system

Abstract

We provide the system. [Solution] This invention relates to a system that uses a generative AI to analyze training history data, proficiency data, and work data collected for each trainee within an organization, and automatically generates and distributes individually optimized training videos. After training, comprehension tests and feedback are obtained, and the generative AI immediately analyzes them to visualize the training effect and automatically makes suggestions for improvement for the next training session. This significantly reduces the burden of individualizing and updating training materials, while efficiently and uniformly raising the skill level of the entire organization and achieving continuous human resource development.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method executed by at least one processor, the method comprising: a step of receiving a user utterance; a step of adding the user utterance to a prompt including an instruction sentence related to a description of a character of a chatbot; a step of encoding the prompt; and a step of inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Literature

Patent Documents

[0003]

Patent Document 1

Summary of Invention

Problems to be Solved by the Invention

[0004] The problem to be solved by the present invention lies in that, while there is a demand for efficiently and uniformly establishing company-wide common knowledge and skills within an organization, the level of understanding and progress vary greatly for each trainee, and variations occur in the progress and proficiency of training for each site or department, which makes it difficult for training managers and human resource development departments to accurately grasp training effects, and requires enormous labor and time for continuous improvement of teaching materials and individual responses. In addition to the conventional problems, it is difficult to quickly and automatically provide learning content optimized for each trainee, and furthermore, it is not possible to immediately reflect effect measurement after training and feedback for the next training, and as a result, the overall improvement of skills across the entire organization and the efficiency of human resource development cannot be sufficiently realized.

Means for Solving the Problems

[0005] The present invention provides a system comprising: a data collection unit that automatically acquires training history data, proficiency data, and work data for each participant; an analysis unit that analyzes the various data acquired by the data collection unit using a generative AI to determine training content optimized for each participant; and a video generation and distribution unit that automatically generates training videos based on the training content determined by the analysis unit and distributes the videos to each participant. This system comprehensively collects past training history of participants across multiple locations and departments within an organization, proficiency evaluations for each field, work performance, work logs, and other work data. It uses a generative AI to analyze each participant's weak areas, work challenges, and the effectiveness of past training from multiple perspectives, determines individually optimized training content, and then automatically selects and edits necessary training material parts from a training content database to generate customized training videos for each participant, which are then distributed to the participants' terminals. This system addresses variations in understanding and progress among participants while achieving uniform and efficient human resource development throughout the company, significantly reducing the burden on training managers and human resource development departments. [Brief explanation of the drawing]

[0006] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of the data processing device and smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0007] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0008] First, let's explain the terminology used in the following explanation.

[0009] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0010] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0011] In the following embodiments, the labeled storage is one or a plurality of non-volatile storage devices that store various programs, various parameters, and the like. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disk (for example, hard disk), magnetic tape, and the like.

[0012] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F manages communication between a plurality of 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), and the like.

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, may be only B, or may be a combination of A and B. Further, in the present specification, when three or more matters are expressed by being connected with "and / or", the same concept as that for "A and / or B" is applied.

[0014] [First Embodiment]

[0015] FIG. 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0016] As shown in FIG. 1, the data processing system 10 includes a data processing apparatus 12 and a smart device 14. An example of the data processing apparatus 12 is a server.

[0017] 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 the "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, RAM 30, and 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), and the like.

[0018] 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, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, camera 42, and communication I / F 44 are also connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A receives user input via contact of an indicator by detecting the contact of the indicator (for example, a pen, a finger, or the like). The microphone 38B receives voice-based user input by detecting the user's voice. A control unit 46A in the processor 46 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, a specific processing unit 290 acquires the data indicating the user input.

[0020] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0022] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0024] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0026] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0027] (Example 1)

[0028] The flow of a specific process in Example 1 will now be described. Each part of the system described below is implemented by a data processing device 12 and a smart device 14. The data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0029] (Embodiments for carrying out the present invention)

[0030] Embodiments of the present invention will be described in more specific and detail. The system of the present invention consists of a server and a terminal, with the server side realizing a data collection unit, an analysis unit, a video generation and distribution unit, a report generation unit, and an improvement suggestion unit, and the terminal side realizing a learner's learning interface, test taking, feedback input, progress confirmation, etc.

[0031] First, the data collection unit operates on a server and connects to multiple internal systems and external cloud services, such as the organization's human resources management system, business management system, learning management system, sales support system, and project management system, via API integration and batch processing. It automatically and comprehensively collects training history data for each participant (e.g., name of the training attended, date of attendance, number of sessions, test results, certificate acquisition status, instructor evaluation, attendance rate, tardiness / early departure history, etc.), proficiency data (e.g., skill evaluation for each field, self-evaluation, supervisor evaluation, 360-degree evaluation, past promotion exam results, qualification acquisition status, mock exam scores, OJT evaluation, etc.), and business data (e.g., type of assigned work, performance indicators, sales performance, customer satisfaction score, project progress rate, work hours, number of errors, customer interaction history, complaint handling history, goal achievement rate, leadership evaluation, team communication evaluation, etc.). For example, if employee A in the sales department has taken "Basic Sales Training," "Proposal Skills Enhancement Training," and "Closing Techniques Training" over the past three years, scoring 80, 65, and 70 points on the respective tests, while their sales performance includes a below-average closing rate and multiple customer satisfaction surveys mentioning that their explanations are difficult to understand, this information will be stored on the server.

[0032] The analysis unit operates on a server and uses generative AI to analyze various collected data from multiple perspectives and hierarchically. The generative AI is composed of a combination of multiple AI technologies, such as natural language processing models (e.g., large-scale language models), image generation models (e.g., deep learning-based automatic illustration generation), speech synthesis models (e.g., multilingual speech synthesis), video editing AI (e.g., automatic cut editing, automatic subtitle generation), and data mining AI (e.g., clustering, anomaly detection, trend analysis). The analysis unit automatically extracts and analyzes each participant's weak areas (e.g., closing techniques, listening skills, IT security, leadership, time management, complaint handling, presentation skills, etc.), business challenges (e.g., low customer satisfaction, slow project progress management, high error rate, low leadership evaluation, etc.), and the effectiveness of past training (e.g., whether performance improved after the previous training, trends in comprehension test scores, changes in feedback content, changes in OJT evaluation, etc.) using generative AI. For example, if a generative AI analyzes participant A's past test results, work logs, and feedback, and determines that "the understanding of closing techniques is insufficient and there are issues with the clarity of customer explanations," it will instruct the AI ​​to focus on that area in the next training session.

[0033] The video generation and distribution unit operates on a server and, based on the individually optimized training content determined by the analysis unit, automatically selects necessary training material components (e.g., slides, narration audio, demonstration videos, case study videos, animations, diagrams, comprehension check questions, role-playing videos, simulation videos, interview videos, FAQ videos, past best case study videos, etc.) from the training content database and edits and generates videos using generative AI. For example, it can create new illustrations and diagrams with image generation AI, automatically generate narration with speech synthesis AI, and combine each component with video editing AI to output a single training video. Furthermore, it can automatically adjust the difficulty level, level of detail in explanations, examples and phrasing used, narration speed, presence or absence of subtitles, language (if multilingual), video length, and interactive elements (such as branching videos that present quizzes or choices midway through) according to the attributes and comprehension level of the participants. For example, the AI ​​automatically optimizes the content, providing detailed explanations from the basics for younger employees, including numerous application examples and practical case studies for managers, automatically generating narration and subtitles in the local language for overseas branches, and automatically inserting audio guides and sign language videos for people with disabilities.

[0034] The generated training videos are delivered from the video distribution server to the participant's device (PC, tablet, smartphone, wearable device, etc.). Participants view the videos through a dedicated application or web browser on their device. For example, it supports a variety of usage scenarios, such as sales representatives viewing on their smartphones while on the go, employees working from home taking the course on their home computers, field workers viewing field manual videos on their tablets, and employees at overseas locations taking the course with local language subtitles.

[0035] After watching a video, a comprehension test (including multiple-choice questions, written answers, case study responses, interactive quizzes within the video, simulation problems, role-playing evaluations, and practical video submissions) is automatically presented on the device, and participants can answer it on the spot. For example, after a closing techniques training session, participants might be asked to write an answer on "how to propose to a customer in this situation," or after an IT security training session, they might be asked to select from multiple-choice questions the key points for identifying suspicious emails. In addition, a feedback input screen is displayed, where participants can freely write or select from options to input their thoughts on the clarity of the video, the appropriateness of the content, areas they would like to learn more about, areas for improvement in the materials, and how they envision using the material in the field.

