaluate students' comments and behavior during class. Furthermore, there was no way to automatically generate individual lesson plans or suggestions for improvement, placing a heavy burden on teachers. Furthermore, there was no system in place to evaluate passenger conversations in autonomous vehicles and make suggestions for improving driving conditions, making improving the riding experience a challenge.

[1324] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1325] In this invention, the server includes: means for receiving and storing evaluation criteria points for each subject or lesson; means for inputting student voice data into a speech recognition system and converting it into text data; means for scoring the converted text data based on the evaluation criteria; means for storing the scored results and the speech content in a database; means for notifying the teacher of the generated lesson plan and suggestions for future development; means for evaluating in-car speech data and analyzing the quality of conversation and the degree of agreement between topics; and means for collecting passenger feedback and making suggestions for improving driving conditions. This makes it possible to efficiently and fairly evaluate students' comments and behavior during class and automatically generate individual lesson plans and suggestions for development. It also evaluates passenger conversations and makes suggestions for improving driving conditions in an autonomous vehicle, improving the riding experience.

[1326] "Evaluation criteria points" refer to the evaluation criteria set for each class or subject, and are indicators used to score students' comments and attitudes.

[1327] A "speech recognition system" refers to a technology that converts voice data into text data, and is a system used to accurately transcribe collected speech.

[1328] "Text data" refers to character string data converted by a speech recognition system and is used to record students' utterances as text.

[1329] "Scoring" refers to the process of numerically evaluating the converted text data based on evaluation criteria points.

[1330] "Database" refers to the information system for storing and managing the scored results and comments.

[1331] "Generative AI" refers to artificial intelligence that generates and analyzes text based on large datasets, and is used to automate evaluations and recommendations.

[1332] A "lesson plan" refers to a specific educational strategy to support the individual growth of each student, and is automatically generated by generative AI.

[1333] "Feedback" refers to the opinions and reactions that a system collects from users, and is information used to evaluate and improve the system.

[1334] "Driving situation improvement proposals" refer to specific proposals for improving driving quality based on data collected within an autonomous vehicle.

[1335] "Interface" refers to the means by which users access and operate the system, and is used by teachers to check assessment results and lesson plans.

[1336] "Conversation quality" refers to an indicator that evaluates how meaningful and constructive the conversation is.

[1337] "Topic alignment" refers to an indicator that evaluates how closely a conversation matches the set theme or purpose.

[1338] This invention is a system that can be used in both the fields of educational evaluation and autonomous driving. Specifically, it manages evaluation criteria points for each subject and lesson, automatically evaluates students' comments and behavior during class, and uses generative AI to provide teachers with lesson plans and suggestions for growth. It is also a system that evaluates passenger conversations and makes suggestions for improving driving conditions inside autonomous vehicles.

[1339] System Configuration

[1340] The system consists of the following elements:

[1341] 1. Server: The central hardware that processes, stores, and analyzes data.

[1342] 2. Devices: Devices with microphones installed on students' desks, microphones in self-driving cars, etc.

[1343] 3. Users: School teachers and passengers in self-driving cars.

[1344] Program Overview

[1345] 1. Collection of audio data

[1346] A microphone installed on the device captures what students say in class and conversations inside the self-driving car in real time, and sends the audio data to a server.

[1347] 2. Speech Recognition and Text Conversion

[1348] The server uses a speech recognition system (for example, the SpeechRecognition library) to convert the received voice data into text data, which records what students say and what passengers say as text.

[1349] 3. Scoring and Evaluation

[1350] The server analyzes the converted text data and scores it based on pre-trained evaluation criteria. This process uses a generative AI model (e.g., GPT-3.5-turbo) to evaluate the student's comments based on criteria such as "responsiveness to the lesson content," "coherence of speech," and "activeness." Additionally, in the self-driving car, the server evaluates the quality of the conversation and the degree of topic agreement.

[1351] 4. Saving to the database

[1352] The scored results and comments are stored in a database. The database includes student IDs, comments, and evaluation scores based on each evaluation criterion. This allows for the management of evaluation history.

[1353] 5. Generating lesson plans and suggestions

[1354] The server uses AI to create individual lesson plans and development suggestions for each student based on the collected score data and speech content. For example, it can generate advice such as, "It would be effective to practice expressing your opinions consistently." Then, within the self-driving car, it generates driving improvement suggestions based on feedback on the driving situation.

