System

Abstract

A system is provided.SOLUTION: Means for capturing audio and video for a user to initiate a presentation, means for transmitting the captured audio and video to a server, means for converting the audio to text in real-time at the server, and means for analyzing the video at the server; A system comprising: means for identifying facial expressions, gaze, and gestures of a user; means for generating questions relevant to the AI based on the analyzed information; means for transmitting the generated questions to terminals for display to the user; means for recapturing the user's answers and transmitting them to a server; means for analyzing the user's answers and evaluating the presentation; and means for generating and displaying feedback to the user based on the evaluation results.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In modern society, presentation skills play an important role in many situations, but there are limited ways to efficiently improve them. Furthermore, existing presentation courses and expert consultations are expensive and time-consuming. This makes it difficult for many people to improve their presentation skills affordably and effectively. Furthermore, while real-time feedback and multifaceted evaluation are required when practicing presentations, there are currently few ways to achieve this. [Means for solving the problem]

[0005] To solve the above problems, a system is provided that includes means for capturing audio and video when a user starts a presentation, means for transmitting the captured audio and video to a server, means for converting the audio to text in real time on the server, means for analyzing the video on the server and identifying the user's facial expressions, eye movements, and gestures, means for an AI to generate appropriate questions based on the analyzed data, means for transmitting the generated questions to a terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation, and means for generating feedback based on the evaluation results and displaying it to the user.This system allows users to improve their presentation skills reasonably and efficiently.

[0006] "User" refers to an individual or group who uses the system to give presentations and improve their presentation skills.

[0007] A "presentation" is a series of actions taken to convey information to an audience and gain their understanding and acceptance.

[0008] "Audio and video capturing means" refers to devices such as microphones and cameras for capturing the user's voice and appearance.

[0009] A "server" is a computer system for receiving, analyzing, and processing captured audio and video data.

[0010] "Real-time speech-to-text conversion means" refers to algorithms and software that use speech recognition technology to instantly convert a user's speech into text data.

[0011] "Means for analyzing video and identifying a user's facial expressions, gaze direction, and gestures" refers to algorithms and software that use video analysis technology to analyze a user's visual information and recognize facial expressions, gaze direction, and body movements.

[0012] "Means for AI to generate appropriate questions" refers to AI technology that uses machine learning models to automatically generate appropriate questions based on the content of a user's presentation and previous data.

[0013] "Means for sending the generated question to the terminal and displaying it to the user" refers to the communication means and display technology for transferring the question generated by the server to the user's terminal and displaying the question on the terminal screen.

[0014] The "means for recapturing the user's answers and transmitting them to the server" refers to a capture device and communication means for once again collecting the audio and video of the user answering the questions and transmitting them to the server.

[0015] The "means for evaluating presentations" refers to algorithms and AI technologies for evaluating the quality of presentations according to multiple evaluation criteria based on the content of the presentation and responses of users.

[0016] The "means for generating feedback and displaying it to the user" refers to software and communication technology for creating a feedback message based on the evaluation results and transmitting it to the user's terminal for display. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0025] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0038] This invention relates to an AI presentation training system that allows users to improve their presentation skills. This system can capture audio and video, and provide real-time analysis and feedback, enabling effective practice in a realistic environment.

[0039] System Configuration

[0040] 1. User Interface

[0041] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[0042] 2. Server Functions

[0043] The server receives the captured audio and video data and performs the following processes:

[0044] 1. Audio analysis:

[0045] The server uses a speech recognition engine to convert the user's speech into text in real time.

[0046] 2. Video analysis:

[0047] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[0048] 3. Question generation:

[0049] The AI ​​on the server generates appropriate questions based on the analyzed data. The questions are generated to match the content of the user's presentation, and the AI's settings take into account personality and expertise.

[0050] 4. Question display:

[0051] The question is sent from the server to the terminal and displayed on the user's screen. The user checks the question and begins answering.

[0052] 5. Answer analysis:

[0053] The user's responses are also captured as audio and video and sent to the server, where they are converted into text using a speech recognition engine and analyzed using a video analysis engine for facial expressions, eye movements, and gestures.

[0054] 6. Rating and Feedback:

[0055] The server comprehensively evaluates the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. Feedback is generated based on the evaluation results and sent to the terminal to be displayed to the user.

[0056] Natural language explanation of program processing

[0057] User starts presenting

[0058] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[0059] The server receives and analyzes the data

[0060] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[0061] AI-generated questions

[0062] The AI ​​on the server generates appropriate questions based on the analyzed data. For example, it might ask, "Please tell me specifically about risk management for this project." The questions are then sent to the device and displayed to the user.

[0063] User answers the question

[0064] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[0065] The server analyzes and evaluates the answers

[0066] The server converts the audio responses back into text and analyzes the video. The server evaluates the user's speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, it may evaluate the user's speaking speed as appropriate, but their voice volume as low.

[0067] Generating and displaying feedback

[0068] The server generates feedback based on the evaluation results, such as "Your speaking speed is appropriate, but your voice volume is low. You should speak a little louder." The feedback is sent to the device and displayed to the user.

[0069] Through this system, users can effectively improve their presentation skills while receiving continuous feedback. Furthermore, repeated practice allows for training in a more realistic environment.

[0070] The processing flow will be explained below.

[0071] Step 1:

[0072] The user starts a dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[0073] Step 2:

[0074] The server inputs the received voice data into a speech recognition engine and converts the voice into text, which is then used to analyze the content of the presentation.

[0075] Step 3:

[0076] The server inputs the received video data into a video analysis engine to analyze the user's facial expressions, gaze, and gestures, for example, to determine whether the user is smiling, looking at the camera, or moving their hands.

[0077] Step 4:

[0078] The AI ​​on the server generates appropriate questions based on the results of audio and video analysis. The questions are related to the content of the presentation, and the difficulty and perspective are adjusted according to the AI ​​settings.

[0079] Step 5:

[0080] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0081] Step 6:

[0082] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[0083] Step 7:

[0084] The server inputs the voice data of the response into a voice recognition engine and converts it into text. It also analyzes the video data of the response using a video analysis engine. The server also analyzes facial expressions, eye movements, and gestures when responding.

[0085] Step 8:

[0086] The server evaluates the presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, the server may give an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not facing the camera."

[0087] Step 9:

[0088] The server generates feedback based on the evaluation results, including specific areas for improvement and advice, such as "If you look more closely at the camera, you will be more appealing to viewers."

[0089] Step 10:

[0090] The generated feedback is sent from the server to the device and displayed on the device screen. The user can review the feedback and use it to practice their next presentation to improve the points pointed out.

[0091] Through this specific processing step, users can effectively learn and improve their presentation skills.

[0092] Example 1

[0093] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0094] Conventional systems for improving presentation skills have the problem that it is difficult for users to receive immediate, specific feedback even when they practice, and because they do not perform real-time analysis, it is difficult to practice in a realistic environment.Furthermore, conventional systems are unable to comprehensively evaluate the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. during a presentation, and therefore can only provide limited feedback.

[0095] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0096] In this invention, the server includes means for converting voice to text, means for analyzing video to identify the user's facial expressions, eye movements, and gestures, means for generating questions using artificial intelligence based on the analyzed data, means for comprehensively evaluating data related to the user's presentation, means for providing sequential feedback based on the evaluation results, and means for enabling the user to practice in a manner close to reality to improve their presentation skills. This allows the user to receive specific feedback in real time and receive advice based on a comprehensive evaluation.

[0097] "Audio capturing means" refers to a microphone or other audio input device that captures the user's speech.

[0098] "Video capture means" refers to a camera or other video input device that captures the user's visual movements and expressions.

[0099] "Server" refers to a computer system for receiving and analyzing captured audio and video data.

[0100] "Means for converting voice to text" refers to technology that uses a voice recognition engine to convert voice data into text data.

[0101] "Means for analyzing video" refers to technology that processes and analyzes video data to identify a user's facial expressions, gaze, gestures, etc.

[0102] "Artificial intelligence" refers to a computer program or system that uses machine learning and natural language processing techniques to make inferences and judgments.

[0103] "Means for generating questions" refers to artificial intelligence technology for creating appropriate questions based on analyzed data.

[0104] "Means for comprehensively evaluating data related to a user's presentation" refers to technology for simultaneously analyzing and evaluating multiple elements, such as the user's speaking speed, voice volume, choice of words, eye contact, facial expressions, and gestures.

[0105] "Means for providing feedback" refers to technology that notifies users of specific improvements and advice in real time based on the analysis and evaluation results.

[0106] "Realistic practice tools" refers to systems that allow users to effectively improve their skills under conditions similar to those experienced in a real presentation.

[0107] This invention relates to a system for enabling users to improve their presentation skills. The system begins when a user launches a dedicated application on a device such as a PC or smartphone and starts a presentation.

[0108] Hardware and Software Configuration

[0109] 1. Hardware

[0110] Voice capture microphone: A microphone is used to capture the user's voice. Examples include a typical condenser microphone or a headset microphone.

[0111] Video capture camera: A camera is used to capture the user's video. Examples include a webcam built into a PC or an external camera.

[0112] 2. Software

[0113] Speech recognition engine: Engines such as Google Cloud Speech-to-Text and IBM Watson Speech to Text are used to convert speech into text.

[0114] Video analysis engine: OpenCV and Google Cloud Vision are used as the engine to analyze video data and identify the user's facial expressions, gaze, and gestures.

[0115] Generative AI models: GPT-3 and BERT are used as AI models to generate appropriate questions based on analyzed data.

[0116] System Operation

[0117] Start your presentation

[0118] The user starts the dedicated application and presses the "Start Presentation" button on their device, such as a PC or smartphone, which then uses the microphone and camera to capture audio and video and transmits the data to the server in real time.

[0119] Data analysis

[0120] The server converts the received voice data into text using a speech recognition engine (Google Cloud Speech-to-Text or IBM Watson Speech to Text), and simultaneously analyzes the video data using a video analysis engine (OpenCV or Google Cloud Vision). For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data, and the video analysis engine analyzes the user's facial expressions and eye direction as they speak.

[0121] question generation

[0122] Based on the analyzed data, a generative AI model (such as GPT-3 or BERT) on the server generates an appropriate question. For example, a question such as "Please tell me specifically about risk management for this project" is generated. The generated question is sent from the server to the device and displayed on the user's screen.

[0123] Answers to questions

[0124] The user checks the question displayed on the device screen and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The server again analyzes this data using its voice recognition engine and video analysis engine to evaluate the user's answer.