[0036] The results of the comprehension test and the feedback are sent from the terminal to the server, where they are immediately analyzed again by a generative AI in the analysis unit. For example, if the generative AI analyzes the test results of participant B and determines that "the correct answer rate for security-related questions is low," "the logical structure of the written answers is weak," and "there are many comments in the feedback that 'the video speed is too fast'," it will instruct the system to automatically incorporate additional explanatory parts and practice problems into the next training video. Furthermore, the generative AI classifies and summarizes the free-response feedback using natural language processing, extracting trends such as "more examples are needed," "the diagrams were easy to understand," and "the narration was difficult to hear," and reflecting these findings in improving the training materials.

[0037] The report generation unit operates on a server and, based on data from all participants aggregated by the analysis unit, automatically generates reports in various formats such as graphs, heatmaps, time-series charts, rankings, radar charts, and scatter plots. These reports cover the progress of the entire organization and each location, proficiency distribution, training effectiveness, progress in understanding, attendance rates, lists of non-attendees, distribution of weaknesses by subject area, comparison of effectiveness for each training material, and trends by participant attribute (by age, job type, location, position, etc.). These reports are displayed in real time on dashboards for training managers and management. For example, it visualizes trends such as "Understanding of closing techniques is generally low at East Japan locations," "Attendance rates for leadership training among managers are low," and "IT literacy among new employees is rapidly improving."

[0038] The Improvement Proposal Department operates on a server, and based on the analysis results from the Analysis Department and Report Generation Department, the Generative AI automatically generates improvement proposals for the next training session and notifies the training staff and the Human Resources Development Department. For example, it proposes specific action plans such as "conduct additional closing technique exercises at all East Japan locations," "recommend individual follow-up training for participants with low comprehension levels," "add new case studies to the general training materials," "adjust narration speed," "increase the number of real-world case studies," "shorten the length of videos," and "strengthen multilingual support." Furthermore, the AI ​​learns from past improvement history and effectiveness measurement results, and continuously optimizes the content of its proposals.

[0039] Thus, the system of the present invention centrally performs data collection, analysis, video generation and distribution, report creation, and improvement suggestions on the server side, while providing interfaces on the terminal side for learners to view videos, take tests, input feedback, and check progress. This simultaneously realizes an optimized learning experience for each learner and improves the efficiency of human resource development across the entire organization. Generative AI plays an essential and central role in all major processes of the system, including data analysis, automatic generation of video content, automatic creation of narration and illustrations, individual optimization, feedback analysis, report creation, and improvement suggestions. For example, by combining and utilizing various generative AI technologies, such as using natural language processing to classify and summarize learners' free-text feedback, image generation to create new teaching material illustrations, speech synthesis to automatically generate multilingual narration, video editing AI to simultaneously generate different versions of videos for each learner attribute, data mining AI to extract trends across the entire organization, and improvement suggestion AI to automatically generate optimal action plans, unprecedented flexibility and automation can be achieved. Furthermore, a major feature of this invention is its ability to automatically provide individually optimized training for all job types, levels, locations, and attributes. For example, it can automatically insert audio guides and sign language videos for people with disabilities, automatically generate local language subtitles and local case studies for overseas locations, deliver manual videos to field workers using AR terminals and wearable devices, and automatically generate management strategy case studies and leadership case studies for managers.

[0040] (System Configuration)

[0041] The system according to this embodiment comprises a data collection unit, an analysis unit, a video generation unit, a distribution unit, a learner interface unit, a comprehension evaluation unit, a feedback collection unit, a report generation unit, and an improvement suggestion unit. The data collection unit automatically and comprehensively collects training history data, proficiency data, and business data for each learner from multiple information sources, such as the organization's human resource management system, learning management system, business management system, sales support system, project management system, and external cloud services. For example, it collects training history data such as the name of the training the learner has previously attended, the date of attendance, the number of times attended, test results, certificate acquisition status, evaluation from the instructor, attendance rate, and tardiness / early departure history; proficiency data such as skill evaluation for each field, self-evaluation, supervisor evaluation, 360-degree evaluation, qualification acquisition status, mock exam scores, and OJT evaluation; and business data such as the type of work assigned, sales performance, customer satisfaction score, project progress rate, working hours, number of errors, customer interaction history, complaint handling history, goal achievement rate, leadership evaluation, and team communication evaluation, using API integration, batch processing, and database queries. The analysis unit feeds various data obtained from the data collection unit into a generative AI to analyze each participant's weak areas, work challenges, and past training effectiveness from multiple perspectives and hierarchically. For example, if a participant in the sales department repeatedly scores low on closing skills tests, an IT participant has a low understanding of security, or a project management participant shows a tendency towards delays in progress management data, the generative AI automatically extracts these trends and determines individually optimized training content. A concrete example of a prompt to be fed into the generative AI is: "Analyze the following participant data, extract areas of low understanding and work challenges, and list the content that should be focused on in the next training session. Participant data: Training history = Sales basic training 80 points, Proposal skills enhancement training 65 points, Closing skills training 70 points, Work data = Below average closing rate, numerous customer surveys indicating that explanations are difficult to understand."

[0042] The video generation unit automatically selects necessary training material components from the training content database based on the individually optimized training content determined by the analysis unit, and edits and generates videos using generative AI. For example, it creates new illustrations and diagrams with image generation AI, automatically generates narration with speech synthesis AI, and outputs a single training video by combining slide materials, demonstration videos, case study videos, animations, diagrams, comprehension check questions, role-playing videos, simulation videos, interview videos, FAQ videos, and past best case study videos with video editing AI. Furthermore, it automatically adjusts the difficulty level, level of detail in explanations, examples and phrasing used, narration speed, presence or absence of subtitles, language, video length, and interactive elements according to the attributes and comprehension level of the participants. For example, the generative AI automatically optimizes the content, such as providing thorough explanations from the basics for young employees, including many applied examples and practical case studies for managers, automatically generating narration and subtitles in the local language for overseas branches, and automatically inserting audio guides and sign language videos for people with disabilities. Specific examples of prompts to be fed into a generative AI include: "Create a video for young employees in the sales department that clearly explains the basics of closing techniques. Combine slides, demonstration footage, diagrams, and narration. The narration should be spoken slowly and using easy-to-understand language, and the examples should focus on situations involving proposals to new customers."

[0043] The distribution department is responsible for delivering training videos generated by the video generation department from the video distribution server to the participants' devices. For example, it supports a variety of devices such as PCs, tablets, smartphones, and wearable devices, enabling participants to watch videos anywhere—at home, in the office, while traveling, or on-site. The distribution department automatically adjusts video quality and file size according to the participant's device environment and communication conditions, and supports various distribution methods such as streaming, download, and offline playback. For example, it enables optimal delivery depending on the usage scenario, such as field workers watching on-site manual videos on tablets, employees at overseas branches taking courses with local language subtitles, or sales representatives watching on their smartphones while traveling.

[0044] The participant interface provides a user interface for participants to watch training videos on their devices, take comprehension tests, enter feedback, and check their progress. For example, the video viewing screen allows for operations such as play, pause, rewind, fast forward, subtitle display, and audio switching. The comprehension test screen allows participants to take tests in various formats, including multiple-choice questions, written questions, case study answers, interactive quizzes within the video, simulation problems, role-playing evaluations, and practical skills video submissions. The feedback input screen allows participants to freely describe or select from options their thoughts on the clarity of the video, the appropriateness of the content, areas they wish to learn more about, areas for improvement in the materials, and how they envision using the training in the field. The progress confirmation screen allows participants to view a list of completed training sessions, test results, proficiency progress, a list of uncompleted training sessions, and goal achievement status in graphs and charts.

[0045] The comprehension evaluation unit receives the results of comprehension tests sent from the learner interface unit on a server and immediately analyzes them using generative AI. For example, the generative AI automatically calculates a variety of evaluation indicators, such as the correct answer rate for multiple-choice questions, the logical structure and keyword frequency of written answers, the practical validity of case studies, the response patterns for simulation problems, the expressiveness and persuasiveness of role-playing videos, and the procedural accuracy of practical skills videos. Furthermore, it also performs correlation analysis with past test results, other learners, and business data to evaluate each learner's strengths and weaknesses in detail. A specific example of a prompt sentence to be fed into the generative AI is: "Evaluate the following written answer and score its logical structure, persuasiveness, and appropriateness of keywords on a scale of 1 to 10. Answer: 'It is important to accurately understand customer needs and explain the proposed content in an easy-to-understand manner.'"

[0046] The feedback collection unit receives feedback sent from the learner interface unit on a server and uses generative AI to classify, summarize, and extract trends through natural language processing. For example, it automatically classifies free-form feedback such as "The video was too fast," "I would like more examples," "The diagrams were easy to understand," and "The narration was difficult to hear," and uses this information to improve the training materials. Furthermore, it extracts learner satisfaction and future learning needs from the feedback and uses this information to optimize the content of the next training session. Specific examples of prompts to be fed into the generative AI include: "Classify the following feedback and extract elements that will be useful for improving the training materials. Feedback: 'The video examples were relevant to the field and easy to understand,' 'The narration was too fast and difficult to hear,' 'There were many diagrams, making it easy to understand.'"