[1355] 6. Notification of Results

[1356] The final lesson plans and driving improvement suggestions are notified to users (teachers and passengers), who can check the evaluation results and suggestions via a dedicated web interface or smartphone app.

[1357] Specific examples

[1358] For example, if a student says, "I think the language in this poem is beautiful" during a Japanese class, the device's microphone captures the speech and sends it to the server. The server then uses a speech recognition system to convert this speech into text data: "I think the language in this poem is beautiful." The generative AI model then scores this text as "Readiness to respond to lesson content: 5 points," "Consistency of speech: 4 points," and "Positiveness: 3 points," and stores the results in a database.

[1359] Prompt Sentence Examples

[1360] "Analyze the following text and rate the quality and consistency of the conversation: 'The scenery is beautiful. Let's take a photo at the next rest stop.'"

[1361] In this way, the present invention provides a specific form for efficiently collecting, analyzing, evaluating, and proposing improvements to data in both educational evaluation and automated driving.

[1362] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1363] Step 1: Collecting audio data

[1364] The device captures students' speech and conversations inside the self-driving car in real time. A microphone installed on the device picks up the audio and sends the audio data to the server. The input is "audio data" and the output is "audio data sent to the server."

[1365] Step 2: Speech recognition and text conversion

[1366] The server uses the SpeechRecognition library to convert the received voice data into text data. The voice data is input into the voice recognition system and the converted text data is obtained. The input is "voice data" and the output is "text data." Specifically, the voice recognition system analyzes the voice sample and generates a corresponding string of characters.

[1367] Step 3: Scoring based on the evaluation criteria

[1368] The server then scores the converted text data based on pre-trained evaluation criteria. Using a generative AI model (e.g., GPT-3.5-turbo), it assigns a score to the given text based on criteria such as "responsiveness to lesson content," "coherence of speech," and "activeness." The input is "text data" and "evaluation criteria points," and the output is "scored data."

[1369] Step 4: Saving to the Database

[1370] The server saves the scoring results and the comments together in a database. The input is the "score data" and "text data," and the output is the "saved data." Specifically, a new record is added to the database, and the record contains the student ID, comments, and evaluation score.

[1371] Step 5: Generate lesson plans and suggestions

[1372] The server uses generative AI to create individual lesson plans and suggestions for development based on the collected score data and utterances. The inputs are "score data" and "text data," and the output is a "generated lesson plan" and "proposal." Specifically, the server provides prompts to the AI ​​model to generate appropriate advice. For example, "Analyze the following text and generate a lesson plan: What the student uttered."

[1373] Step 6: Notification of results

[1374] The user (teacher) checks the generated lesson plans and proposals through a dedicated web interface or smartphone app. The server sends the evaluation results and proposals to the notification system. The inputs are the "generated lesson plans" and "proposals," and the output is the "notified results." Specifically, the results for each student are displayed on the dashboard of the web interface.

[1375] These steps will enable educational assessment and a conversational assessment and feedback system within an autonomous vehicle.

[1376] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1377] This invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and even emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the invention are described below.

[1378] System Configuration

[1379] The system is broadly composed of the following elements:

[1380] 1. Server

[1381] 2. Terminal (e.g., a device with a microphone installed on a student's desk)

[1382] 3. Web interface accessed by users (teachers)

[1383] 4. Emotion Engine

[1384] Entering and learning evaluation criteria points

[1385] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to class content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the terminal and stored in a database.

[1386] The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these criteria. Through this learning process, the AI ​​model can evaluate students' comments and emotions during class based on these criteria.

[1387] Collection of speech and emotion data during class

[1388] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device uses an emotion engine to collect emotional data from the student's voice and facial expressions.

[1389] Speech recognition system for text conversion and emotion recognition

[1390] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful."

[1391] At the same time, the emotion engine analyzes emotional information from voice and students' facial expressions, generating data such as "positive emotions are expressed when speaking."

[1392] Scoring

[1393] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each comment and the emotion it conveys is given a score, such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points."

[1394] Recording scores, comments, and emotions

[1395] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[1396] Lesson plan generation and notification

[1397] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend that you provide instruction to increase self-esteem in order to increase positive comments."