[0125] Generating and Providing Feedback

[0126] The server performs a comprehensive evaluation of the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. For example, it obtains a specific evaluation result such as "Your speaking speed is appropriate, but your voice volume is low." Based on this evaluation result, the server generates specific feedback and displays it on the user's screen. For example, it may provide feedback such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[0127] Specific examples

[0128] Prompt Sentence Examples

[0129] Example presentation: "The goal of this project is to increase sales by 20%."

[0130] Example question: "What strategies do you have in mind to achieve this goal?"

[0131] Through this system, users can improve their presentation skills while receiving specific feedback in real time, and by repeatedly practicing, they can improve the quality of their presentations.

[0132] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0133] Step 1:

[0134] The user starts the dedicated application and presses the "Start Presentation" button, which causes the device to capture the user's voice and video and send the data to the server.

[0135] Input: User's audio and video

[0136] Output: Audio and video data sent to the server

[0137] Specific operation: The device's microphone and camera capture audio and video in real time, encode the data, and send it to the server.

[0138] Step 2:

[0139] The server analyzes the received voice data using a speech recognition engine and converts it into text, for example, using Google Cloud Speech-to-Text.

[0140] Input: Audio data sent to the server

[0141] Output: Text data

[0142] Specific operation: The speech recognition engine analyzes the voice data and converts the user's speech into text. For example, it extracts the utterance "The goal of this project is..." as text data.

[0143] Step 3:

[0144] The server analyzes the video data using a video analysis engine, such as OpenCV.

[0145] Input: Video data sent to the server

[0146] Output: Analysis results of user's facial expressions, gaze, and gestures

[0147] Specific operation: The video analysis engine analyzes the video data frame by frame and identifies the user's facial expressions (e.g., smiling or surprised), gaze (e.g., looking toward the camera), and gestures (e.g., hand movements and gestures).

[0148] Step 4:

[0149] A generative AI model on the server generates appropriate questions based on the results of audio and video analysis, using, for example, GPT-3.

[0150] Input: Text data and video analysis results

[0151] Output: Generated question text

[0152] Specific behavior: The generative AI model generates relevant questions based on the analysis results, such as "Please tell me specifically about risk management for this project."

[0153] Step 5:

[0154] The server sends the generated question to the terminal and displays it on the user's screen.

[0155] Input: Generated question text

[0156] Output: The question displayed on the user's screen

[0157] Specific operation: The server sends the generated question text to the terminal, and the dedicated application displays the received question text on the user's screen.

[0158] Step 6:

[0159] The user checks the questions displayed on the terminal and begins to answer them. The terminal again captures audio and video and sends them to the server.

[0160] Input: User's answer audio and video

[0161] Output: Answer audio and video data sent to the server

[0162] Specific operation: When the user answers a question, the device's microphone and camera again capture audio and video and send the data to the server.

[0163] Step 7:

[0164] The server converts the response voice into text again using a voice recognition engine and analyzes the video again.

[0165] Input: Answer audio and video data sent to the server

[0166] Output: Answer text and video analysis results

[0167] Specific operation: The speech recognition engine converts the answer into text, and the video analysis engine reanalyzes the user's facial expressions, gaze, and gestures. For example, the speech "Risk management is..." is extracted as text, and the facial expressions and gaze are reanalyzed.

[0168] Step 8:

[0169] The server will provide an overall rating and generate feedback.

[0170] Input: Answer text and video analysis results

[0171] Output: Feedback text

[0172] Specific behavior: The server comprehensively evaluates the text data and video analysis results and generates specific feedback, such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[0173] Step 9:

[0174] The server sends the generated feedback to the terminal and displays it on the user's screen.

[0175] Input: Feedback text

[0176] Output: Feedback displayed on the user's screen

[0177] Specific operation: The server sends the generated feedback text to the terminal, and the dedicated application displays the received feedback on the user's screen.

[0178] (Application example 1)

[0179] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0180] In modern factories and production sites, workers are required to have the skills to effectively deliver various presentations, such as introducing new products and providing safety training. These skills contribute to worker growth and improved productivity, but traditional educational methods often struggle to provide real-time feedback, and one-way lectures often fail to provide effective instruction. The present invention aims to provide a system that allows workers to efficiently improve their presentation and education / training skills in the field.

[0181] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0182] In this invention, the server includes means for capturing audio and images to start a user's presentation or training, means for transmitting the captured audio and images to the server, means for converting the audio to text in real time in the server, means for analyzing the images in the server and identifying the user's facial expressions, line of sight, and gestures, means for an AI to generate appropriate prompts based on the analyzed data, means for transmitting the generated prompts to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or training, and means for generating feedback based on the evaluation results and displaying them to the user. This allows workers to improve their presentation and training skills while receiving feedback in real time.

[0183] "Users" are workers who use this system to give presentations or provide training.

[0184] "Voice capture" is the act of recording a user's speech as voice data.

[0185] "Video capture" is the act of recording a user's posture and movements as video data.

[0186] A "server" is a computer system that receives and analyzes audio and video data and provides feedback based on the results.

[0187] "Speech-to-text" is the process of converting captured voice data into text data.

[0188] "Image analysis" is the process of identifying a user's facial expressions, gaze, and gestures based on captured video data.

[0189] "Prompts" are appropriate questions or instructions generated by AI based on analyzed data.

[0190] "Feedback" refers to guidance and advice provided to users based on the results of evaluation of their presentations and training.

[0191] "Real-time" refers to the immediacy of time in which the results of a user action are provided as analysis and feedback immediately after the action is taken.

[0192] The present invention is an AI system that enables users to effectively deliver presentations or educational training, and includes functions for capturing audio and video, analyzing them in real time, and generating feedback.

[0193] System Configuration

[0194] 1. User Interface

[0195] The user starts a dedicated application on a device (smartphone, tablet, PC, etc.) and starts a presentation or training session. By pressing the start button, the device captures audio and video and transmits the data to the server in real time.

[0196] 2. Server Functions

[0197] Audio Analysis:

[0198] The server uses a speech recognition engine (e.g., Google Speech Recognition API) to convert the voice data into text. For example, if a user says, "The operating procedure for this machine is...", the server extracts this as text data.

[0199] Image analysis:

[0200] The server uses a video analysis engine (e.g., OpenCV) to analyze the user's facial expressions, gaze, and gestures, for example, to evaluate whether the user is smiling when speaking and in what direction their eyes are pointing.

[0201] Prompt generation:

[0202] Based on the analyzed data, the AI ​​generates appropriate prompts (questions or instructions), such as "Please explain the emergency shutdown procedure for this machine."

[0203] Question display:

[0204] The generated prompts are sent from the server to the terminal and displayed to the user, who then initiates an answer based on the prompts.

[0205] Response analysis and evaluation:

[0206] The user's answers are also captured as audio and video and sent to the server. The server converts the audio back into text and analyzes the video to evaluate the answer. For example, the server may evaluate the answer as "appropriate, but the user's gaze is not focused."

[0207] Feedback generation and display:

[0208] Based on the evaluation results, the server generates feedback, such as "It would be more effective if you directed your gaze more at the other person," and sends it to the device and displays it to the user.

[0209] Program processing explanation

[0210] The specific hardware and software used in this system are as follows:

[0211] Hardware:

[0212] Camera and microphone: Use the device's built-in or external camera and microphone.

[0213] Server: A dedicated server for high-performance data analysis.

[0214] software:

[0215] OpenCV: A library for video capture and analysis.

[0216] SpeechRecognition: An engine for converting speech to text.

[0217] Flask: A framework for server-side data processing.

[0218] Generative AI model: An AI algorithm for generating prompts (questions or instructions).

[0219] Examples of specific examples and prompts

[0220] Consider a case where a user is giving a presentation on how to operate a new machine in a factory. When the user presses the "start button," the camera and microphone are activated to capture audio and video, which are then sent to the server. The server analyzes this and generates a prompt such as "Please explain the emergency stop procedure for this machine," which is displayed on the terminal. When the user responds to the prompt and begins their explanation, their answer is captured again and sent to the server. The server evaluates the user's answer and provides feedback such as "It would be more effective if you paid more attention to the other person."

[0221] Specific prompt examples:

[0222] "Please explain the emergency shutdown procedure for this machine."

[0223] "Please elaborate on the content of the next slide and explain the risk factors that should be considered."

[0224] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0225] Step 1:

[0226] The user launches the dedicated application on their device and presses the "Start Presentation" button. This activates the device's camera and microphone, and audio and video capture begins. The input is the user's audio and video, and the output is the captured audio and video data. Specifically, the device's application starts the capture function of the specified camera and microphone.

[0227] Step 2:

[0228] The device transmits the captured audio and video data to the server in real time. Specifically, the device uploads the data to the server via a network. The data is often compressed and encrypted. The input is the captured audio and video data, and the output is the data transmitted to the server.

[0229] Step 3:

[0230] The server converts the received voice data into text using a voice recognition engine (e.g., Google Speech Recognition API). The input is voice data, and the output is text data. Specifically, the server calls the voice recognition engine, analyzes the voice data, and converts it into text.

[0231] Step 4:

[0232] The server analyzes the received video data using a video analysis engine (e.g., OpenCV) to identify the user's facial expressions, gaze, and gestures. The input is video data, and the output is analyzed feature data of the user's facial expressions, gaze, and gestures. Specifically, the video analysis engine processes the video frame by frame, identifies facial features, and extracts their features.

[0233] Step 5:

[0234] The server uses a generative AI model to generate appropriate prompts based on the analyzed voice text and video data. The input is text data and characteristic data on the user's facial expressions, eye movements, and gestures, and the output is the generated prompt. Specifically, the generative AI model analyzes this data and generates appropriate questions and instructions.

[0235] Step 6:

[0236] The server sends the generated prompt to the terminal, and the terminal displays the prompt on the user's screen. The input is the prompt sent from the server, and the output is the prompt displayed on the user's screen. Specifically, the server sends the generated prompt to the terminal via the network, and the terminal receives it and displays it on the screen.

[0237] Step 7:

[0238] The user responds based on the prompts displayed on the device screen, and the device again captures audio and video and sends them to the server. The input is the user's new audio and video, and the output is the captured data sent to the server. Specifically, the device again captures audio and video using the camera and microphone and sends them to the server.

[0239] Step 8:

[0240] The server analyzes the received audio and video again, converts the user's audio response into text, and analyzes the video to evaluate it. The input is the newly received audio and video data, and the output is the analyzed text data and the evaluation results. Specifically, the server uses a voice recognition engine to convert the audio into text, and a video analysis engine to analyze the user's facial expressions, gaze, and gestures.

[0241] Step 9:

[0242] The server generates feedback based on the evaluation results and sends it to the terminal. The terminal displays the feedback to the user. The input is the evaluation results, and the output is the generated feedback sentence and its display to the user by the terminal that received it. Specifically, the server generates feedback and sends it to the terminal via the network, and the terminal displays the feedback on its screen.