[0047] The report generation unit automatically generates data from all participants obtained from the analysis unit, comprehension evaluation unit, and feedback collection unit, showing progress for the entire organization and each location, proficiency distribution, training effectiveness, comprehension trends, attendance rates, lists of non-participants, weakness distribution by subject area, effectiveness comparisons for each training material, and trends by participant attribute in various formats such as graphs, heatmaps, time-series charts, rankings, radar charts, and scatter plots. This data is displayed in real time on a dashboard for training managers and management. For example, it visualizes trends such as "Overall understanding of closing techniques is low at East Japan locations," "The attendance rate for leadership training among managers is low," and "IT literacy among new employees is rapidly improving." A specific example of a prompt to be fed into the generation AI is, "Based on the following participant data, visualize the comprehension distribution by location using a heatmap and extract areas of weakness."

[0048] The Improvement Proposal Department uses the Report Generation Department's analysis results to automatically generate improvement proposals for the next training session using a generation AI, and notifies training staff and the Human Resources Development Department. For example, it proposes specific action plans such as "conduct additional closing technique exercises across all East Japan locations," "recommend individual follow-up training for participants with low comprehension levels," "add new case studies to the general training materials," "adjust narration speed," "increase the number of real-world case studies," "shorten the length of videos," and "strengthen multilingual support." Furthermore, the AI ​​learns from past improvement history and effectiveness measurement results to continuously optimize the proposals. Specific examples of prompts to be fed into the generation AI include, "Based on the following report data, list five improvement proposals for the next training session. Pay particular attention to areas with low comprehension levels and to approaches for groups with low participation rates."

[0049] Thus, the system according to this embodiment enables the automatic generation and distribution of training videos optimized for each participant through the cooperation of each component, comprehension evaluation, feedback analysis, visualization of progress across the entire organization, and continuous improvement of training materials. By making generative AI essential and central to all processes, it provides unprecedented flexibility and automation to the organization's human resource development.

[0050] (Implementation Steps)

[0051] Step 1: Data Collection In this step, the data collection department automatically and comprehensively collects training history data, proficiency data, and work data for each trainee from multiple sources, including the organization's HR management system, learning management system, business management system, sales support system, project management system, and external cloud services. For example, training history data such as the name of the training the trainee has previously attended, the date of attendance, the number of times the training was attended, test results, certificate acquisition status, instructor evaluations, attendance rate, and tardiness / early departure history; proficiency data such as skill evaluations for each field, self-evaluation, supervisor evaluation, 360-degree evaluation, qualification acquisition status, mock exam scores, and OJT evaluations; and work data such as the type of work assigned, sales performance, customer satisfaction score, project progress rate, working hours, number of errors, customer interaction history, complaint handling history, goal achievement rate, leadership evaluation, and team communication evaluation are collected using API integration, batch processing, and database queries.

[0052] Step 2: Individual analysis and problem identification (using generation AI) The analysis department feeds various data obtained from the data collection department into a generative AI, which then analyzes each participant's weak areas, work challenges, and past training effectiveness from multiple perspectives and hierarchically. For example, if a participant in the sales department repeatedly scores low on closing skills tests, an IT participant has a low understanding of security, or a project management participant shows a tendency towards delays in progress management data, the generative AI automatically extracts these trends and determines individually optimized training content. A concrete example of a prompt to be fed into the generative AI is: "Analyze the following participant data, extract areas of low understanding and work challenges, and list the content that should be focused on in the next training session. Participant data: Training history = Sales basic training 80 points, Proposal skills enhancement training 65 points, Closing skills training 70 points, Work data = Below average closing rate, numerous customer surveys indicating that explanations are difficult to understand."

[0053] Step 3: Automated generation of individually optimized training videos (using AI generation) The video generation unit automatically selects necessary training material components from the training content database based on the individually optimized training content determined by the analysis unit, and edits and generates videos using a generative AI. For example, the image generation AI creates new illustrations and diagrams, the speech synthesis AI automatically generates narration, and the video editing AI combines slide materials, demonstration videos, case study videos, animations, diagrams, comprehension check questions, role-playing videos, simulation videos, interview videos, FAQ videos, and past best case study videos to output a single training video. Furthermore, the difficulty level, level of detail in explanations, examples and phrasing used, narration speed, presence or absence of subtitles, language, video length, and interactive elements are automatically adjusted according to the attributes and comprehension level of the participants. A specific example of a prompt sentence to be fed into the generative AI is: "Create a video for young employees in the sales department that clearly explains the basics of closing techniques. Combine slide materials, demonstration videos, diagrams, and narration. The narration should be spoken slowly, using easy-to-understand language, and the examples should focus on situations involving proposals to new customers."

[0054] Step 4: Distribution and participation of training videos The distribution department distributes training videos generated by the video generation department from the video distribution server to the participants' devices. Participants can view the videos on a variety of devices, including PCs, tablets, smartphones, and wearable devices. The distribution department automatically adjusts the video quality and file size according to the participant's device environment and communication status, and supports various distribution methods such as streaming, download, and offline playback. For example, it enables optimal distribution depending on the usage scenario, such as field workers viewing field manual videos on tablets, employees at overseas branches taking courses with local language subtitles, or sales representatives viewing videos on their smartphones while on the go.

[0055] Step 5: Comprehension test and feedback The learner interface displays comprehension tests on the terminal (multiple-choice questions, written answers, case study responses, interactive quizzes within videos, simulation problems, role-playing evaluations, practical skills video submissions, etc.), and learners can answer them on the spot. Furthermore, a feedback input screen is displayed, where learners can freely write or select options to provide feedback on the clarity of the videos, the appropriateness of the content, areas they would like to learn more about in the future, areas for improvement in the materials, and how they envision applying the material in practice.

[0056] Step 6: Test feedback analysis and individual re-optimization (using generative AI) The comprehension assessment unit and feedback collection unit receive the results of comprehension tests and feedback content sent from the learner interface unit to a server and immediately analyze them using generative AI. For example, the generative AI automatically calculates a variety of evaluation indicators, such as the correct answer rate for multiple-choice questions, the logical structure and keyword frequency of written answers, the practical validity of case studies, response patterns for simulation problems, the expressiveness and persuasiveness of role-playing videos, and the procedural accuracy of practical skills videos. Furthermore, it also performs comparisons with past test results and other learners, and correlation analysis with business data to evaluate each learner's strengths and weaknesses in detail. Specific examples of prompts to be fed into a generative AI include: "Evaluate the following written response and score its logical structure, persuasiveness, and appropriateness of keywords on a scale of 1 to 10. Response: 'It is important to accurately understand customer needs and explain the proposed content in an easy-to-understand manner.'" and "Classify the following feedback and extract elements that can be used to improve the teaching materials. Feedback: 'The video examples were relevant to the field and easy to understand.' 'The narration was spoken too quickly and was difficult to understand.' 'There were many diagrams, which made it easy to understand.'"

[0057] Step 7: Visualize progress and effects across the entire organization and generate reports (using AI generation). The report generation unit automatically generates data from all participants obtained from the analysis unit, comprehension evaluation unit, and feedback collection unit, showing progress for the entire organization and each location, proficiency distribution, training effectiveness, comprehension trends, attendance rates, lists of non-participants, weakness distribution by subject area, effectiveness comparisons for each training material, and trends by participant attribute in various formats such as graphs, heatmaps, time-series charts, rankings, radar charts, and scatter plots. This data is then displayed in real time on a dashboard for training managers and management. An example of a prompt message to be fed into the generation AI is, "Based on the following participant data, visualize the comprehension distribution by location using a heatmap and extract areas of weakness."

[0058] Step 8: Automatic generation and notification of improvement suggestions (using AI generation) Based on the analysis results from the report generation department, the improvement proposal department uses a generation AI to automatically generate improvement proposals for the next training session and notify training staff and the human resources development department. For example, it proposes specific action plans such as "conduct additional closing technique exercises at all East Japan locations," "recommend individual follow-up training for participants with low comprehension levels," "add new case studies to the general training materials," "adjust narration speed," "increase the number of real-world case studies," "shorten the length of videos," and "strengthen multilingual support." Specific examples of prompt sentences to be fed into the generation AI include, "Based on the following report data, list five improvement proposals for the next training session. Pay particular attention to areas with low comprehension levels and to approaches for groups with low participation rates."

[0059] (Specific use cases)

[0060] For example, consider a manufacturing organization with multiple locations nationwide that implements a training program for new employees in its sales department. First, the data collection department automatically retrieves data on each new employee's past training history from the personnel management system and learning management system. This includes attendance records for training courses such as "Basic Sales Training," "Product Knowledge Training," and "Proposal Skills Enhancement Training," as well as the dates of attendance, test results, attendance rates, and instructor evaluations for each training course. Furthermore, the department collects business data from the business management system, such as each new employee's assigned tasks, sales performance, customer interaction history, complaint handling history, goal achievement rate, and customer satisfaction score. In addition, proficiency data such as self-evaluations, supervisor evaluations, 360-degree evaluations, mock exam scores, and on-the-job training (OJT) evaluations are also acquired.