[1398] The created lesson plans and proposals are notified to the user (teacher) after the lesson. The user can check the student's assessment results and lesson plans through a web interface.

[1399] Specific examples

[1400] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[1401] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[1402] The processing flow will be explained below.

[1403] Step 1:

[1404] Users (teachers) access a dedicated web interface and input evaluation criteria points for each subject and class, including "responsiveness to class content," "coherence of speech," "activeness," "emotional responsiveness," "emotional consistency," and "emotional activeness."

[1405] Step 2:

[1406] The terminal transmits the input evaluation criteria points to the server in real time, and the server stores the received evaluation criteria points in a database.

[1407] Step 3:

[1408] The server inputs the saved evaluation criteria points into the AI ​​model, which then learns these criteria to understand the evaluation criteria.

[1409] Step 4:

[1410] When the lesson begins, the device (such as a microphone-equipped device installed on the student's desk) records what the student says in real time and collects audio data. At the same time, the device uses a camera to capture the student's facial expressions and collect emotional information in real time.

[1411] Step 5:

[1412] The device sends the collected voice data and emotion data to a server, which inputs the voice data into a voice recognition system and converts it into text data.

[1413] Step 6:

[1414] The emotion engine analyzes the emotion data sent to the server and generates emotional information about the student's speech. For example, it recognizes emotions such as "positive" or "negative" from the tone of voice and facial expressions.

[1415] Step 7:

[1416] The server inputs the converted text data and the recognized emotional information into the AI ​​generator, which then scores each utterance based on the evaluation criteria. For example, the evaluation scores could be "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "emotional responsiveness: 4 points," "emotional consistency: 5 points," and "emotional activeness: 3 points."

[1417] Step 8:

[1418] The server stores the scored results in a database along with the comments and student IDs, thereby recording which students made what comments and how they were evaluated.

[1419] Step 9:

[1420] The server uses generative AI to create lesson plans and suggestions for future growth for each student based on the collected score data, comment content, and emotional data. For example, if the comments lack consistency or the emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing consistent opinions. We recommend providing instruction to increase self-esteem in order to increase positive comments."

[1421] Step 10:

[1422] After the lesson, the server notifies the user (teacher) of the generated lesson plan and the evaluation results. The user can check the evaluation results and lesson plan of the relevant student through a web interface.

[1423] Step 11:

[1424] The user (teacher) can use the displayed evaluation results and lesson plans to plan the next lesson and develop individual instruction strategies, thereby effectively supporting the growth of each student.

[1425] Example 2

[1426] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1427] Conventional classroom evaluation systems evaluate students based on subjectivity, which creates issues with the fairness and efficiency of evaluation. In addition, it is difficult to evaluate students based on their emotions and create individual lesson plans, making it difficult to support students' overall growth.

[1428] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for collecting and analyzing students' emotional data, means for scoring the converted text data and emotional data based on the evaluation criteria, means for saving the scored results, the content of the comments, and the emotional data in a database, and means for notifying the teacher of the generated specific lesson plan and suggestions for future growth. This makes it possible to fairly and efficiently evaluate students' comments, attitudes, and even emotions during lessons and provide individual lesson plans.

[1429] "Criteria points" are specific criteria established to assess a student's performance for a particular subject or lesson.

[1430] A "voice recognition system" is a technical system that analyzes input voice data and converts it into corresponding text data.

[1431] "Emotional data" is data that indicates the emotional state of students analyzed from their comments and facial expressions, and includes types of emotions such as positive and negative.

[1432] "Generative AI" refers to artificial intelligence technology that automatically generates evaluations and suggestions based on collected data.

[1433] "Scoring" is the process of numerically evaluating students' comments, attitudes, emotions, etc. based on evaluation criteria.

[1434] A "database" is a system for systematically storing and managing collected data.

[1435] A "lesson plan" is a document that outlines specific plans and policies for teachers to effectively teach students.

[1436] "Notifying" refers to the action of informing teachers of the generated lesson plans and evaluation results.

[1437] The present invention is a system that combines a speech recognition system, a generative AI, and an emotion engine to fairly and efficiently evaluate students' comments, attitudes, and emotions during elementary and junior high school classes, and provides teachers with lesson plans and suggestions for future growth. Specific embodiments of the present invention are described below.