[0243] Example prompt sentence:

[0244] "Please explain the emergency shutdown procedure for this machine."

[0245] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0246] The present invention relates to an AI presentation trainer system that helps users improve their presentation skills. The system combines audio and video capture with an emotion engine that recognizes the user's emotions. This provides more detailed feedback based on the user's emotions, which can be expected to improve presentation skills.

[0247] System Configuration

[0248] 1. User Interface

[0249] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[0250] 2. Server Functions

[0251] The server receives the captured audio and video data and performs the following processes:

[0252] 1. Audio analysis:

[0253] The server uses a speech recognition engine to convert the user's speech into text in real time.

[0254] 2. Video analysis:

[0255] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[0256] 3. Emotion recognition:

[0257] The server uses an emotion engine to recognize the user's emotions, for example, determining whether the user is nervous or relaxed during a presentation.

[0258] 4. Question generation:

[0259] The AI ​​on the server generates appropriate questions based on the analyzed data and emotion recognition results, and adjusts the content and difficulty of the questions depending on the emotional state.

[0260] 5. Question display:

[0261] The question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0262] 6. Answer analysis:

[0263] The user's answers are also captured as audio and video and sent to the server. The server then converts the answers into text using a speech recognition engine, analyzes facial expressions, eye movements, and gestures using a video analysis engine, and analyzes the emotions expressed when answering using an emotion engine.

[0264] 7. Rating and Feedback:

[0265] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may evaluate the user's speaking speed as appropriate, but the user's eyes are not directed toward the camera and the user's emotional state is tense.

[0266] 8. Feedback Generation and Display:

[0267] The server generates feedback based on the evaluation results. The feedback includes specific areas for improvement and advice. For example, the server might generate advice such as, "You should look more closely at the camera. Also, take a deep breath and relax to reduce tension." The feedback is sent from the server to the device and displayed to the user.

[0268] Natural language explanation of program processing

[0269] User starts presenting

[0270] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[0271] The server receives and analyzes the data

[0272] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[0273] emotion recognition

[0274] The server's emotion engine analyzes the user's emotional state from the video data, determining whether they are tense, relaxed, enjoying themselves, etc.

[0275] AI-generated questions

[0276] The AI ​​on the server generates appropriate questions based on the results of voice and video analysis and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The question is then sent to the device and displayed to the user.

[0277] User answers the question

[0278] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[0279] The server analyzes and evaluates the answers

[0280] The server converts the voice response back into text and analyzes the video. It also uses an emotion engine to analyze the user's emotional state. For example, it may evaluate the user's speaking speed as appropriate, but their eyes are not looking at the camera and their emotional state seems tense.

[0281] Generating and displaying feedback

[0282] The server generates feedback based on the evaluation results. For example, it might say, "Your appeal to the audience will increase if you look more closely at the camera. Also, try taking deep breaths to relax." The feedback is sent to the device and displayed to the user.

[0283] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[0284] The processing flow will be explained below.

[0285] Step 1:

[0286] The user starts the dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[0287] Step 2:

[0288] The server inputs the received voice data into a speech recognition engine to convert the speech into text, which is used to identify what the user is saying.

[0289] Step 3:

[0290] The server inputs the received video data into a video analysis engine, which analyzes the user's facial expressions, gaze direction, and gestures. The video analysis engine identifies the user's facial expressions, gaze direction, hand and body movements, etc.

[0291] Step 4:

[0292] The server uses an emotion engine to recognize the user's emotions from the video data, for example, determining whether the user is nervous, relaxed, or having fun.

[0293] Step 5:

[0294] The AI ​​on the server generates appropriate questions based on the results of voice, video, and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The content and difficulty of the questions are adjusted according to the user's emotional state.

[0295] Step 6:

[0296] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0297] Step 7:

[0298] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[0299] Step 8:

[0300] The server inputs the voice data of the response into a speech recognition engine and converts it into text. It also analyzes the video data of the response with a video analysis engine to identify the user's facial expressions, eye movements, and gestures. The server then uses an emotion engine to analyze the user's emotional state.

[0301] Step 9:

[0302] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may provide an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera. Also, your emotional state seems tense."

[0303] Step 10:

[0304] The server generates feedback based on the evaluation results. The feedback includes specific improvements and advice. For example, advice such as "Turn your eyes more closely into the camera. Also, try taking deep breaths to relax" is generated. The feedback is sent from the server to the device and displayed to the user.

[0305] Through this specific processing step, users can effectively learn and improve their presentation skills. Multifaceted feedback, including emotional state, is provided, allowing users to gain a deeper understanding of how their presentations are perceived.

[0306] Example 2

[0307] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0308] Conventional presentation training systems are limited to capturing and analyzing audio and video data, and are unable to provide feedback that takes into account the user's emotional state. This limits the extent to which users can improve their presentation skills. Specifically, the user's emotional state, such as tension or relaxation, is not reflected in the evaluation or feedback of the presentation, making it difficult to provide training that closely resembles a real presentation situation.

[0309] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0310] In this invention, the server includes means for capturing audio and video to start a presentation by a user, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for an AI model to generate appropriate questions based on the analyzed data, means for transmitting the generated questions to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation, means for generating feedback based on the evaluation results and displaying them to the user, means for recognizing and analyzing the user's emotions, and means for adjusting the content of the feedback based on the user's emotional state, thereby enabling multifaceted feedback that takes the user's emotional state into consideration.

[0311] "User" refers to an individual or group that makes a presentation using a dedicated application.

[0312] "Terminal" refers to a device such as a PC, smartphone, or tablet that a user uses when giving a presentation.

[0313] "Means for capturing audio and video" refers to the ability to record audio and video data of a user using the microphone and camera installed on the device.

[0314] "Means for transmitting audio and video to a server" refers to a technology for transmitting captured audio and video data to a server via a network in real time.

[0315] "Means for converting speech to text" refers to a function that utilizes a speech recognition engine to convert captured speech data into corresponding text data.

[0316] "Means for analyzing video and identifying a user's facial expressions, gaze direction, and gestures" refers to a technology that uses a video analysis engine to detect and analyze a user's facial expressions, gaze direction, and gestures.

[0317] "Means for an AI model to generate appropriate questions" refers to the function of automatically generating appropriate questions for a user using a generative AI model based on the results of audio and video analysis.

[0318] "Means for transmitting the generated question to the terminal and displaying it to the user" refers to a function for transmitting a question generated by the server to the terminal and displaying it on the screen of the terminal.

[0319] "Means for recapturing the user's answers and sending them to the server" refers to a function for recapturing audio and video data when the user gives an answer and sending that data to the server.

[0320] "Means for analyzing the user's answers and evaluating the presentation" refers to a technology for analyzing audio and video data and evaluating the content of the user's presentation.

[0321] "Means for generating feedback based on the evaluation results and displaying it to the user" refers to a function for generating useful feedback based on the evaluation results of the presentation, transmitting it to the terminal, and displaying it to the user.

[0322] "Means for recognizing and analyzing user emotions" refers to technology that recognizes and analyzes a user's emotional state from video data.

[0323] The "means for adjusting the feedback content based on the user's emotional state" refers to a function for appropriately adjusting and providing the feedback content based on the emotion recognition result.

[0324] The present invention is a system for improving users' presentation skills, and in particular, by combining emotion recognition technology, it is possible to provide feedback based on the user's emotional state, allowing users to acquire more effective presentation skills.

[0325] First, the user launches a dedicated application on a device such as a PC or smartphone. When the user presses the "Start Presentation" button, the device's built-in microphone and camera start up and begin capturing audio and video data. The captured audio and video data is then sent to a server in real time via the Internet.

[0326] The server converts the received voice data into text data in real time using the Google Cloud Speech-to-Text API. At the same time, the received video data is analyzed using OpenCV. The video analysis engine detects and analyzes the user's facial expressions, eye direction, and gestures. This analysis allows the system to understand the context in which the user is speaking.

[0327] Furthermore, the server uses the Microsoft Azure Emotion API to analyze the user's emotional state from the video data. Based on facial expressions, eye movements, and changes in facial muscles, it can determine whether the user is tense, relaxed, or enjoying themselves. For example, the level of tension and stress can also be calculated.

[0328] The server uses OpenAI GPT-4 to generate appropriate questions based on these analysis results. The generated questions are sent from the server to the device and displayed on the user's screen. For example, a question such as "Please tell me specifically about risk management for this project" may be displayed.

[0329] The user begins to answer the displayed question. The device again uses the microphone and camera to capture audio and video data of the user's answer. The captured data is sent to the server, which again performs speech recognition and video analysis. The answer is converted to text using the Google Cloud Speech-to-Text API, and the video analysis engine analyzes facial expressions, eye movements, and gestures. The emotion engine is also used to analyze the user's emotional state.

[0330] The server comprehensively evaluates this data and makes a rating based on the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, a specific rating may be given such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera and your emotional state is tense."

[0331] Based on the evaluation results, the server generates feedback, which includes specific areas for improvement and advice. For example, advice such as "You can appeal more to the audience by looking more closely at the camera. Also, it would be a good idea to take deep breaths to relieve tension" is provided. The generated feedback is sent from the server to the device and displayed to the user.

[0332] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[0333] Specific examples

[0334] Prompt Sentence Examples

[0335] "Please tell me more about the risk management for this project."

[0336] Example of user analysis results

[0337] Speech text: "The first step in managing the risks of this project is..."

[0338] Facial expression analysis result: nervous

[0339] Eye analysis results: Not facing the camera

[0340] Emotion recognition result: tension

[0341] Feedback example

[0342] "Your speech is clear, but it would be better if you looked directly into the camera. To relax, try taking a deep breath before you start speaking."

[0343] In this way, the system can improve users' presentation skills in a variety of ways.

[0344] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0345] Step 1:

[0346] User starts presenting

[0347] The user launches the dedicated app on a device such as a PC or smartphone and presses the "Start Presentation" button. The input is the user's operation (pressing the button), which triggers the device's microphone and camera to start up. The output is the device starting to capture audio and video data.

[0348] Step 2:

[0349] Sending captured data

[0350] The terminal transmits the captured audio and video data to the server in real time. The input is the captured audio and video data, which are transmitted to the server via the network. The output is the real-time audio and video streams that arrive at the server.

[0351] Step 3:

[0352] Converting audio data to text

[0353] The server converts the received audio stream into text data using the Google Cloud Speech-to-Text API. The input is the received audio data, which is then subjected to speech recognition to generate text data. The output is text data that shows what the user said.