[0061] Next, the analysis unit feeds this data into the generative AI to extract weak areas and work challenges for each new employee. For example, if a new employee's test results in the "Proposal Skills Enhancement Training" are significantly lower than in other training sessions, their sales performance is below average in terms of closing rates, and there are multiple instances of customer surveys indicating that their explanations are difficult to understand, the generative AI will analyze these trends and individually determine training content that focuses on "clear explanations of proposals" and "closing techniques." A concrete example of a prompt to be fed into the generative AI would be: "Analyze the following new employee data, extract areas of low understanding and work challenges, and list the content that should be focused on in the next training session. New employee data: Training history = Sales basic training 80 points, Product knowledge training 85 points, Proposal skills enhancement training 60 points, Work data = Closing rate below average, numerous customer surveys indicating that explanations are difficult to understand."

[0062] The video generation unit automatically selects training material components such as "clear explanation methods for proposals," "closing techniques," "actual sales negotiation examples," and "customer interaction role-playing" from the training content database, based on the individually optimized training content determined by the analysis unit. It then creates new diagrams and illustrations using image generation AI, automatically generates narration using speech synthesis AI, and combines slides, demonstration videos, case study videos, animations, diagrams, and comprehension check questions using video editing AI to produce a single training video. For example, for young employees, it carefully explains the basics, inserts role-playing videos simulating actual sales negotiation situations, uses slow, easy-to-understand narration, and visually explains complex concepts with diagrams and animations. A concrete example of a prompt sentence to be fed into the generation AI is: "Create a video for new employees in the sales department focusing on clear explanation methods for proposals and closing techniques. Combine slides, demonstration videos, diagrams, and narration. The narration should be slow, use easy-to-understand language, and the case studies should focus on proposal situations with new customers."

[0063] The distribution department will deliver the generated training videos to each new employee's device, allowing them to view the videos on their PCs, tablets, smartphones, etc. This will accommodate a variety of usage scenarios, such as new employees working in sales fields viewing the videos on their smartphones while on the go, new employees stationed at branch offices taking the courses on their office computers, or new employees working from home taking the courses on their tablets at home.

[0064] After completing the course, the participant interface team presents a comprehension test, asking participants to select key points from a list of options for clearly explaining a proposal, to write down appropriate responses during a closing, and to select the appropriate response based on actual business negotiation examples. Participants answer these questions on the spot. Furthermore, a feedback input screen allows participants to freely write or select from options to provide feedback on topics such as "clarity of the video," "usefulness of the case studies," "areas they would like to learn more about," and "areas for improvement in the course materials."

[0065] The comprehension evaluation unit and the feedback collection unit feed these test results and feedback content into a generative AI, which automatically analyzes things like "the logical structure and persuasiveness of written answers," "the correct answer rate for multiple-choice questions," and "the most frequent requests and points raised in the feedback," and uses this information to re-optimize the content of the next training session and improve the training materials. Specific examples of prompts to be fed into the generative AI include: "Evaluate the following written answer and score its logical structure, persuasiveness, and appropriateness of keywords on a scale of 1 to 10. Answer: 'It is important to accurately understand customer needs and explain the proposed content clearly.'" and "Classify the following feedback and extract elements that will be useful for improving the training materials. Feedback: 'The video examples were relevant to the field and easy to understand,' 'The narration was too fast and difficult to understand,' 'There were many diagrams, making it easy to understand.'"

[0066] The report generation unit aggregates the comprehension test results, feedback, training status, and work data of all new employees, and visualizes them in real time using graphs, heatmaps, rankings, and time-series charts, such as "comprehension distribution by location," "weakness distribution by subject area," "trends in training rates," "comparison of effectiveness for each training material," and "trends by participant attributes," providing this information to training managers and human resource development departments. A concrete example of a prompt message to be fed into the generation AI is: "Based on the following participant data, visualize the comprehension distribution by location using a heatmap and extract areas of weakness."

[0067] Based on the analysis results from the report generation department, the improvement proposal department uses a generation AI to automatically generate specific action plans, such as "recommend additional role-playing training for new employees with weak explanatory skills regarding proposals," "conduct group training at locations with low understanding of closing techniques," "add new case studies to video materials," "adjust narration speed," "increase the number of field examples," and "shorten the length of videos," and then notifies the training staff. An example of a prompt sentence to be fed into the generation AI is, "Based on the following report data, list five suggestions for improving the next training session. Pay particular attention to approaching areas with low understanding or low participation rates."

[0068] Thus, the system of the present invention can be used in new employee training for nationwide organizations to automatically generate and distribute individually optimized training videos, assess understanding levels, analyze feedback, visualize overall organizational progress, and continuously improve training materials by utilizing generative AI according to the understanding level and work challenges of individual participants.

[0069] (Application Example 1) The flow of a specific process in Application Example 1 will be explained below. Each part of the system described below is implemented by a data processing device 12 and a smart device 14. The data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0070] (Embodiments for carrying out the present invention)

[0071] Embodiments of the present invention will be described in more detail. The system of the present invention is intended for a variety of personnel, including factory workers, robot operators, maintenance personnel, and site supervisors, and is operated by linking a server with various terminals (tablets, personal computers, smartphones, wearable devices, etc.). The server side is equipped with a data collection unit, an analysis unit, a video generation and distribution unit, a report generation unit, and an improvement suggestion unit, while the terminal side is equipped with a worker interface unit, a test taking unit, a feedback input unit, and a progress confirmation unit.

[0072] The data collection unit is connected to multiple information sources, including the robot operation management system, maintenance management system, worker management system, trouble recording system, skill evaluation system, and training management system within the factory, via API integration, batch processing, and real-time data streams. For example, the robot operation management system retrieves information such as which robot model each worker operated and how many times, error history during operation, work time, detailed logs of operation content, safety check status during operation, whether or not abnormal stops occurred, history of emergency stop button usage, and history of operation panel setting changes. For example, if worker A operates a new robot 5 times, records an operation error 3 times, and uses the emergency stop button once, this detailed history will be accumulated. The maintenance management system retrieves information such as the history of periodic maintenance, compliance with maintenance procedures, points raised during maintenance, time required for maintenance work, contents of maintenance records, parts replacement history, and photos and videos of maintenance work. For example, if worker B records two procedural errors during periodic maintenance work, and there are omissions in the maintenance record, or if the wrong part is used during parts replacement, a detailed history of the maintenance work will be accumulated. The trouble recording system retrieves information such as the details of the response when a trouble occurs, the response time, the result of the response, whether or not the trouble recurred, a record of the on-site situation at the time of the trouble, on-site video and audio recordings at the time of the trouble, and the results of the trouble cause analysis. For example, if worker C takes 15 minutes to respond to an initial trouble and the trouble frequently recurs after the response, or if a violation of work procedures is identified from on-site video at the time of the trouble, individual response trends will be recorded. The skills evaluation system retrieves information such as the results of in-house skills tests, evaluations by on-site leaders and supervisors, OJT evaluations, external qualification acquisition status, the progress of skills evaluation, and comments and points raised during evaluation. For example, if worker D receives a B rank evaluation in an in-house test, or if worker E has acquired multiple external qualifications, or if evaluation comments include "high safety awareness" or "thorough adherence to procedures," a multifaceted evaluation of skill levels becomes possible. The training management system retrieves information such as the name of the training attended, the date of attendance, test results, completion certificate acquisition status, frequency of attendance, progress of understanding during training, history of questions asked during training, and results of post-training questionnaires.For example, if worker F has completed safety training and scored 60 points on the comprehension test, or if worker G has taken the same training multiple times, or if there are comments in the post-training survey such as "the on-site video was helpful" or "the narration was difficult to understand," then the training history will be recorded in detail.

[0073] The analysis unit inputs the vast amount of data obtained from the data collection unit into a generative AI, and performs a multifaceted and hierarchical analysis of each worker's operational error trends for each robot model, procedural error trends in maintenance work, delays and challenges in troubleshooting, understanding of safety training, and the status of skill transfer at each site. For example, it can extract the challenges and weaknesses of individual workers, such as when worker A has little experience operating new robots and makes many operational errors, when worker B makes noticeable procedural errors in routine maintenance work, when worker C is slow to respond to initial problems when they occur, or when worker D consistently scores low on safety training comprehension tests. Furthermore, it is also possible to analyze trends at the site or team level, and can extract organizational-wide issues, such as "many operational errors with new robots are occurring at a specific site," "there are many maintenance procedural errors in the night shift team," or "delays in troubleshooting are noticeable on a specific line." The generative AI classifies and summarizes worker feedback and evaluation comments using a natural language processing model, creates new illustrations and diagrams for educational materials using an image generation model, automatically generates multilingual narration and explanations of technical terms using a speech synthesis model, simultaneously generates different versions of videos for each worker attribute using a video editing AI, extracts trends across the entire factory using a data mining AI, and automatically generates optimal action plans using an improvement suggestion AI.