[1438] System Configuration

[1439] The system is broadly composed of the following elements:

[1440] 1. Server

[1441] 2. Terminal (device with microphone installed on student's desk)

[1442] 3. Web interface accessed by users (teachers)

[1443] 4. Emotion engine (e.g., Microsoft Azure Emotion API)

[1444] Entering and learning evaluation criteria points

[1445] Users (teachers) use a dedicated web interface to input evaluation criteria points for each subject and class. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "activeness," as well as "emotional responsiveness," "emotional consistency," and "emotional activeness." These criteria are sent to the server via the device and stored in a database. The server then inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. This learning process enables the AI ​​model to evaluate students' comments and emotions during class based on these criteria.

[1446] Collection of speech and emotion data during class

[1447] During class, a device equipped with a microphone installed on each student's desk captures what the student says in real time. The collected voice data is sent from the device to a server. At the same time, the device also uses an emotion engine to collect emotional data from the student's voice and facial expressions. This is done using a microphone device and a facial recognition camera.

[1448] Speech recognition system for text conversion and emotion recognition

[1449] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," the voice is converted into text data: "I think the language in this poem is beautiful." At the same time, the emotion engine analyzes emotional information from the voice and the student's facial expressions, generating data such as "positive emotion is expressed in the statement."

[1450] Scoring

[1451] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria. Each utterance and its associated emotion are given scores such as "responsiveness to lesson content: 5 points," "consistency of speech: 4 points," "activeness: 3 points," "responsiveness to emotions: 4 points," "consistency of emotions: 5 points," and "activeness of emotions: 3 points." The prompt is as follows:

[1452] "Based on the student's comments and emotional data, please calculate this student's grade for the lesson. The evaluation criteria are as follows: responsiveness to lesson content, coherence of speech, assertiveness, emotional responsiveness, emotional coherence, and emotional assertiveness."

[1453] Recording scores, comments, and emotions

[1454] The scoring results are stored in a database on the server. This includes the student ID, the content of the comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students made what comments and how they were evaluated.

[1455] Lesson plan generation and notification

[1456] The server analyzes the collected score data, comment content, and emotional data, and uses a generative AI to create lesson plans and development suggestions for each student. For example, if a student's comments lack consistency or their emotions are negative, the generative AI will generate advice such as, "It would be effective to practice expressing your opinions consistently. We recommend providing instruction to increase self-esteem in order to increase positive comments." The created lesson plans and suggestions are notified to the user (teacher) after the lesson. The user can check the student's evaluation results and lesson plans through a web interface.

[1457] Specific examples

[1458] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the statement relevant to the lesson content?", "Is the speech consistent?", "Is the attitude positive?", and "Emotional readiness," "Emotional consistency," and "Emotional assertiveness." When a student speaks during class, a device with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the statement is recognized as "positive." The generating AI assigns scores as follows: "Readiness to lesson content: 5 points," "Speech consistency: 4 points," "Proactiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," and "Emotional assertiveness: 3 points." Finally, these results are notified to the user, and the teacher plans the next lesson based on the evaluation results and lesson plan.

[1459] This allows teachers to efficiently and fairly evaluate students' comments, attitudes, and even emotions, and with the support of generative AI, can enhance individual instruction.

[1460] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1461] Step 1: Enter and save your evaluation criteria points

[1462] Users (teachers) access a dedicated web interface and enter evaluation criteria points for each subject and class.

[1463] Specific actions: Open a web browser, enter your username and password in the login form to log in, enter your evaluation criteria points in the input form, and click the submit button.

[1464] Input: Subject, lesson, evaluation criteria points (e.g., "Readiness to follow lesson content: 5", "Consistency of speech: 4", "Proactiveness: 3")

[1465] Data processing: Send the evaluation criteria points to the server.

[1466] Output: The server stores the metric points in the database.

[1467] Step 2: Learning the evaluation criteria points

[1468] The server inputs the received evaluation criteria points into the AI ​​model, which then learns from them.

[1469] How it works: The server retrieves the evaluation criteria points from the database and passes the data to the AI ​​model, which then performs the learning process based on the evaluation criteria.

[1470] Input: Evaluation criteria point data

[1471] Data calculation: The AI ​​model analyzes the evaluation criteria points and learns.

[1472] Output: The AI ​​model obtains the ability to make an evaluation based on the evaluation criteria.