[0354] Step 4:

[0355] Video data analysis

[0356] The server analyzes the received video stream using OpenCV. The input is the received video data, and video analysis is performed on this data to identify the user's facial expressions, gaze, and gestures. The output is the analysis result data on the user's facial expressions, gaze, and gestures.

[0357] Step 5:

[0358] emotion recognition

[0359] The server uses the Microsoft Azure Emotion API to analyze the user's emotional state from video data. The input is the video analysis results data, and emotion recognition is performed based on this data. The output is data on the user's emotional state, such as whether they are tense or relaxed.

[0360] Step 6:

[0361] Question Generation

[0362] The server uses OpenAI GPT-4 to generate appropriate questions based on the results of voice and video analysis and emotion recognition. The input is voice text data, analysis result data, and emotion recognition data, and the question is generated based on these. The output is the generated question.

[0363] Step 7:

[0364] Show Questions

[0365] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and this data is sent to the terminal via the network. The output is the question displayed on the user's application screen. Example: "Please tell me specifically about risk management for this project."

[0366] Step 8:

[0367] User response capture

[0368] The user checks the displayed question and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The input is the audio and video data of the user's answer, which is captured and sent to the server. The output is the audio and video data that reaches the server.

[0369] Step 9:

[0370] Analysis of responses

[0371] The server converts the retransmitted voice data into text using a voice recognition engine, and analyzes the video data using a video analysis engine. The input is the user's response voice and video data, which are then subjected to voice recognition and video analysis. The output is the text of the response and the analysis results of the user's facial expressions, gaze, and gestures.

[0372] Step 10:

[0373] Emotion recognition when answering

[0374] The server again uses the Microsoft Azure Emotion API to analyze the user's emotional state at the time of answering. The input is the video analysis result data, and emotion recognition is performed based on this. The output is data indicating the user's emotional state.

[0375] Step 11:

[0376] Evaluation of presentation content

[0377] The server evaluates the presentation based on the content of the answers, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. The input is audio-text data, video analysis data, and emotion recognition data, and a comprehensive evaluation is made based on these. The output is the evaluation result of the user's presentation.

[0378] Step 12:

[0379] Generate feedback

[0380] Based on the evaluation results, the server generates feedback including specific areas for improvement and advice. The input is the evaluation result data, and feedback statements are generated based on this. The output is the generated feedback statements. Example: "Your appeal to the audience will increase if you look more closely at the camera. Try taking deep breaths to relax."

[0381] Step 13:

[0382] View Feedback

[0383] The server sends the generated feedback to the terminal and displays it on the user's screen. The input is the generated feedback sentence, and this data is sent to the terminal via the network. The output is the feedback sentence displayed on the user's application screen.

[0384] The above is the flow of processing for this system program. We have explained the specific operations performed at each step, as well as the processing or calculation of input and output data.

[0385] (Application example 2)

[0386] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0387] To improve the customer service skills of store staff, practice in a realistic environment is necessary. However, conventional methods provide limited feedback, and areas for improvement are not clearly identified. Furthermore, training does not take into account the emotional state of staff, and emotional upset during customer service can affect customer satisfaction. To address these issues, a system is needed that analyzes staff's customer service skills from multiple angles and provides real-time feedback.

[0388] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0389] In this invention, the server includes means for capturing audio and video to allow a user to start presentation or customer service training, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for recognizing the user's emotions in real time and generating appropriate questions using AI based on the analyzed data, means for transmitting the generated questions to a terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or customer service, and means for generating feedback based on the evaluation results and displaying them to the user. This allows staff to improve their customer service skills in real time while receiving multifaceted feedback including facial expressions and emotional information at the customer service site.

[0390] An "audio capture means" is a device or software for capturing audio data.

[0391] "Video capture means" refers to a device or software for acquiring video data.

[0392] A "server" is a computer system or computer on a network that analyzes and stores data.

[0393] "Real-time conversion means" is a technology or device that converts speech into text almost instantly.

[0394] "Video analysis means" refers to a technique or device that analyzes video data and extracts features.

[0395] The "facial expression identification means" is a technique or device that identifies the facial expression of a user from video data.

[0396] The "gaze direction identification means" is a technique or device that identifies the direction of a user's gaze from video data.

[0397] The "gesture recognition means" is a technique or device that recognizes a user's gesture action from video data.

[0398] "Emotion recognition means" refers to a technique or device that recognizes the user's emotional state from video data and audio data.

[0399] A "question generator" is a technique or device that generates appropriate questions based on the analyzed data.

[0400] The "question display means" is a technique or device that visually displays the generated question to the user.

[0401] An "answer capture means" is a device or software for recapturing a user's answers.

[0402] The "evaluation means" is a technology or device that analyzes and evaluates the user's answers and the performance of customer service and presentation.

[0403] The "feedback generation means" is a technique or device that generates feedback to the user based on the evaluation results.

[0404] A "feedback display means" is a technique or device that visually displays the generated feedback to the user.

[0405] This invention relates to a system for improving customer service skills in brick-and-mortar stores, which has the function of capturing audio and video data and analyzing the user's facial expressions, gaze, gestures, and emotional state. Based on the analysis results, AI generates appropriate questions and provides feedback to the user.

[0406] The server includes an audio capture means, a video capture means, a real-time conversion means, a video analysis means, a facial expression identification means, a gaze identification means, a gesture identification means, an emotion recognition means, a question generation means, a question display means, an answer capture means, an evaluation means, and a feedback generation means.

[0407] The user launches a dedicated application on a smartphone or other device and presses the "Start Training" button. The device then captures audio and video data and sends it to the server in real time. The server then converts the audio into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text), analyzes the user's facial expressions, gaze, and gestures using a video analysis engine (e.g., OpenCV), and recognizes the user's emotional state using an emotion recognition engine (e.g., Affectiva).

[0408] Based on the analyzed data, the AI ​​on the server uses a question generation engine to generate appropriate questions. For example, a question such as "Please tell me specifically about risk management for this project" may be generated. The generated question is sent to the user's device and displayed on the screen.

[0409] The user answers the displayed questions. The device captures the answer again as audio and video data and sends it to the server. The server receives the answer data and analyzes it again using its voice recognition engine and video analysis engine.

[0410] The server evaluates the results of the analysis and generates feedback using a feedback generation engine, including specific suggestions for improvement such as, "Your speaking speed is appropriate, but you don't seem to be looking at the camera enough. You should be more aware of the camera."

[0411] The generated feedback is sent to the user's device and displayed in real time, allowing the user to improve their customer service skills in multiple ways.

[0412] Prompt Sentence Examples

[0413] "We have a new product in stock. Please explain in detail how it differs from similar products we have available so far, and what you recommend it for customers."

[0414] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0415] Step 1:

[0416] The user launches the dedicated application on a device such as a smartphone and presses the "Start Training" button. In this step, the device starts capturing audio and video data and sends this data to the server in real time. The input is the audio and video made by the user, and the output is the captured data sent to the server.

[0417] Step 2:

[0418] The server inputs the received voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into text in real time. The input is the captured voice data, and the output is the converted text data. The specific operation involves signal processing of the voice data and generation of text data.

[0419] Step 3:

[0420] The server inputs the received video data into a video analysis engine (e.g., OpenCV) to analyze facial expressions, gaze, and gestures. The input is the captured video data, and the output is the analyzed facial expression data, gaze data, and gesture data. The specific operation is to analyze the video frames and extract features.

[0421] Step 4:

[0422] The server uses an emotion recognition engine (e.g., Affectiva) to recognize the user's emotional state from video and audio data. The input is video and audio data, and the output is the recognized emotional state data. The specific operation is to classify and label the emotional state.

[0423] Step 5:

[0424] The AI ​​on the server generates appropriate questions based on the analyzed data and emotional state. The input is the analyzed text data, facial expression data, eye gaze data, gesture data, and emotional state data, and the output is the generated question. The specific operations are natural language generation and text generation.

[0425] Step 6:

[0426] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and the output is the question displayed on the terminal screen. The specific operations are data transmission and UI update.

[0427] Step 7:

[0428] The user answers the displayed questions. The device recaptures the answers as audio and video data and sends them to the server. The input is the user's audio and video answers, and the output is the captured data sent to the server.

[0429] Step 8:

[0430] The server then inputs the received voice data into a voice recognition engine and converts it into text. The input is the captured response voice, and the output is the converted text data. The specific operations are voice recognition and text conversion.

[0431] Step 9:

[0432] The server inputs the received video data into the video analysis engine and analyzes facial expressions, gaze, and gestures. The input is the captured response video, and the output is analyzed facial expression data, gaze data, and gesture data. The specific operations are video analysis and feature extraction.

[0433] Step 10:

[0434] The server evaluates the answer data and generates feedback using a feedback generation engine. The input is the analyzed answer data, and the output is the generated feedback. The specific operations are evaluation analysis and feedback generation.

[0435] Step 11:

[0436] The server sends the generated feedback to the device and displays it to the user in real time. The input is the generated feedback, and the output is the feedback displayed on the device screen. The specific operations are data transmission and UI update.

[0437] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0438] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0439] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0440] [Second embodiment]

[0441] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0442] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0443] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0444] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0445] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0446] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0447] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0448] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0449] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0450] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0451] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0452] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0453] This invention relates to an AI presentation training system that allows users to improve their presentation skills. This system can capture audio and video, and provide real-time analysis and feedback, enabling effective practice in a realistic environment.

[0454] System Configuration

[0455] 1. User Interface

[0456] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[0457] 2. Server Functions

[0458] The server receives the captured audio and video data and performs the following processes:

[0459] 1. Audio analysis:

[0460] The server uses a speech recognition engine to convert the user's speech into text in real time.

[0461] 2. Video analysis:

[0462] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[0463] 3. Question generation:

[0464] The AI ​​on the server generates appropriate questions based on the analyzed data. The questions are generated to match the content of the user's presentation, and the AI's settings take into account personality and expertise.

[0465] 4. Question display:

[0466] The question is sent from the server to the terminal and displayed on the user's screen. The user checks the question and begins answering.

[0467] 5. Answer analysis:

[0468] The user's responses are also captured as audio and video and sent to the server, where they are converted into text using a speech recognition engine and analyzed using a video analysis engine for facial expressions, eye movements, and gestures.

[0469] 6. Rating and Feedback:

[0470] The server comprehensively evaluates the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. Feedback is generated based on the evaluation results and sent to the terminal to be displayed to the user.

[0471] Natural language explanation of program processing

[0472] User starts presenting

[0473] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[0474] The server receives and analyzes the data

[0475] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[0476] AI-generated questions

[0477] The AI ​​on the server generates appropriate questions based on the analyzed data. For example, it might ask, "Please tell me specifically about risk management for this project." The questions are then sent to the device and displayed to the user.