[0074] The video generation and distribution department automatically selects necessary training material components from the training content database based on the individually optimized training content determined by the analysis department, and edits and generates videos using generative AI. For example, it creates new diagrams and illustrations with image generation AI, automatically generates narration with speech synthesis AI, and outputs a single training video by combining slide materials, demonstration videos, case study videos, animations, diagrams, comprehension check questions, role-playing videos, simulation videos, interview videos, FAQ videos, and past best case study videos with video editing AI. For example, for worker A, the video would focus on explaining "basic operation procedures for the new robot," "common mistakes during operation and how to avoid them," "explanation of the control panel interface," and "on-site safety check procedures," while for worker B, it would focus on "standard procedures for regular maintenance," "points where procedural errors are likely to occur," "how to write maintenance records," and "safety measures during maintenance." Furthermore, the difficulty level, level of detail in explanations, examples and phrasing used, narration speed, presence or absence of subtitles, language, video length, and interactive elements (such as branching videos that present quizzes or choices midway through) are automatically adjusted according to the worker's attributes and level of comprehension. For example, the AI ​​automatically optimizes the content by providing detailed explanations from the basics for younger workers, incorporating numerous application examples and practical case studies for veteran workers, automatically generating narration and subtitles in the local language for overseas sites, and automatically inserting audio guides and sign language videos for people with disabilities.

[0075] The generated training videos are delivered from the video distribution server to the workers' devices (such as tablets, smartphones, PCs, and wearable devices at the worksite). Workers view the videos through a dedicated application or web browser on their devices. For example, it supports a variety of usage scenarios, such as workers watching on their tablets during breaks after night shifts, group viewing during pre-work meetings at the worksite, maintenance personnel reviewing operation procedure videos while standing in front of robots at the worksite, and workers at overseas sites taking the course with local language subtitles.

[0076] After watching the video, a comprehension test (including multiple-choice questions, written answers, case study responses, interactive quizzes within the video, simulation problems, role-playing evaluations, and practical video submissions) is automatically presented on the device, and workers can answer it on the spot. For example, the system automatically generates questions directly related to on-site work, such as "selecting the first safety check procedure to be performed on the control panel of a new robot from the given options," "answering in writing about points to note during routine maintenance work," and "arranging the initial response flow in chronological order when a problem occurs." Furthermore, a feedback input screen is displayed, allowing workers to freely describe or select from options their thoughts on the clarity of the video, whether the content is useful in the field, which robot models they would like to learn about in the future, areas for improvement in the teaching materials, and how they envision using the material in the field.

[0077] The results of the comprehension test and the feedback are sent from the terminal to the server, where they are immediately analyzed again by a generative AI in the analysis unit. For example, the generative AI analyzes the worker's test results and determines things like "insufficient understanding of operating procedures," "vague memory of maintenance procedures," "misunderstanding of troubleshooting flow," and "lack of safety awareness," and instructs the system to automatically incorporate additional explanatory parts and practice problems into the next training video. In addition, trends such as "the on-site video was easy to understand," "the explanation of the control panel was detailed and helpful," and "the narration contained too many technical terms and was difficult" are extracted from the feedback and reflected in the improvement of the training materials.

[0078] The report generation unit operates on a server and, based on data from all workers collected by the analysis unit, automatically generates reports in various formats such as graphs, heatmaps, time-series charts, rankings, radar charts, and scatter plots. These reports cover the progress of the entire factory and individual work sites, skill distribution, training effectiveness, progress in understanding, attendance rates, lists of non-attendees, distribution of weaknesses by field, comparison of the effectiveness of different training materials, and trends by worker attribute (by age, job type, work site, team, etc.). These reports are displayed in real time on dashboards for site supervisors and managers. For example, it visualizes trends such as "work sites with many operating errors of new robots," "teams with many maintenance procedure errors," "distribution of workers who are slow to respond to problems," and "lines with low levels of understanding of safety training."

[0079] The improvement suggestion unit operates on a server, and based on the analysis results from the report generation unit, the generation AI automatically generates improvement suggestions for the next training session and notifies site supervisors and managers. For example, it proposes specific action plans such as "recommend additional simulation training for workers who frequently make mistakes operating the new robot," "conduct on-site demonstration training for workers who frequently make mistakes in maintenance procedures," "add role-playing training using past cases for workers who are slow to respond to problems," and "strengthen group safety education at sites with low safety awareness." Furthermore, the AI ​​learns from past improvement history and effectiveness measurement results, and continuously optimizes the content of its suggestions.

[0080] Thus, the system of the present invention centrally performs data collection, analysis, video generation and distribution, report creation, and improvement suggestions on the server side, while providing interfaces on the terminal side for workers to view videos, take tests, input feedback, and check progress. This simultaneously achieves an optimized learning experience for each worker and raises the overall level of skills transfer, safety education, and troubleshooting capabilities in the factory. Generative AI plays an essential and central role in all major processes of the system, including data analysis, automatic generation of video content, automatic creation of narration and illustrations, individual optimization, feedback analysis, report creation, and improvement suggestions. For example, by combining and utilizing various generative AI technologies, such as using natural language processing to classify and summarize workers' free-text feedback, image generation to create new teaching material illustrations, speech synthesis to automatically generate multilingual narration, video editing AI to simultaneously generate different versions of videos for each worker attribute, data mining AI to extract trends across the entire factory, and improvement suggestion AI to automatically generate optimal action plans, it is possible to provide unprecedented flexibility and automation to on-site education and skills transfer in the factory. Furthermore, a major feature of this invention is its ability to automatically provide individually optimized training for all job types, levels, workplaces, and attributes. For example, it can automatically insert audio guides and sign language videos for people with disabilities, automatically generate local language subtitles and local case studies for overseas sites, deliver manual videos to field workers using AR terminals and wearable devices, and automatically generate on-site improvement examples and leadership case studies for managers.

[0081] (System Configuration)

[0082] The system according to this embodiment comprises a data collection unit, an analysis unit, a video generation and distribution unit, a report generation unit, an improvement proposal unit, a worker interface unit, a test evaluation unit, and a feedback analysis unit. The data collection unit is connected to multiple information sources such as a robot operation management system, a maintenance management system, a worker management system, a trouble recording system, a skills evaluation system, and a training management system within the factory via API linkage, batch processing, and real-time data streams, and records the number of robot operations per worker by robot model, error history during operation, work time, detailed logs of operation content, safety check status during operation, presence or absence of abnormal stops, history of emergency stop button usage, history of operation panel setting changes, history of periodic maintenance, compliance with maintenance procedures, points raised during maintenance, time required for maintenance work, and maintenance records. The system comprehensively and automatically collects an extremely wide range of data, including: written content, parts replacement history, photos and videos of maintenance work, details of responses to troubles, response time, response results, recurrence status, on-site situation records at the time of trouble, on-site video and audio recordings at the time of trouble, results of trouble cause analysis, results of in-house skills tests, evaluations by on-site leaders and supervisors, OJT evaluations, external qualification acquisition status, progress in skills evaluation, comments and points raised during evaluations, names of training courses attended, dates of attendance, test results, completion certificate acquisition status, frequency of attendance, progress in understanding during training, history of questions asked during training, and results of post-training questionnaires. For example, if worker A operates a new robot 5 times and records operational errors in 3 of those times, and uses the emergency stop button once; if worker B records procedural errors twice during regular maintenance work and there is a omission in the maintenance record; or if worker C takes 15 minutes to respond to an initial trouble and experiences frequent recurrences after the response, detailed history tailored to the actual situation on-site will be accumulated.

[0083] The analysis unit inputs the vast amount of data obtained from the data collection unit into a generative AI, and performs a multifaceted and hierarchical analysis of each worker's operational error trends for each robot model, procedural error trends in maintenance work, delays and challenges in troubleshooting, understanding of safety training, and the status of skill transfer at each site. For example, it can extract the challenges and weaknesses of individual workers, such as when worker A has little experience operating new robots and makes many operational errors, when worker B makes noticeable procedural errors in routine maintenance work, when worker C is slow to respond to initial problems when they occur, or when worker D consistently scores low on safety training comprehension tests. Furthermore, it is also possible to analyze trends at the site or team level, and can extract organizational-wide issues, such as "many operational errors with new robots are occurring at a specific site," "there are many maintenance procedural errors in the night shift team," or "delays in troubleshooting are noticeable on a specific line." Specific examples of prompts to be fed into the generating AI include: "Analyze the following worker data to extract the robot models with the most frequent operation errors and their causes, the tendency for procedural errors in maintenance work, and the factors contributing to delays in troubleshooting. List the topics that should be emphasized in the next training session. Worker data: Robot operation history = Number of times operating the new robot: 5, Operation errors: 3, Maintenance record = Number of errors in regular maintenance procedures: 2, Troubleshooting history = Average initial response time: 15 minutes, Skill evaluation = Company certification rank B, Training history = Safety education training completed, Comprehension test score: 60 points."