[1473] Step 3: Collecting audio and emotion data during class

[1474] The device captures students' comments in real time during class, collecting audio data, and simultaneously collects students' emotional data using an emotion engine.

[1475] How it works: The device's built-in microphone records speech in real time, and the device uses an emotion engine to analyze emotions from the voice data.

[1476] Input: Student speech and facial expression data

[1477] Data processing: Record audio data and perform sentiment analysis.

[1478] Output: Voice data and emotion data are obtained.

[1479] Step 4: Text conversion by speech recognition system

[1480] The server inputs the received voice data into a voice recognition system and converts it into text data.

[1481] Specific operation: The server sends the audio file to the speech recognition API and receives text data as the API response.

[1482] Input: Audio data file

[1483] Data processing: The voice recognition system analyzes the voice data and converts it into text data.

[1484] Output: Text data (e.g., "I think the language in this poem is beautiful.")

[1485] Step 5: Analyze the sentiment data

[1486] The server receives the emotion data analyzed by the emotion engine.

[1487] Specific operation: The emotion engine analyzes voice and facial expression data and sends the emotion data to the server.

[1488] Input: Voice data, facial expression data

[1489] Data processing: Emotion analysis using an emotion engine

[1490] Output: Emotion data (e.g. "positive")

[1491] Step 6: Scoring

[1492] The server uses generative AI to score the converted text data and emotional data based on evaluation criteria.

[1493] Specific operation: The server sends text data and emotion data to the generative AI model and obtains an evaluation score from the model.

[1494] Input: Text data, emotion data, evaluation criteria points

[1495] Data calculation: The generating AI performs the scoring.

[1496] Output: Score evaluation (e.g., "Readiness to content: 5 points," "Spoken consistency: 4 points," "Positiveness: 3 points," "Emotional readiness: 4 points," "Emotional consistency: 5 points," "Emotional positiveness: 3 points")

[1497] Step 7: Record your scores, comments, and feelings

[1498] The server stores the scored results in a database.

[1499] Specific behavior: The server executes an SQL query to insert the evaluation result into the database.

[1500] Input: Scored evaluation, student ID, speech content, emotion data

[1501] Data processing: Processing to save data in the database.

[1502] Output: Recording to database completed

[1503] Step 8: Generate and notify lesson plans

[1504] The server analyzes the collected score data, speech content, and emotional data, and uses generative AI to create lesson plans and suggestions for growth, which are then notified to the user.

[1505] Specific operations: The server analyzes the score data, sends prompts to the generative AI model, obtains the generated lesson plans and suggestions, and sends a request to notify the user of the obtained lesson plans and suggestions.

[1506] Input: Score data, speech content, emotion data

[1507] Data calculation: Generative AI generates lesson plans and suggestions.

[1508] Output: Lesson plans, suggestions, and user notifications

[1509] Through these processing steps, the present invention realizes a system that can effectively and fairly evaluate students' comments and emotions and provide optimal teaching plans.

[1510] (Application example 2)

[1511] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1512] Traditional methods for evaluating students' comments and behavior during class rely on the teacher's subjective judgment and lack objectivity and efficiency. Furthermore, evaluations of store staff's customer service are often subjective, making it difficult to provide appropriate guidance for improving staff skills and training. There is a need to solve these problems and provide fair and efficient evaluations and lesson plans.

[1513] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving and storing evaluation criteria points for each subject and lesson, means for inputting voice data collected from students during lessons into a voice recognition system and converting it into text data, means for scoring the converted text data based on the evaluation criteria, means for saving the scored results and the content of the comments in a database, means for notifying the teacher of the generated specific lesson plans and suggestions for future development, means for applying the voice recognition system to the customer service of employees in a physical store and converting the comments made during customer service into text data, means for analyzing the emotions of employees and customers using an emotion recognition system, means for evaluating the customer service response based on the content of the comments and emotions, and means for generating and notifying points for improvement and training suggestions from the evaluation results. This enables fair and efficient evaluation and the generation and notification of specific lesson plans and training suggestions.

[1514] "Evaluation criteria points for each subject and class" are the evaluation criteria points set for each subject and class, and refer to specific evaluation items and standards.

[1515] "Audio data" refers to collected audio information from students and employees, including speech during classes and in physical stores.