[0478] User answers the question

[0479] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[0480] The server analyzes and evaluates the answers

[0481] The server converts the audio responses back into text and analyzes the video. The server evaluates the user's speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, it may evaluate the user's speaking speed as appropriate, but their voice volume as low.

[0482] Generating and displaying feedback

[0483] The server generates feedback based on the evaluation results, such as "Your speaking speed is appropriate, but your voice volume is low. You should speak a little louder." The feedback is sent to the device and displayed to the user.

[0484] Through this system, users can effectively improve their presentation skills while receiving continuous feedback. Furthermore, repeated practice allows for training in a more realistic environment.

[0485] The processing flow will be explained below.

[0486] Step 1:

[0487] The user starts a dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[0488] Step 2:

[0489] The server inputs the received voice data into a speech recognition engine and converts the voice into text, which is then used to analyze the content of the presentation.

[0490] Step 3:

[0491] The server inputs the received video data into a video analysis engine to analyze the user's facial expressions, gaze, and gestures, for example, to determine whether the user is smiling, looking at the camera, or moving their hands.

[0492] Step 4:

[0493] The AI ​​on the server generates appropriate questions based on the results of audio and video analysis. The questions are related to the content of the presentation, and the difficulty and perspective are adjusted according to the AI ​​settings.

[0494] Step 5:

[0495] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0496] Step 6:

[0497] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[0498] Step 7:

[0499] The server inputs the voice data of the response into a voice recognition engine and converts it into text. It also analyzes the video data of the response using a video analysis engine. The server also analyzes facial expressions, eye movements, and gestures when responding.

[0500] Step 8:

[0501] The server evaluates the presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, the server may give an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not facing the camera."

[0502] Step 9:

[0503] The server generates feedback based on the evaluation results, including specific areas for improvement and advice, such as "If you look more closely at the camera, you will be more appealing to viewers."

[0504] Step 10:

[0505] The generated feedback is sent from the server to the device and displayed on the device screen. The user can review the feedback and use it to practice their next presentation to improve the points pointed out.

[0506] Through this specific processing step, users can effectively learn and improve their presentation skills.

[0507] Example 1

[0508] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0509] Conventional systems for improving presentation skills have the problem that it is difficult for users to receive immediate, specific feedback even when they practice, and because they do not perform real-time analysis, it is difficult to practice in a realistic environment.Furthermore, conventional systems are unable to comprehensively evaluate the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. during a presentation, and therefore can only provide limited feedback.

[0510] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0511] In this invention, the server includes means for converting voice to text, means for analyzing video to identify the user's facial expressions, eye movements, and gestures, means for generating questions using artificial intelligence based on the analyzed data, means for comprehensively evaluating data related to the user's presentation, means for providing sequential feedback based on the evaluation results, and means for enabling the user to practice in a manner close to reality to improve their presentation skills. This allows the user to receive specific feedback in real time and receive advice based on a comprehensive evaluation.

[0512] "Audio capturing means" refers to a microphone or other audio input device that captures the user's speech.

[0513] "Video capture means" refers to a camera or other video input device that captures the user's visual movements and expressions.

[0514] "Server" refers to a computer system for receiving and analyzing captured audio and video data.

[0515] "Means for converting voice to text" refers to technology that uses a voice recognition engine to convert voice data into text data.

[0516] "Means for analyzing video" refers to technology that processes and analyzes video data to identify a user's facial expressions, gaze, gestures, etc.

[0517] "Artificial intelligence" refers to a computer program or system that uses machine learning and natural language processing techniques to make inferences and judgments.

[0518] "Means for generating questions" refers to artificial intelligence technology for creating appropriate questions based on analyzed data.

[0519] "Means for comprehensively evaluating data related to a user's presentation" refers to technology for simultaneously analyzing and evaluating multiple elements, such as the user's speaking speed, voice volume, choice of words, eye contact, facial expressions, and gestures.

[0520] "Means for providing feedback" refers to technology that notifies users of specific improvements and advice in real time based on the analysis and evaluation results.

[0521] "Realistic practice tools" refers to systems that allow users to effectively improve their skills under conditions similar to those experienced in a real presentation.

[0522] This invention relates to a system for enabling users to improve their presentation skills. The system begins when a user launches a dedicated application on a device such as a PC or smartphone and starts a presentation.

[0523] Hardware and Software Configuration

[0524] 1. Hardware

[0525] Voice capture microphone: A microphone is used to capture the user's voice. Examples include a typical condenser microphone or a headset microphone.

[0526] Video capture camera: A camera is used to capture the user's video. Examples include a webcam built into a PC or an external camera.

[0527] 2. Software

[0528] Speech recognition engine: Engines such as Google Cloud Speech-to-Text and IBM Watson Speech to Text are used to convert speech into text.

[0529] Video analysis engine: OpenCV and Google Cloud Vision are used as the engine to analyze video data and identify the user's facial expressions, gaze, and gestures.

[0530] Generative AI models: GPT-3 and BERT are used as AI models to generate appropriate questions based on analyzed data.

[0531] System Operation

[0532] Start your presentation

[0533] The user starts the dedicated application and presses the "Start Presentation" button on their device, such as a PC or smartphone, which then uses the microphone and camera to capture audio and video and transmits the data to the server in real time.

[0534] Data analysis

[0535] The server converts the received voice data into text using a speech recognition engine (Google Cloud Speech-to-Text or IBM Watson Speech to Text), and simultaneously analyzes the video data using a video analysis engine (OpenCV or Google Cloud Vision). For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data, and the video analysis engine analyzes the user's facial expressions and eye direction as they speak.

[0536] question generation

[0537] Based on the analyzed data, a generative AI model (such as GPT-3 or BERT) on the server generates an appropriate question. For example, a question such as "Please tell me specifically about risk management for this project" is generated. The generated question is sent from the server to the device and displayed on the user's screen.

[0538] Answers to questions

[0539] The user checks the question displayed on the device screen and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The server again analyzes this data using its voice recognition engine and video analysis engine to evaluate the user's answer.

[0540] Generating and Providing Feedback

[0541] The server performs a comprehensive evaluation of the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. For example, it obtains a specific evaluation result such as "Your speaking speed is appropriate, but your voice volume is low." Based on this evaluation result, the server generates specific feedback and displays it on the user's screen. For example, it may provide feedback such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[0542] Specific examples

[0543] Prompt Sentence Examples

[0544] Example presentation: "The goal of this project is to increase sales by 20%."

[0545] Example question: "What strategies do you have in mind to achieve this goal?"

[0546] Through this system, users can improve their presentation skills while receiving specific feedback in real time, and by repeatedly practicing, they can improve the quality of their presentations.

[0547] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0548] Step 1:

[0549] The user starts the dedicated application and presses the "Start Presentation" button, which causes the device to capture the user's voice and video and send the data to the server.

[0550] Input: User's audio and video

[0551] Output: Audio and video data sent to the server

[0552] Specific operation: The device's microphone and camera capture audio and video in real time, encode the data, and send it to the server.

[0553] Step 2:

[0554] The server analyzes the received voice data using a speech recognition engine and converts it into text, for example, using Google Cloud Speech-to-Text.

[0555] Input: Audio data sent to the server

[0556] Output: Text data

[0557] Specific operation: The speech recognition engine analyzes the voice data and converts the user's speech into text. For example, it extracts the utterance "The goal of this project is..." as text data.

[0558] Step 3:

[0559] The server analyzes the video data using a video analysis engine, such as OpenCV.

[0560] Input: Video data sent to the server

[0561] Output: Analysis results of user's facial expressions, gaze, and gestures

[0562] Specific operation: The video analysis engine analyzes the video data frame by frame and identifies the user's facial expressions (e.g., smiling or surprised), gaze (e.g., looking toward the camera), and gestures (e.g., hand movements and gestures).

[0563] Step 4:

[0564] A generative AI model on the server generates appropriate questions based on the results of audio and video analysis, using, for example, GPT-3.

[0565] Input: Text data and video analysis results

[0566] Output: Generated question text

[0567] Specific behavior: The generative AI model generates relevant questions based on the analysis results, such as "Please tell me specifically about risk management for this project."

[0568] Step 5:

[0569] The server sends the generated question to the terminal and displays it on the user's screen.

[0570] Input: Generated question text

[0571] Output: The question displayed on the user's screen

[0572] Specific operation: The server sends the generated question text to the terminal, and the dedicated application displays the received question text on the user's screen.

[0573] Step 6:

[0574] The user checks the questions displayed on the terminal and begins to answer them. The terminal again captures audio and video and sends them to the server.

[0575] Input: User's answer audio and video

[0576] Output: Answer audio and video data sent to the server

[0577] Specific operation: When the user answers a question, the device's microphone and camera again capture audio and video and send the data to the server.

[0578] Step 7:

[0579] The server converts the response voice into text again using a voice recognition engine and analyzes the video again.

[0580] Input: Answer audio and video data sent to the server

[0581] Output: Answer text and video analysis results

[0582] Specific operation: The speech recognition engine converts the answer into text, and the video analysis engine reanalyzes the user's facial expressions, gaze, and gestures. For example, the speech "Risk management is..." is extracted as text, and the facial expressions and gaze are reanalyzed.

[0583] Step 8:

[0584] The server will provide an overall rating and generate feedback.

[0585] Input: Answer text and video analysis results

[0586] Output: Feedback text

[0587] Specific behavior: The server comprehensively evaluates the text data and video analysis results and generates specific feedback, such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[0588] Step 9:

[0589] The server sends the generated feedback to the terminal and displays it on the user's screen.

[0590] Input: Feedback text

[0591] Output: Feedback displayed on the user's screen

[0592] Specific operation: The server sends the generated feedback text to the terminal, and the dedicated application displays the received feedback on the user's screen.

[0593] (Application example 1)

[0594] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0595] In modern factories and production sites, workers are required to have the skills to effectively deliver various presentations, such as introducing new products and providing safety training. These skills contribute to worker growth and improved productivity, but traditional educational methods often struggle to provide real-time feedback, and one-way lectures often fail to provide effective instruction. The present invention aims to provide a system that allows workers to efficiently improve their presentation and education / training skills in the field.

[0596] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0597] In this invention, the server includes means for capturing audio and images to start a user's presentation or training, means for transmitting the captured audio and images to the server, means for converting the audio to text in real time in the server, means for analyzing the images in the server and identifying the user's facial expressions, line of sight, and gestures, means for an AI to generate appropriate prompts based on the analyzed data, means for transmitting the generated prompts to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or training, and means for generating feedback based on the evaluation results and displaying them to the user. This allows workers to improve their presentation and training skills while receiving feedback in real time.