[0084] The video generation and distribution department automatically selects necessary training material components from the training content database based on the individually optimized training content determined by the analysis department, and edits and generates videos using generative AI. For example, it creates new diagrams and illustrations with image generation AI, automatically generates narration with speech synthesis AI, and outputs a single training video by combining slide materials, demonstration videos, case study videos, animations, diagrams, comprehension check questions, role-playing videos, simulation videos, interview videos, FAQ videos, and past best case study videos with video editing AI. For example, for worker A, the video would focus on explaining "basic operation procedures for the new robot," "common mistakes during operation and how to avoid them," "explanation of the control panel interface," and "on-site safety check procedures," while for worker B, it would focus on "standard procedures for regular maintenance," "points where procedural errors are likely to occur," "how to write maintenance records," and "safety measures during maintenance." Furthermore, the difficulty level, level of detail in explanations, examples and phrasing used, narration speed, presence or absence of subtitles, language, video length, and interactive elements (such as branching videos that present quizzes or choices midway through) are automatically adjusted according to the worker's attributes and level of comprehension. Specific examples of prompt text to be fed into the generating AI include: "Create a video that clearly explains the basic operating procedures and how to avoid common mistakes for workers who have little experience operating the new robot and make many mistakes. Combine slides, on-site footage, diagrams, and narration, with the narration spoken slowly and including on-site safety check points."

[0085] The report generation unit uses data from all workers collected by the analysis unit to automatically generate reports in various formats such as graphs, heatmaps, time-series charts, rankings, radar charts, and scatter plots. These reports cover the overall progress of the factory and individual work sites, skill distribution, training effectiveness, progress in understanding, attendance rates, lists of non-attendees, distribution of weaknesses by field, comparison of effectiveness of different training materials, and trends by worker attribute (by age, job type, work site, team, etc.). These reports are displayed in real time on dashboards for site supervisors and managers. For example, it visualizes trends such as "work sites with many operating errors of new robots," "teams with many maintenance procedure errors," "distribution of workers who are slow to respond to problems," and "lines with low levels of understanding of safety training." A concrete example of a prompt message to be fed into the generation AI is, "Based on the following worker data, visualize the understanding distribution by robot model using a heatmap and extract the models and workers with many operating errors."

[0086] The Improvement Proposal Department uses the Report Generation Department's analysis results to automatically generate improvement proposals for the next training session using a generation AI, and notifies site supervisors and managers. For example, it proposes specific action plans such as "Recommend additional simulation training for workers who frequently make mistakes operating the new robot," "Implement on-site demonstration training for workers who frequently make mistakes in maintenance procedures," "Add role-playing training using past cases for workers who are slow to respond to problems," and "Strengthen group safety education at sites with low safety awareness." Furthermore, the AI ​​learns from past improvement history and effectiveness measurement results to continuously optimize the proposals. An example of a prompt message to be fed into the generation AI is, "Based on the following report data, list three additional training proposals for workers who frequently make operating errors. Pay particular emphasis on improving practical response skills on-site."

[0087] The worker interface section provides a user interface for workers to watch training videos on their terminals, take comprehension tests, input feedback, and check their progress. For example, the video viewing screen allows operations such as play, pause, rewind, fast forward, subtitle display, and audio switching. The comprehension test screen allows workers to take tests in various formats, including multiple-choice questions, written questions, case study answers, interactive quizzes within the video, simulation problems, role-playing evaluations, and practical skills video submissions. The feedback input screen allows workers to input information on the clarity of the video, the appropriateness of the content, areas they would like to learn in the future, areas for improvement in the materials, and how they envision using the materials in the field, either in free-form text or multiple-choice format. The progress confirmation screen allows workers to view a list of completed training, test results, proficiency progress, a list of uncompleted training, and goal achievement status in graphs and charts.

[0088] The test evaluation unit receives the results of comprehension tests transmitted from the worker interface unit to a server and immediately analyzes them using generative AI. For example, the generative AI automatically calculates a variety of evaluation indicators, such as the correct answer rate for multiple-choice questions, the logical structure and keyword frequency of written answers, the practical validity of case studies, the response patterns for simulation problems, the expressiveness and persuasiveness of role-playing videos, and the procedural accuracy of practical skills videos. Furthermore, it performs comparisons with past test results and other workers, and correlation analysis with work data to evaluate the strengths and weaknesses of each worker in detail. A concrete example of a prompt sentence to be fed into the generative AI is: "Evaluate the following written answer and score it on a scale of 1 to 10 for accuracy of procedure, safety awareness, and on-site applicability. Answer: 'Before operating the robot, check the location of the emergency stop button and ensure the safety of the surrounding area before starting panel operation.'"

[0089] The Feedback Analysis Unit receives feedback sent from the Worker Interface Unit on a server and uses generative AI to classify, summarize, and extract trends through natural language processing. For example, it automatically classifies free-form feedback such as "The on-site video was easy to understand," "The explanation of the control panel was detailed and helpful," and "The narration contained a lot of technical jargon and was difficult to understand," and uses this information to improve training materials. Furthermore, it extracts worker satisfaction and future learning needs from the feedback content and uses this to optimize the content of the next training session. A concrete example of a prompt sentence to be fed into the generative AI is: "Classify the following feedback and extract elements that will be useful for improving training materials. Feedback: 'The on-site video was easy to understand,' 'The explanation of the control panel was detailed and helpful,' 'The narration contained a lot of technical jargon and was difficult to understand.'"

[0090] Thus, the system according to this embodiment enables the automatic generation and distribution of training videos optimized for each worker through the cooperation of each component, comprehension evaluation, feedback analysis, visualization of progress across the entire factory, and continuous improvement of training materials. By making generative AI essential and central to all processes, it provides unprecedented flexibility and automation to on-site training and skills transfer in factories.

[0091] (Implementation Steps)

[0092] Step 1: Data Collection The data collection unit connects to multiple information sources within the factory, including robot operation management systems, maintenance management systems, worker management systems, trouble recording systems, skill evaluation systems, and training management systems, via API integration, batch processing, and real-time data streams. It collects data such as the number of robot operations per worker for each model, error history during operation, work time, detailed logs of operation content, safety check status during operation, presence or absence of abnormal stops, history of emergency stop button usage, history of operation panel setting changes, history of periodic maintenance, compliance with maintenance procedures, points raised during maintenance, time required for maintenance work, and maintenance records. The system comprehensively and automatically collects an extremely wide range of data, including: content of the system, parts replacement history, photos and videos of maintenance work, details of how problems were handled, response time, response results, whether or not the problem recurred, records of the situation at the time of the problem, on-site video and audio recordings of the time of the problem, results of the problem cause analysis, results of in-house skills tests, evaluations by on-site leaders and supervisors, OJT evaluations, status of external qualifications obtained, trends in skills evaluation, comments and points raised during evaluations, names of training courses attended, dates of attendance, test results, status of obtaining completion certificates, frequency of attendance, progress in understanding during training, history of questions asked during training, and results of post-training questionnaires.

[0093] Step 2: Identifying and analyzing individual issues (using generation AI) The analysis unit inputs the vast amount of data obtained from the data collection unit into the generation AI, which then performs a multifaceted and hierarchical analysis of each worker's operational error tendencies for each robot model, procedural error tendencies in maintenance work, delays and challenges in troubleshooting, understanding of safety training, and the status of skill transfer at each site. For example, it extracts the challenges and weaknesses of each individual worker, such as when worker A has little experience operating new robots and makes many operational errors, worker B makes noticeable procedural errors in routine maintenance work, worker C is slow to respond to initial problems when they occur, or worker D consistently scores low on safety training comprehension tests. Specific examples of prompts to be fed into the generating AI include: "Analyze the following worker data to extract the robot models with the most frequent operation errors and their causes, the tendency for procedural errors in maintenance work, and the factors contributing to delays in troubleshooting. List the topics that should be emphasized in the next training session. Worker data: Robot operation history = Number of times operating the new robot: 5, Operation errors: 3, Maintenance record = Number of errors in regular maintenance procedures: 2, Troubleshooting history = Average initial response time: 15 minutes, Skill evaluation = Company certification rank B, Training history = Safety education training completed, Comprehension test score: 60 points."