[1516] A "voice recognition system" is a system that converts collected voice data into text data.

[1517] "Text data" is character information converted from voice data by a voice recognition system.

[1518] "Evaluation criteria" refers to the items and indicators that serve as the basis for scoring based on text data.

[1519] "Scoring" is the process of quantitatively evaluating text data based on evaluation criteria and expressing it as a number.

[1520] A "database" is a data storage system for centrally storing and managing information such as scored results and comment content.

[1521] "Generative AI" refers to an artificial intelligence model that generates specific advice, such as lesson plans and training suggestions, based on collected data.

[1522] An "emotion recognition system" is a system that analyzes emotions from voice and facial expressions and generates the results as data.

[1523] "Customer service" refers to the contact and service provided by employees to customers in physical stores.

[1524] "Training proposals" are proposals based on the evaluation results for specific improvements and practice methods to improve and develop the skills of employees or students.

[1525] This invention is a system that combines a voice recognition system, generative AI, and an emotion recognition system to efficiently and fairly evaluate the comments and attitudes of students and employees during classes or in physical stores, and provide specific lesson plans and training suggestions.

[1526] System Configuration

[1527] The system roughly consists of the following elements:

[1528] 1. Server

[1529] 2. Terminals (such as devices with microphones installed on students' or employees' desks or in physical stores)

[1530] 3. Web interface accessed by users (teachers and store managers)

[1531] 4. Emotion Recognition System

[1532] Entering and learning evaluation criteria points

[1533] Users (teachers and store managers) use a dedicated web interface to input evaluation criteria points for each lesson or customer service. These evaluation criteria points include "responsiveness to lesson content," "consistency of speech," "proactiveness," as well as "emotional responsiveness," "emotional consistency," and "emotional proactiveness." These criteria are sent to the server via the terminal and stored in a database.

[1534] The server inputs the received evaluation criteria points into the AI ​​model, which then learns these evaluation criteria. Through this learning process, the AI ​​model can evaluate the comments and emotions of students in class or employees in a physical store based on these criteria.

[1535] Collecting speech and emotion data during classes and in-store

[1536] During classes or in brick-and-mortar stores, the devices capture students' and employees' speech in real time. The collected voice data is sent from the devices to a server. At the same time, the devices also use an emotion recognition system to collect emotional data from voice and facial expressions.

[1537] Speech recognition system for text conversion and emotion recognition

[1538] The server inputs the received voice data into a voice recognition system and converts it into text data. For example, if a student says, "I think the language in this poem is beautiful," or an employee says, "Hello, customer. Please take your time and enjoy your shopping," the voice is converted into text data. At the same time, the emotion recognition system analyzes emotional information from the voice and facial expressions and generates data such as, "Positive emotions are expressed in the statement."

[1539] Scoring

[1540] The server uses a generative AI to score the converted text data and emotional data based on evaluation criteria, such as "responsiveness to lesson content: 5 points," "coherence of speech: 4 points," and "activeness: 3 points" for each comment and its associated emotion.

[1541] Recording scores, comments, and emotions

[1542] The scoring results are stored in a database on the server. This includes student and employee IDs, comments, emotional data, and scores based on each evaluation criterion. This allows for centralized management of which students and employees made what comments and how they were evaluated.

[1543] Generation and notification of lesson plans and training proposals

[1544] The server analyzes the collected score data, comment content, and emotional data, and uses generative AI to create lesson plans for each student and employee training suggestions. For example, if a student's comments lack consistency, the generative AI might advise, "It would be effective to practice expressing your opinions consistently." If an employee's emotions are negative, the AI ​​might suggest, "We recommend training to increase self-esteem in order to increase positive expressions."

[1545] The created lesson plans and suggestions are notified to the user (teacher or store manager), who can then check the evaluation results and suggestions for the relevant student or employee through a web interface.

[1546] Specific examples

[1547] For example, in a Japanese class, a teacher might input evaluation criteria such as "Is the content of the comment relevant to the lesson content?", "Is the speech consistent?", "Is the student's attitude positive?", and also "Emotional readiness," "Emotional consistency," and "Emotional positivity." When a student speaks during class, a device equipped with a microphone captures the audio and sends it to a server. The audio data is converted into text, such as "I think the language in this poem is beautiful," and the emotion expressed at the time of the comment is recognized as "positive."