[0598] "Users" are workers who use this system to give presentations or provide training.

[0599] "Voice capture" is the act of recording a user's speech as voice data.

[0600] "Video capture" is the act of recording a user's posture and movements as video data.

[0601] A "server" is a computer system that receives and analyzes audio and video data and provides feedback based on the results.

[0602] "Speech-to-text" is the process of converting captured voice data into text data.

[0603] "Image analysis" is the process of identifying a user's facial expressions, gaze, and gestures based on captured video data.

[0604] "Prompts" are appropriate questions or instructions generated by AI based on analyzed data.

[0605] "Feedback" refers to guidance and advice provided to users based on the results of evaluation of their presentations and training.

[0606] "Real-time" refers to the immediacy of time in which the results of a user action are provided as analysis and feedback immediately after the action is taken.

[0607] The present invention is an AI system that enables users to effectively deliver presentations or educational training, and includes functions for capturing audio and video, analyzing them in real time, and generating feedback.

[0608] System Configuration

[0609] 1. User Interface

[0610] The user starts a dedicated application on a device (smartphone, tablet, PC, etc.) and starts a presentation or training session. By pressing the start button, the device captures audio and video and transmits the data to the server in real time.

[0611] 2. Server Functions

[0612] Audio Analysis:

[0613] The server uses a speech recognition engine (e.g., Google Speech Recognition API) to convert the voice data into text. For example, if a user says, "The operating procedure for this machine is...", the server extracts this as text data.

[0614] Image analysis:

[0615] The server uses a video analysis engine (e.g., OpenCV) to analyze the user's facial expressions, gaze, and gestures, for example, to evaluate whether the user is smiling when speaking and in what direction their eyes are pointing.

[0616] Prompt generation:

[0617] Based on the analyzed data, the AI ​​generates appropriate prompts (questions or instructions), such as "Please explain the emergency shutdown procedure for this machine."

[0618] Question display:

[0619] The generated prompts are sent from the server to the terminal and displayed to the user, who then initiates an answer based on the prompts.

[0620] Response analysis and evaluation:

[0621] The user's answers are also captured as audio and video and sent to the server. The server converts the audio back into text and analyzes the video to evaluate the answer. For example, the server may evaluate the answer as "appropriate, but the user's gaze is not focused."

[0622] Feedback generation and display:

[0623] Based on the evaluation results, the server generates feedback, such as "It would be more effective if you directed your gaze more at the other person," and sends it to the device and displays it to the user.

[0624] Program processing explanation

[0625] The specific hardware and software used in this system are as follows:

[0626] Hardware:

[0627] Camera and microphone: Use the device's built-in or external camera and microphone.

[0628] Server: A dedicated server for high-performance data analysis.

[0629] software:

[0630] OpenCV: A library for video capture and analysis.

[0631] SpeechRecognition: An engine for converting speech to text.

[0632] Flask: A framework for server-side data processing.

[0633] Generative AI model: An AI algorithm for generating prompts (questions or instructions).

[0634] Examples of specific examples and prompts

[0635] Consider a case where a user is giving a presentation on how to operate a new machine in a factory. When the user presses the "start button," the camera and microphone are activated to capture audio and video, which are then sent to the server. The server analyzes this and generates a prompt such as "Please explain the emergency stop procedure for this machine," which is displayed on the terminal. When the user responds to the prompt and begins their explanation, their answer is captured again and sent to the server. The server evaluates the user's answer and provides feedback such as "It would be more effective if you paid more attention to the other person."

[0636] Specific prompt examples:

[0637] "Please explain the emergency shutdown procedure for this machine."

[0638] "Please elaborate on the content of the next slide and explain the risk factors that should be considered."

[0639] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0640] Step 1:

[0641] The user launches the dedicated application on their device and presses the "Start Presentation" button. This activates the device's camera and microphone, and audio and video capture begins. The input is the user's audio and video, and the output is the captured audio and video data. Specifically, the device's application starts the capture function of the specified camera and microphone.

[0642] Step 2:

[0643] The device transmits the captured audio and video data to the server in real time. Specifically, the device uploads the data to the server via a network. The data is often compressed and encrypted. The input is the captured audio and video data, and the output is the data transmitted to the server.

[0644] Step 3:

[0645] The server converts the received voice data into text using a voice recognition engine (e.g., Google Speech Recognition API). The input is voice data, and the output is text data. Specifically, the server calls the voice recognition engine, analyzes the voice data, and converts it into text.

[0646] Step 4:

[0647] The server analyzes the received video data using a video analysis engine (e.g., OpenCV) to identify the user's facial expressions, gaze, and gestures. The input is video data, and the output is analyzed feature data of the user's facial expressions, gaze, and gestures. Specifically, the video analysis engine processes the video frame by frame, identifies facial features, and extracts their features.

[0648] Step 5:

[0649] The server uses a generative AI model to generate appropriate prompts based on the analyzed voice text and video data. The input is text data and characteristic data on the user's facial expressions, eye movements, and gestures, and the output is the generated prompt. Specifically, the generative AI model analyzes this data and generates appropriate questions and instructions.

[0650] Step 6:

[0651] The server sends the generated prompt to the terminal, and the terminal displays the prompt on the user's screen. The input is the prompt sent from the server, and the output is the prompt displayed on the user's screen. Specifically, the server sends the generated prompt to the terminal via the network, and the terminal receives it and displays it on the screen.

[0652] Step 7:

[0653] The user responds based on the prompts displayed on the device screen, and the device again captures audio and video and sends them to the server. The input is the user's new audio and video, and the output is the captured data sent to the server. Specifically, the device again captures audio and video using the camera and microphone and sends them to the server.

[0654] Step 8:

[0655] The server analyzes the received audio and video again, converts the user's audio response into text, and analyzes the video to evaluate it. The input is the newly received audio and video data, and the output is the analyzed text data and the evaluation results. Specifically, the server uses a voice recognition engine to convert the audio into text, and a video analysis engine to analyze the user's facial expressions, gaze, and gestures.

[0656] Step 9:

[0657] The server generates feedback based on the evaluation results and sends it to the terminal. The terminal displays the feedback to the user. The input is the evaluation results, and the output is the generated feedback sentence and its display to the user by the terminal that received it. Specifically, the server generates feedback and sends it to the terminal via the network, and the terminal displays the feedback on its screen.

[0658] Example prompt sentence:

[0659] "Please explain the emergency shutdown procedure for this machine."

[0660] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0661] The present invention relates to an AI presentation trainer system that helps users improve their presentation skills. The system combines audio and video capture with an emotion engine that recognizes the user's emotions. This provides more detailed feedback based on the user's emotions, which can be expected to improve presentation skills.

[0662] System Configuration

[0663] 1. User Interface

[0664] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[0665] 2. Server Functions

[0666] The server receives the captured audio and video data and performs the following processes:

[0667] 1. Audio analysis:

[0668] The server uses a speech recognition engine to convert the user's speech into text in real time.

[0669] 2. Video analysis:

[0670] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[0671] 3. Emotion recognition:

[0672] The server uses an emotion engine to recognize the user's emotions, for example, determining whether the user is nervous or relaxed during a presentation.

[0673] 4. Question generation:

[0674] The AI ​​on the server generates appropriate questions based on the analyzed data and emotion recognition results, and adjusts the content and difficulty of the questions depending on the emotional state.

[0675] 5. Question display:

[0676] The question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0677] 6. Answer analysis:

[0678] The user's answers are also captured as audio and video and sent to the server. The server then converts the answers into text using a speech recognition engine, analyzes facial expressions, eye movements, and gestures using a video analysis engine, and analyzes the emotions expressed when answering using an emotion engine.

[0679] 7. Rating and Feedback:

[0680] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may evaluate the user's speaking speed as appropriate, but the user's eyes are not directed toward the camera and the user's emotional state is tense.

[0681] 8. Feedback Generation and Display:

[0682] The server generates feedback based on the evaluation results. The feedback includes specific areas for improvement and advice. For example, the server might generate advice such as, "You should look more closely at the camera. Also, take a deep breath and relax to reduce tension." The feedback is sent from the server to the device and displayed to the user.

[0683] Natural language explanation of program processing

[0684] User starts presenting

[0685] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[0686] The server receives and analyzes the data

[0687] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[0688] emotion recognition

[0689] The server's emotion engine analyzes the user's emotional state from the video data, determining whether they are tense, relaxed, enjoying themselves, etc.

[0690] AI-generated questions

[0691] The AI ​​on the server generates appropriate questions based on the results of voice and video analysis and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The question is then sent to the device and displayed to the user.

[0692] User answers the question

[0693] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[0694] The server analyzes and evaluates the answers

[0695] The server converts the voice response back into text and analyzes the video. It also uses an emotion engine to analyze the user's emotional state. For example, it may evaluate the user's speaking speed as appropriate, but their eyes are not looking at the camera and their emotional state seems tense.

[0696] Generating and displaying feedback

[0697] The server generates feedback based on the evaluation results. For example, it might say, "Your appeal to the audience will increase if you look more closely at the camera. Also, try taking deep breaths to relax." The feedback is sent to the device and displayed to the user.

[0698] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[0699] The processing flow will be explained below.

[0700] Step 1:

[0701] The user starts the dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[0702] Step 2:

[0703] The server inputs the received voice data into a speech recognition engine to convert the speech into text, which is used to identify what the user is saying.

[0704] Step 3:

[0705] The server inputs the received video data into a video analysis engine, which analyzes the user's facial expressions, gaze direction, and gestures. The video analysis engine identifies the user's facial expressions, gaze direction, hand and body movements, etc.

[0706] Step 4:

[0707] The server uses an emotion engine to recognize the user's emotions from the video data, for example, determining whether the user is nervous, relaxed, or having fun.

[0708] Step 5:

[0709] The AI ​​on the server generates appropriate questions based on the results of voice, video, and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The content and difficulty of the questions are adjusted according to the user's emotional state.

[0710] Step 6:

[0711] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0712] Step 7:

[0713] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[0714] Step 8:

[0715] The server inputs the voice data of the response into a speech recognition engine and converts it into text. It also analyzes the video data of the response with a video analysis engine to identify the user's facial expressions, eye movements, and gestures. The server then uses an emotion engine to analyze the user's emotional state.

[0716] Step 9:

[0717] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may provide an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera. Also, your emotional state seems tense."

[0718] Step 10:

[0719] The server generates feedback based on the evaluation results. The feedback includes specific improvements and advice. For example, advice such as "Turn your eyes more closely into the camera. Also, try taking deep breaths to relax" is generated. The feedback is sent from the server to the device and displayed to the user.