[0094] Step 3: Automated generation of individually optimized training videos (using AI generation) The video generation and distribution department automatically selects necessary training material components from the training content database based on the individually optimized training content determined by the analysis department, and edits and generates videos using generative AI. For example, it creates new diagrams and illustrations with image generation AI, automatically generates narration with speech synthesis AI, and outputs a single training video by combining slide materials, demonstration videos, case study videos, animations, diagrams, comprehension check questions, role-playing videos, simulation videos, interview videos, FAQ videos, and past best case study videos with video editing AI. The difficulty level, level of detail in explanations, examples and phrasing used, narration speed, presence or absence of subtitles, language, video length, and interactive elements are automatically adjusted according to the attributes and comprehension level of the workers. Specific examples of prompt text to be fed into the generation AI include: "Create a video that clearly explains the basic operation procedures and how to avoid common mistakes for workers with little experience operating the new robot and who make many operating errors. Combine slide materials, on-site footage, diagrams, and narration, with the narration spoken slowly and including on-site safety check points."

[0095] Step 4: Distribution and viewing of training videos The video generation and distribution department creates training videos which are then distributed to workers' devices (tablets, smartphones, PCs, wearable devices, etc.) allowing workers to watch the videos at any time and place they choose, such as at the worksite, in the break room, or at home. For example, it supports a variety of usage scenarios, such as workers watching on their tablets during breaks after night shifts, group viewing during pre-work meetings at the worksite, maintenance personnel reviewing operation procedure videos while standing in front of robots at the worksite, or workers at overseas sites receiving training with local language subtitles.

[0096] Step 5: Comprehension test and feedback The worker interface unit presents comprehension tests on the terminal (multiple-choice questions, written answers, case study responses, interactive quizzes in videos, simulation problems, role-playing evaluations, practical skills video submissions, etc.), and workers can answer them on the spot. Furthermore, a feedback input screen is displayed, where workers can freely describe or select from options their thoughts on the clarity of the video, whether the content is useful in the field, which robot models they would like to learn about in the future, areas for improvement in the teaching materials, and how they envision using the material in the field.

[0097] Step 6: Test feedback analysis and individual re-optimization (using generative AI) The Test Evaluation Department and Feedback Analysis Department receive the results of comprehension tests and feedback content transmitted from the Worker Interface Department to a server and immediately analyze them using generative AI. For example, the generative AI automatically calculates a variety of evaluation indicators, such as the correct answer rate for multiple-choice questions, the accuracy of the procedures in written answers, safety awareness, on-site applicability, the practical validity of case studies, response patterns for simulation problems, the expressiveness and persuasiveness of role-playing videos, and the accuracy of the procedures in practical skills videos. Specific examples of prompts to be fed into the generative AI include: "Evaluate the following written answer and score the accuracy of the procedure, safety awareness, and on-site applicability on a scale of 1 to 10. Answer: 'Before operating the robot, check the location of the emergency stop button and ensure the safety of the surrounding area before starting panel operation.'" and "Classify the following feedback and extract elements that will be useful for improving the training materials. Feedback: 'The on-site video was easy to understand,' 'The explanation of the operation panel was detailed and helpful,' 'The narration contained a lot of technical jargon and was difficult to understand.'"

[0098] Step 7: Visualization of overall factory progress and effectiveness, and generation of reports (using AI generation). The report generation unit automatically generates data on the progress of the entire factory and individual sites, skill distribution, training effectiveness, progress in understanding, attendance rates, lists of non-attendees, distribution of weaknesses by subject area, comparison of effectiveness of different training materials, and trends by worker attribute, in various formats such as graphs, heatmaps, time-series charts, rankings, radar charts, and scatter plots, based on data from all workers obtained from the analysis unit, test evaluation unit, and feedback analysis unit. This data is then displayed in real time on a dashboard for site supervisors and managers. An example of a prompt message to be fed into the generation AI is, "Based on the following worker data, visualize the understanding distribution by robot model using a heatmap and extract the models and workers with the most operating errors."

[0099] Step 8: Automatic generation and notification of improvement suggestions (using AI generation) Based on the analysis results from the report generation department, the improvement proposal department uses a generation AI to automatically generate improvement proposals for the next training session and notify site supervisors and managers. For example, it proposes specific action plans such as "Recommend additional simulation training for workers who frequently make mistakes operating the new robot," "Implement on-site demonstration training for workers who frequently make mistakes in maintenance procedures," "Add role-playing training using past cases for workers who are slow to respond to problems," and "Strengthen group safety education at sites with low safety awareness." An example of a prompt message to be fed into the generation AI is, "Based on the following report data, list three additional training proposals for workers who frequently make operating errors. Pay particular emphasis on improving practical response skills on-site."

[0100] (Specific use cases)

[0101] For example, consider a manufacturing plant that produces a wide variety of products in small quantities, where multiple types of industrial robots are in operation. As new robots are introduced into the plant, there is a diverse workforce including factory workers, robot operators, maintenance personnel, and site supervisors, each with different robot models, tasks, years of experience, past troubleshooting experience, maintenance history, skill assessments, and training history. On-site challenges arise, such as production stoppages due to operating errors of newly introduced robots, procedural errors in regular maintenance, delays in initial response to problems, uneven skill transfer among workers, and variations in safety awareness and understanding among workers.

[0102] The system of the present invention first has a data collection unit that automatically collects data for each worker from a robot operation management system, maintenance management system, worker management system, trouble recording system, skills evaluation system, training management system, etc., including robot operation history (which model was operated and how many times, history of errors that occurred during operation, work time, detailed log of operation content), trouble response history (how troubles were dealt with and how, response time, response result, whether it recurred), maintenance records (history of regular maintenance, compliance with maintenance procedures, points raised during maintenance, time required for maintenance work, parts replacement history), skills evaluation (results of in-house skills tests, evaluations by on-site leaders and supervisors, OJT evaluations, status of external qualifications obtained, comments at the time of evaluation), and past training attendance history (name of training attended, date of attendance, test results, status of obtaining completion certificates, frequency of attendance, results of post-training questionnaires). For example, detailed data is accumulated such as: for worker A, the number of times he operated the new robot was 5, with 3 operational errors, one of which involved using the emergency stop button; for worker B, there were 2 procedural errors during routine maintenance work and 1 omission in recording maintenance records; for worker C, the average initial response time when a problem occurred was 15 minutes, and recurrence was frequent; and for worker D, he had completed safety education training and scored 60 points on the comprehension test.

[0103] The analysis unit inputs this data into a generative AI to perform a multifaceted and hierarchical analysis of each worker's operational error tendencies for each robot model, procedural error tendencies in maintenance work, delays and challenges in troubleshooting, understanding of safety training, and the status of skill transfer at each site. For example, it extracts the challenges and weaknesses of each individual worker, such as when worker A has little experience operating new robots and makes many operational errors, worker B makes noticeable procedural errors in routine maintenance work, worker C is slow to respond to initial problems when they occur, or worker D consistently scores low on safety training comprehension tests. Specific examples of prompts to be fed into the generating AI include: "Analyze the following worker data to extract the robot models with the most frequent operation errors and their causes, the tendency for procedural errors in maintenance work, and the factors contributing to delays in troubleshooting. List the topics that should be emphasized in the next training session. Worker data: Robot operation history = Number of times operating the new robot: 5, Operation errors: 3, Maintenance record = Number of errors in regular maintenance procedures: 2, Troubleshooting history = Average initial response time: 15 minutes, Skill evaluation = Company certification rank B, Training history = Safety education training completed, Comprehension test score: 60 points."

[0104] The video generation and distribution department automatically selects training material components based on the individually optimized training content determined by the analysis department. These components include, for example, "Basic operation procedures for the new robot," "Common operating errors and how to avoid them," "Explanation of the control panel interface," "On-site safety check procedures," "Standard procedures for regular maintenance," "Points where procedural errors are likely to occur," "How to write maintenance records," "Safety measures during maintenance," "Initial response flow in case of trouble," "Past trouble cases and solutions," "Practical role-playing videos in the field," and "Simulations using actual field footage." The image generation AI creates new diagrams and illustrations, the speech synthesis AI generates narration, and the video editing AI combines slide materials, demonstration videos, animations, and comprehension check questions to generate customized training videos for each worker. A specific example of a prompt message to be fed into the generation AI is: "Create a video that clearly explains the basic operation procedures and how to avoid common errors for workers with little experience operating the new robot and who make many operating errors. Combine slide materials, field footage, diagrams, and narration. The narration should be spoken slowly, and on-site safety check points should also be included."

[0105] Workers can watch videos at any location and time they choose, such as on-site, in break rooms, or at home. After watching the videos, they are presented with comprehension tests (multiple-choice questions, written answers, case study responses, interactive quizzes within the videos, simulation problems, role-playing evaluations, practical skills video submissions, etc.). For example, questions directly related to on-site work are automatically generated, such as "selecting the first safety check procedure to be performed on the control panel of a new robot from the given options," "answering in writing about points to note during routine maintenance work," and "arranging the initial response flow in chronological order when a problem occurs." Furthermore, a feedback input screen allows users to freely write or select from options to provide feedback on topics such as "clarity of the video," "whether the content is useful on-site," "robot models they would like to learn about in the future," "areas for improvement in the teaching materials," and "how they envision using the material on-site."