[1548] In a physical store, if a store clerk says, "Hello, customer. Please take your time and enjoy your shopping," a microphone-equipped device captures the voice and sends it to a server. The voice data is converted into text, and the emotion expressed at the time of the statement is recognized as "positive."

[1549] The generative AI assigns a score based on this data, for example, "Responsiveness to customer service: 5 points," "Consistency of speech: 4 points," and "Proactiveness: 3 points." Finally, these results are notified to the user, and teachers and store managers can use the evaluation results and suggestions to plan the next response or training.

[1550] Prompt Sentence Examples

[1551] Customer service rating: Hello, customer. We hope you enjoy your shopping experience. Sentiment rating: [{'label': 'POSITIVE', 'score': 0.99}] Generate improvement suggestions.

[1552] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1553] Step 1:

[1554] Users input evaluation criteria points for each lesson or customer service through a web interface. The input evaluation criteria points include "responsiveness to lesson content," "coherence of speech," "proactiveness," etc. These evaluation criteria points are sent to the server via the terminal and stored in a database. The input data is in text format, and the output is the evaluation criteria points stored in the database.

[1555] Step 2:

[1556] The server inputs the received evaluation criteria points into the AI ​​model, and the AI ​​learns these criteria. Through this learning process, the AI ​​model acquires the ability to score statements made in class or at a physical store based on the evaluation criteria. The input data are the evaluation criteria points, and the output data is the trained AI model.

[1557] Step 3:

[1558] During classes or in physical stores, devices use microphones to capture speech from students or employees in real time. The audio data is sent from the devices to a server. The collected audio data is the input data, and the output data is the audio file sent to the server.

[1559] Step 4:

[1560] The server inputs the received voice data into a voice recognition system and converts the voice data into text data. Specifically, a voice recognition system (such as Google Speech Recognition API) is used. The input data is an audio file, and the output data is text data.

[1561] Step 5:

[1562] The device uses an emotion recognition system to analyze emotional data from voice and facial expressions. Using an emotion recognition system (e.g., Hugging Face emotion analysis model), the device receives voice recognition results and voice and facial expression data as input data and generates analyzed emotional data as output data.

[1563] Step 6:

[1564] The server uses generative AI to score the converted text data and emotion data based on the evaluation criteria. The generative AI model (e.g., GPT-2) performs the process of assigning scores based on the input data. The input data is the converted text data and emotion data, and the output data is the scored data based on the evaluation criteria.

[1565] Step 7:

[1566] The scoring results are stored in a database on the server. This includes the specific student or employee ID, the content of the comments, emotional data, and scores based on each evaluation criterion. The input data is the scoring results, and the output data is the evaluation information stored in the database.

[1567] Step 8:

[1568] The server analyzes the collected score data, comment content, and emotion data, and uses a generative AI to create lesson plans for each student or training suggestions for employees. For example, if a student's comments lack consistency, the generative AI generates advice such as, "It would be effective to practice expressing your opinions consistently." Specific operations involve inputting a prompt into the generative AI model and using the output suggestion. The input data are score data, comment content, and emotion data, and the output data are the generated lesson plans or training suggestions.

[1569] Step 9:

[1570] Finally, the created lesson plans and suggestions are notified to the user. The user can check the evaluation results and suggestions for the relevant students or employees through a web interface. The input data are the generated lesson plans and training suggestions, and the output data are the notified lesson plans and suggestions.

[1571] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1572] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1573] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1574] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1575] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1576] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1577] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1578] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1579] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1580] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1581] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1582] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1583] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1584] 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.

[1585] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1586] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1587] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1588] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1589] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1590] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist o...

Claims

1. A means to receive and save evaluation criteria points for each subject and class, A means for inputting voice data collected from students during class into a voice recognition system and converting it into text data; means for scoring the converted text data based on evaluation criteria; A means for storing the scored results and comments in a database; A means of informing teachers of the specific lesson plans and suggestions for future development that have been generated; A system including:

2. A means of inputting evaluation criteria points into an AI model and letting it learn; a means for utilizing generative AI to analyze the converted text data; The system of claim 1 , comprising:

3. A means for recording the scoring results of each specific student's comments in a database; Based on the collected score data and comments, we will use generative AI to create individual lesson plans and advice. The system of claim 1 , comprising:

Images