[0720] Through this specific processing step, users can effectively learn and improve their presentation skills. Multifaceted feedback, including emotional state, is provided, allowing users to gain a deeper understanding of how their presentations are perceived.

[0721] Example 2

[0722] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0723] Conventional presentation training systems are limited to capturing and analyzing audio and video data, and are unable to provide feedback that takes into account the user's emotional state. This limits the extent to which users can improve their presentation skills. Specifically, the user's emotional state, such as tension or relaxation, is not reflected in the evaluation or feedback of the presentation, making it difficult to provide training that closely resembles a real presentation situation.

[0724] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0725] In this invention, the server includes means for capturing audio and video to start a presentation by a user, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for an AI model to generate appropriate questions based on the analyzed data, means for transmitting the generated questions to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation, means for generating feedback based on the evaluation results and displaying them to the user, means for recognizing and analyzing the user's emotions, and means for adjusting the content of the feedback based on the user's emotional state, thereby enabling multifaceted feedback that takes the user's emotional state into consideration.

[0726] "User" refers to an individual or group that makes a presentation using a dedicated application.

[0727] "Terminal" refers to a device such as a PC, smartphone, or tablet that a user uses when giving a presentation.

[0728] "Means for capturing audio and video" refers to the ability to record audio and video data of a user using the microphone and camera installed on the device.

[0729] "Means for transmitting audio and video to a server" refers to a technology for transmitting captured audio and video data to a server via a network in real time.

[0730] "Means for converting speech to text" refers to a function that utilizes a speech recognition engine to convert captured speech data into corresponding text data.

[0731] "Means for analyzing video and identifying a user's facial expressions, gaze direction, and gestures" refers to a technology that uses a video analysis engine to detect and analyze a user's facial expressions, gaze direction, and gestures.

[0732] "Means for an AI model to generate appropriate questions" refers to the function of automatically generating appropriate questions for a user using a generative AI model based on the results of audio and video analysis.

[0733] "Means for transmitting the generated question to the terminal and displaying it to the user" refers to a function for transmitting a question generated by the server to the terminal and displaying it on the screen of the terminal.

[0734] "Means for recapturing the user's answers and sending them to the server" refers to a function for recapturing audio and video data when the user gives an answer and sending that data to the server.

[0735] "Means for analyzing the user's answers and evaluating the presentation" refers to a technology for analyzing audio and video data and evaluating the content of the user's presentation.

[0736] "Means for generating feedback based on the evaluation results and displaying it to the user" refers to a function for generating useful feedback based on the evaluation results of the presentation, transmitting it to the terminal, and displaying it to the user.

[0737] "Means for recognizing and analyzing user emotions" refers to technology that recognizes and analyzes a user's emotional state from video data.

[0738] The "means for adjusting the feedback content based on the user's emotional state" refers to a function for appropriately adjusting and providing the feedback content based on the emotion recognition result.

[0739] The present invention is a system for improving users' presentation skills, and in particular, by combining emotion recognition technology, it is possible to provide feedback based on the user's emotional state, allowing users to acquire more effective presentation skills.

[0740] First, the user launches a dedicated application on a device such as a PC or smartphone. When the user presses the "Start Presentation" button, the device's built-in microphone and camera start up and begin capturing audio and video data. The captured audio and video data is then sent to a server in real time via the Internet.

[0741] The server converts the received voice data into text data in real time using the Google Cloud Speech-to-Text API. At the same time, the received video data is analyzed using OpenCV. The video analysis engine detects and analyzes the user's facial expressions, eye direction, and gestures. This analysis allows the system to understand the context in which the user is speaking.

[0742] Furthermore, the server uses the Microsoft Azure Emotion API to analyze the user's emotional state from the video data. Based on facial expressions, eye movements, and changes in facial muscles, it can determine whether the user is tense, relaxed, or enjoying themselves. For example, the level of tension and stress can also be calculated.

[0743] The server uses OpenAI GPT-4 to generate appropriate questions based on these analysis results. The generated questions are sent from the server to the device and displayed on the user's screen. For example, a question such as "Please tell me specifically about risk management for this project" may be displayed.

[0744] The user begins to answer the displayed question. The device again uses the microphone and camera to capture audio and video data of the user's answer. The captured data is sent to the server, which again performs speech recognition and video analysis. The answer is converted to text using the Google Cloud Speech-to-Text API, and the video analysis engine analyzes facial expressions, eye movements, and gestures. The emotion engine is also used to analyze the user's emotional state.

[0745] The server comprehensively evaluates this data and makes a rating based on the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, a specific rating may be given such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera and your emotional state is tense."

[0746] Based on the evaluation results, the server generates feedback, which includes specific areas for improvement and advice. For example, advice such as "You can appeal more to the audience by looking more closely at the camera. Also, it would be a good idea to take deep breaths to relieve tension" is provided. The generated feedback is sent from the server to the device and displayed to the user.

[0747] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[0748] Specific examples

[0749] Prompt Sentence Examples

[0750] "Please tell me more about the risk management for this project."

[0751] Example of user analysis results

[0752] Speech text: "The first step in managing the risks of this project is..."

[0753] Facial expression analysis result: nervous

[0754] Eye analysis results: Not facing the camera

[0755] Emotion recognition result: tension

[0756] Feedback example

[0757] "Your speech is clear, but it would be better if you looked directly into the camera. To relax, try taking a deep breath before you start speaking."

[0758] In this way, the system can improve users' presentation skills in a variety of ways.

[0759] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0760] Step 1:

[0761] User starts presenting

[0762] The user launches the dedicated app on a device such as a PC or smartphone and presses the "Start Presentation" button. The input is the user's operation (pressing the button), which triggers the device's microphone and camera to start up. The output is the device starting to capture audio and video data.

[0763] Step 2:

[0764] Sending captured data

[0765] The terminal transmits the captured audio and video data to the server in real time. The input is the captured audio and video data, which are transmitted to the server via the network. The output is the real-time audio and video streams that arrive at the server.

[0766] Step 3:

[0767] Converting audio data to text

[0768] The server converts the received audio stream into text data using the Google Cloud Speech-to-Text API. The input is the received audio data, which is then subjected to speech recognition to generate text data. The output is text data that shows what the user said.

[0769] Step 4:

[0770] Video data analysis

[0771] The server analyzes the received video stream using OpenCV. The input is the received video data, and video analysis is performed on this data to identify the user's facial expressions, gaze, and gestures. The output is the analysis result data on the user's facial expressions, gaze, and gestures.

[0772] Step 5:

[0773] emotion recognition

[0774] The server uses the Microsoft Azure Emotion API to analyze the user's emotional state from video data. The input is the video analysis results data, and emotion recognition is performed based on this data. The output is data on the user's emotional state, such as whether they are tense or relaxed.

[0775] Step 6:

[0776] Question Generation

[0777] The server uses OpenAI GPT-4 to generate appropriate questions based on the results of voice and video analysis and emotion recognition. The input is voice text data, analysis result data, and emotion recognition data, and the question is generated based on these. The output is the generated question.

[0778] Step 7:

[0779] Show Questions

[0780] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and this data is sent to the terminal via the network. The output is the question displayed on the user's application screen. Example: "Please tell me specifically about risk management for this project."

[0781] Step 8:

[0782] User response capture

[0783] The user checks the displayed question and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The input is the audio and video data of the user's answer, which is captured and sent to the server. The output is the audio and video data that reaches the server.

[0784] Step 9:

[0785] Analysis of responses

[0786] The server converts the retransmitted voice data into text using a voice recognition engine, and analyzes the video data using a video analysis engine. The input is the user's response voice and video data, which are then subjected to voice recognition and video analysis. The output is the text of the response and the analysis results of the user's facial expressions, gaze, and gestures.

[0787] Step 10:

[0788] Emotion recognition when answering

[0789] The server again uses the Microsoft Azure Emotion API to analyze the user's emotional state at the time of answering. The input is the video analysis result data, and emotion recognition is performed based on this. The output is data indicating the user's emotional state.

[0790] Step 11:

[0791] Evaluation of presentation content

[0792] The server evaluates the presentation based on the content of the answers, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. The input is audio-text data, video analysis data, and emotion recognition data, and a comprehensive evaluation is made based on these. The output is the evaluation result of the user's presentation.

[0793] Step 12:

[0794] Generate feedback

[0795] Based on the evaluation results, the server generates feedback including specific areas for improvement and advice. The input is the evaluation result data, and feedback statements are generated based on this. The output is the generated feedback statements. Example: "Your appeal to the audience will increase if you look more closely at the camera. Try taking deep breaths to relax."

[0796] Step 13:

[0797] View Feedback

[0798] The server sends the generated feedback to the terminal and displays it on the user's screen. The input is the generated feedback sentence, and this data is sent to the terminal via the network. The output is the feedback sentence displayed on the user's application screen.

[0799] The above is the flow of processing for this system program. We have explained the specific operations performed at each step, as well as the processing or calculation of input and output data.

[0800] (Application example 2)

[0801] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0802] To improve the customer service skills of store staff, practice in a realistic environment is necessary. However, conventional methods provide limited feedback, and areas for improvement are not clearly identified. Furthermore, training does not take into account the emotional state of staff, and emotional upset during customer service can affect customer satisfaction. To address these issues, a system is needed that analyzes staff's customer service skills from multiple angles and provides real-time feedback.

[0803] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0804] In this invention, the server includes means for capturing audio and video to allow a user to start presentation or customer service training, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for recognizing the user's emotions in real time and generating appropriate questions using AI based on the analyzed data, means for transmitting the generated questions to a terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or customer service, and means for generating feedback based on the evaluation results and displaying them to the user. This allows staff to improve their customer service skills in real time while receiving multifaceted feedback including facial expressions and emotional information at the customer service site.

[0805] An "audio capture means" is a device or software for capturing audio data.

[0806] "Video capture means" refers to a device or software for acquiring video data.

[0807] A "server" is a computer system or computer on a network that analyzes and stores data.

[0808] "Real-time conversion means" is a technology or device that converts speech into text almost instantly.

[0809] "Video analysis means" refers to a technique or device that analyzes video data and extracts features.

[0810] The "facial expression identification means" is a technique or device that identifies the facial expression of a user from video data.

[0811] The "gaze direction identification means" is a technique or device that identifies the direction of a user's gaze from video data.

[0812] The "gesture recognition means" is a technique or device that recognizes a user's gesture action from video data.

[0813] "Emotion recognition means" refers to a technique or device that recognizes the user's emotional state from video data and audio data.

[0814] A "question generator" is a technique or device that generates appropriate questions based on the analyzed data.

[0815] The "question display means" is a technique or device that visually displays the generated question to the user.

[0816] An "answer capture means" is a device or software for recapturing a user's answers.