[0106] The Test Evaluation Department and the Feedback Analysis Department feed these test results and feedback into a generative AI, which automatically analyzes things like "the accuracy of the procedures, safety awareness, and on-site applicability of the written answers," "the correct answer rate for multiple-choice questions," and "the most frequent requests and points raised in the feedback," and uses this information to re-optimize the content of the next training and improve the training materials. Specific examples of prompts to be fed into the generative AI include: "Evaluate the following written answer and score the accuracy of the procedure, safety awareness, and on-site applicability on a scale of 1 to 10. Answer: 'Before operating the robot, check the location of the emergency stop button and ensure the safety of the surroundings before starting to operate the panel.'" and "Classify the following feedback and extract elements that will be useful for improving the training materials. Feedback: 'The on-site video was easy to understand,' 'The explanation of the operation panel was detailed and helpful,' 'The narration contained a lot of technical jargon and was difficult to understand.'"

[0107] The report generation unit aggregates the results of comprehension tests, feedback, training status, and operation, maintenance, and troubleshooting history of all workers. It then visualizes "comprehension distribution by robot model," "skill evaluation trends by worker," "procedure compliance rate for maintenance work," "distribution of initial response time for troubleshooting," "comparison of effectiveness of training materials," and "safety awareness trends by site" using graphs, heatmaps, rankings, and time-series charts, and provides this information to site supervisors and managers in real time. A concrete example of a prompt message to be fed into the generation AI is: "Based on the following worker data, visualize the comprehension distribution by robot model using a heatmap, and extract the models and workers with the most operational errors."

[0108] Based on the analysis results from the report generation department, the improvement proposal department uses a generation AI to automatically generate specific action plans, such as "recommend additional simulation training for workers who frequently make mistakes operating the new robot," "conduct on-site demonstration training for workers who frequently make mistakes in maintenance procedures," "add role-playing training using past cases for workers who are slow to respond to problems," and "strengthen group safety education at sites with low safety awareness," and then notify site supervisors and managers. An example of a prompt message to be fed into the generation AI is, "Based on the following report data, list three suggestions for additional training for workers who frequently make operating errors. Pay particular emphasis on improving practical response skills at the site."

[0109] Thus, the system of the present invention automatically generates and delivers individually optimized training using generative AI, tailored to the specific challenges of each worksite and the skills and understanding of each worker. This enables efficient and continuous improvement of skills transfer, safety education, and troubleshooting capabilities. The system flexibly responds to diverse challenges at worksites, significantly contributing to the prevention of human errors and troubles, and the enhancement of on-site capabilities.

[0110] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0111] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0112] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0113] For example, the data collection unit is implemented by the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the smart device 14 and processes it with the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the output device 40 of the smart device 14 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the devices and control units is not limited to the examples described above and can be modified in various ways.

[0114] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0115] [Second Embodiment]

[0116] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0117] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0118] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0119] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and communication interface 44 are also connected to the bus 52.

[0120] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0121] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0122] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0123] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0124] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0125] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0126] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the identification processing unit 290 using these models.

[0127] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is realized by the data processing device 12 and the smart glasses 214. 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".

[0128] (Example 1)

[0129] Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.

[0130] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.

[0131] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0132] Furthermore, the processing by the data processing system 10 described above is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0133] For example, the data collection unit is implemented by the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the smart glasses 214 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the speaker 240 of the smart glasses 214 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0134] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0135] [Third Embodiment]

[0136] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0137] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0138] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0139] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, display 343, and communication interface 44 are also connected to the bus 52.

[0140] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0141] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0142] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0143] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0144] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0145] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0146] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0147] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is implemented by the data processing device 12 and the headset terminal 314. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0148] (Example 1) Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.

[0149] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.

[0150] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0151] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0152] For example, the data collection unit is implemented by the control unit 46A of the headset terminal 314 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the headset terminal 314 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the speaker 240 and display 343 of the headset terminal 314 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0153] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0154] [Fourth Embodiment]

[0155] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0156] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0157] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0158] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, controlled object 443, and communication interface 44 are also connected to the bus 52.

[0159] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0160] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0161] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0162] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0163] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0164] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0165] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0166] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0167] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is implemented by the data processing device 12 and the robot 414. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0168] (Example 1) Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.

[0169] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.

[0170] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0171] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or external devices, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or external devices.

[0172] For example, the data collection unit is implemented by the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the robot 414 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using a generation AI. For example, the serving unit is implemented by the speaker 240 and controlled object 443 of the robot 414 and the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0173] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0174] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0175] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0176] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0177] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0178] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0179] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0180] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0181] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0182] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0183] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0184] 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.

[0185] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0186] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using this memory.

[0187] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0188] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0189] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0190] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0191] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0192] Furthermore, the following additional information is disclosed regarding the above explanation.

[0193] (Claim 1)

[0194] The system comprises a data collection unit that acquires robot operation history, troubleshooting history, maintenance records, skill evaluations, and training history for each worker in the factory; an analysis unit that analyzes the various data acquired by the data collection unit using a generation AI to determine training content optimized for each worker; and a video generation and distribution unit that automatically generates training videos based on the training content determined by the analysis unit and distributes the videos to each worker, wherein the data collection unit acquires robot operation history by model for each worker, operation error history, work time, detailed logs of operation content, troubleshooting history from multiple robot management systems, maintenance management systems, worker management systems, troubleshooting systems, etc. within the factory, and The system comprehensively collects information such as content, response time, results, recurrence status, regular maintenance implementation history, procedure compliance status, maintenance issues, in-house skills test results, OJT evaluations, external qualification acquisition status, names of training courses attended, dates attended, test results, completion certificate acquisition status, and frequency of attendance. The analysis unit uses a generative AI to analyze from multiple angles each worker's robot type operation error tendencies, procedure error tendencies in maintenance work, delays and challenges in troubleshooting, understanding of safety education, and skill transfer status at each site. Based on this analysis, the system determines individually optimized training content for each worker. The video generation and distribution unit automatically selects and edits necessary training material parts from a training content database to generate customized training videos for each worker and distributes them to the worker's terminal. (Claim 2)

[0195] The system according to claim 1, characterized in that, after a worker watches a training video distributed by the video generation and distribution unit, it automatically presents a comprehension test and feedback, the analysis unit immediately analyzes the comprehension test results and feedback content obtained from the worker using a generation AI to evaluate each worker's level of comprehension and the effectiveness of the training materials, aggregates data from all workers to visualize the progress, skill distribution, and training effectiveness for the entire factory and each work site, and further automatically generates suggestions for improvement for the next training and notifies the manager. (Claim 3)

[0196] The system according to claim 1, characterized in that the analysis unit uses a generative AI to comprehensively analyze the results of comprehension tests and feedback content for each worker, robot operation history, troubleshooting history, maintenance records, skill evaluations, past training history, etc., and automatically generates and distributes additional individual follow-up training videos for workers with low comprehension levels or insufficient troubleshooting skills, and also automatically updates and adds to the content of training materials for the entire factory to raise the overall skill level and improve safety awareness. [Explanation of symbols]

[0197] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. The system comprises a data collection unit that acquires training history data, proficiency data, and work data for each participant; an analysis unit that uses generative AI to analyze the various data acquired by the data collection unit and determine training content optimized for each participant; and a video generation and distribution unit that automatically generates training videos based on the training content determined by the analysis unit and distributes the videos to each participant. The system is characterized in that the data collection unit comprehensively collects past training history of participants across multiple locations and departments within the organization, proficiency evaluations for each field, and work data such as work performance and work logs; the analysis unit uses generative AI to comprehensively analyze each participant's weak areas, work challenges, and past training effectiveness, and individually determines the optimal training content for each participant; and the video generation and distribution unit automatically selects and edits necessary training material parts from a training content database to generate customized training videos for each participant and distributes them to the participant's terminal.

2. The system according to claim 1, characterized in that, after a participant watches a training video distributed by the video generation and distribution unit, it automatically presents a comprehension test and feedback, the analysis unit immediately analyzes the comprehension test results and feedback content obtained from the participant using a generative AI to evaluate each participant's level of comprehension and the effectiveness of the training materials, aggregates data from all participants to visualize the progress and proficiency distribution for the entire organization and each location, and further automatically generates suggestions for improvement for the next training session and notifies the training manager.

3. The system according to claim 1, characterized in that the analysis unit uses a generative AI to comprehensively analyze each participant's comprehension test results, feedback content, work data, past training history, etc., and automatically generates and distributes additional individual follow-up training videos for participants who have low levels of comprehension or work challenges, and also automatically updates and adds to the overall training materials to raise the skill level of the entire organization.

Images