[0817] The "evaluation means" is a technology or device that analyzes and evaluates the user's answers and the performance of customer service and presentation.

[0818] The "feedback generation means" is a technique or device that generates feedback to the user based on the evaluation results.

[0819] A "feedback display means" is a technique or device that visually displays the generated feedback to the user.

[0820] This invention relates to a system for improving customer service skills in brick-and-mortar stores, which has the function of capturing audio and video data and analyzing the user's facial expressions, gaze, gestures, and emotional state. Based on the analysis results, AI generates appropriate questions and provides feedback to the user.

[0821] The server includes an audio capture means, a video capture means, a real-time conversion means, a video analysis means, a facial expression identification means, a gaze identification means, a gesture identification means, an emotion recognition means, a question generation means, a question display means, an answer capture means, an evaluation means, and a feedback generation means.

[0822] The user launches a dedicated application on a smartphone or other device and presses the "Start Training" button. The device then captures audio and video data and sends it to the server in real time. The server then converts the audio into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text), analyzes the user's facial expressions, gaze, and gestures using a video analysis engine (e.g., OpenCV), and recognizes the user's emotional state using an emotion recognition engine (e.g., Affectiva).

[0823] Based on the analyzed data, the AI ​​on the server uses a question generation engine to generate appropriate questions. For example, a question such as "Please tell me specifically about risk management for this project" may be generated. The generated question is sent to the user's device and displayed on the screen.

[0824] The user answers the displayed questions. The device captures the answer again as audio and video data and sends it to the server. The server receives the answer data and analyzes it again using its voice recognition engine and video analysis engine.

[0825] The server evaluates the results of the analysis and generates feedback using a feedback generation engine, including specific suggestions for improvement such as, "Your speaking speed is appropriate, but you don't seem to be looking at the camera enough. You should be more aware of the camera."

[0826] The generated feedback is sent to the user's device and displayed in real time, allowing the user to improve their customer service skills in multiple ways.

[0827] Prompt Sentence Examples

[0828] "We have a new product in stock. Please explain in detail how it differs from similar products we have available so far, and what you recommend it for customers."

[0829] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0830] Step 1:

[0831] The user launches the dedicated application on a device such as a smartphone and presses the "Start Training" button. In this step, the device starts capturing audio and video data and sends this data to the server in real time. The input is the audio and video made by the user, and the output is the captured data sent to the server.

[0832] Step 2:

[0833] The server inputs the received voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into text in real time. The input is the captured voice data, and the output is the converted text data. The specific operation involves signal processing of the voice data and generation of text data.

[0834] Step 3:

[0835] The server inputs the received video data into a video analysis engine (e.g., OpenCV) to analyze facial expressions, gaze, and gestures. The input is the captured video data, and the output is the analyzed facial expression data, gaze data, and gesture data. The specific operation is to analyze the video frames and extract features.

[0836] Step 4:

[0837] The server uses an emotion recognition engine (e.g., Affectiva) to recognize the user's emotional state from video and audio data. The input is video and audio data, and the output is the recognized emotional state data. The specific operation is to classify and label the emotional state.

[0838] Step 5:

[0839] The AI ​​on the server generates appropriate questions based on the analyzed data and emotional state. The input is the analyzed text data, facial expression data, eye gaze data, gesture data, and emotional state data, and the output is the generated question. The specific operations are natural language generation and text generation.

[0840] Step 6:

[0841] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and the output is the question displayed on the terminal screen. The specific operations are data transmission and UI update.

[0842] Step 7:

[0843] The user answers the displayed questions. The device recaptures the answers as audio and video data and sends them to the server. The input is the user's audio and video answers, and the output is the captured data sent to the server.

[0844] Step 8:

[0845] The server then inputs the received voice data into a voice recognition engine and converts it into text. The input is the captured response voice, and the output is the converted text data. The specific operations are voice recognition and text conversion.

[0846] Step 9:

[0847] The server inputs the received video data into the video analysis engine and analyzes facial expressions, gaze, and gestures. The input is the captured response video, and the output is analyzed facial expression data, gaze data, and gesture data. The specific operations are video analysis and feature extraction.

[0848] Step 10:

[0849] The server evaluates the answer data and generates feedback using a feedback generation engine. The input is the analyzed answer data, and the output is the generated feedback. The specific operations are evaluation analysis and feedback generation.

[0850] Step 11:

[0851] The server sends the generated feedback to the device and displays it to the user in real time. The input is the generated feedback, and the output is the feedback displayed on the device screen. The specific operations are data transmission and UI update.

[0852] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0853] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0854] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0855] [Third embodiment]

[0856] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0857] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0858] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0859] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0860] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0861] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0862] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0863] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0864] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0865] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0866] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0867] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0868] This invention relates to an AI presentation training system that allows users to improve their presentation skills. This system can capture audio and video, and provide real-time analysis and feedback, enabling effective practice in a realistic environment.

[0869] System Configuration

[0870] 1. User Interface

[0871] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[0872] 2. Server Functions

[0873] The server receives the captured audio and video data and performs the following processes:

[0874] 1. Audio analysis:

[0875] The server uses a speech recognition engine to convert the user's speech into text in real time.

[0876] 2. Video analysis:

[0877] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[0878] 3. Question generation:

[0879] The AI ​​on the server generates appropriate questions based on the analyzed data. The questions are generated to match the content of the user's presentation, and the AI's settings take into account personality and expertise.

[0880] 4. Question display:

[0881] The question is sent from the server to the terminal and displayed on the user's screen. The user checks the question and begins answering.

[0882] 5. Answer analysis:

[0883] The user's responses are also captured as audio and video and sent to the server, where they are converted into text using a speech recognition engine and analyzed using a video analysis engine for facial expressions, eye movements, and gestures.

[0884] 6. Rating and Feedback:

[0885] The server comprehensively evaluates the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. Feedback is generated based on the evaluation results and sent to the terminal to be displayed to the user.

[0886] Natural language explanation of program processing

[0887] User starts presenting

[0888] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[0889] The server receives and analyzes the data

[0890] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[0891] AI-generated questions

[0892] The AI ​​on the server generates appropriate questions based on the analyzed data. For example, it might ask, "Please tell me specifically about risk management for this project." The questions are then sent to the device and displayed to the user.

[0893] User answers the question

[0894] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[0895] The server analyzes and evaluates the answers

[0896] The server converts the audio responses back into text and analyzes the video. The server evaluates the user's speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, it may evaluate the user's speaking speed as appropriate, but their voice volume as low.

[0897] Generating and displaying feedback

[0898] The server generates feedback based on the evaluation results, such as "Your speaking speed is appropriate, but your voice volume is low. You should speak a little louder." The feedback is sent to the device and displayed to the user.

[0899] Through this system, users can effectively improve their presentation skills while receiving continuous feedback. Furthermore, repeated practice allows for training in a more realistic environment.

[0900] The processing flow will be explained below.

[0901] Step 1:

[0902] The user starts a dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[0903] Step 2:

[0904] The server inputs the received voice data into a speech recognition engine and converts the voice into text, which is then used to analyze the content of the presentation.

[0905] Step 3:

[0906] The server inputs the received video data into a video analysis engine to analyze the user's facial expressions, gaze, and gestures, for example, to determine whether the user is smiling, looking at the camera, or moving their hands.

[0907] Step 4:

[0908] The AI ​​on the server generates appropriate questions based on the results of audio and video analysis. The questions are related to the content of the presentation, and the difficulty and perspective are adjusted according to the AI ​​settings.

[0909] Step 5:

[0910] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[0911] Step 6:

[0912] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[0913] Step 7:

[0914] The server inputs the voice data of the response into a voice recognition engine and converts it into text. It also analyzes the video data of the response using a video analysis engine. The server also analyzes facial expressions, eye movements, and gestures when responding.

[0915] Step 8:

[0916] The server evaluates the presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, the server may give an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not facing the camera."

[0917] Step 9:

[0918] The server generates feedback based on the evaluation results, including specific areas for improvement and advice, such as "If you look more closely at the camera, you will be more appealing to viewers."

[0919] Step 10:

[0920] The generated feedback is sent from the server to the device and displayed on the device screen. The user can review the feedback and use it to practice their next presentation to improve the points pointed out.

[0921] Through this specific processing step, users can effectively learn and improve their presentation skills.

[0922] Example 1

[0923] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0924] Conventional systems for improving presentation skills have the problem that it is difficult for users to receive immediate, specific feedback even when they practice, and because they do not perform real-time analysis, it is difficult to practice in a realistic environment.Furthermore, conventional systems are unable to comprehensively evaluate the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. during a presentation, and therefore can only provide limited feedback.

[0925] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0926] In this invention, the server includes means for converting voice to text, means for analyzing video to identify the user's facial expressions, eye movements, and gestures, means for generating questions using artificial intelligence based on the analyzed data, means for comprehensively evaluating data related to the user's presentation, means for providing sequential feedback based on the evaluation results, and means for enabling the user to practice in a manner close to reality to improve their presentation skills. This allows the user to receive specific feedback in real time and receive advice based on a comprehensive evaluation.

[0927] "Audio capturing means" refers to a microphone or other audio input device that captures the user's speech.

[0928] "Video capture means" refers to a camera or other video input device that captures the user's visual movements and expressions.

[0929] "Server" refers to a computer system for receiving and analyzing captured audio and video data.

[0930] "Means for converting voice to text" refers to technology that uses a voice recognition engine to convert voice data into text data.

[0931] "Means for analyzing video" refers to technology that processes and analyzes video data to identify a user's facial expressions, gaze, gestures, etc.

[0932] "Artificial intelligence" refers to a computer program or system that uses machine learning and natural language processing techniques to make inferences and judgments.

[0933] "Means for generating questions" refers to artificial intelligence technology for creating appropriate questions based on analyzed data.

[0934] "Means for comprehensively evaluating data related to a user's presentation" refers to technology for simultaneously analyzing and evaluating multiple elements, such as the user's speaking speed, voice volume, choice of words, eye contact, facial expressions, and gestures.

[0935] "Means for providing feedback" refers to technology that notifies users of specific improvements and advice in real time based on the analysis and evaluation results.

[0936] "Realistic practice tools" refers to systems that allow users to effectively improve their skills under conditions similar to those experienced in a real presentation.

[0937] This invention relates to a system for enabling users to improve their presentation skills. The system begins when a user launches a dedicated application on a device such as a PC or smartphone and starts a presentation.

[0938] Hardware and Software Configuration

[0939] 1. Hardware

[0940] Voice capture microphone: A microphone is used to capture the user's voice. Examples include a typical condenser microphone or a headset microphone.

[0941] Video capture camera: A camera is used to capture the user's video. Examples include a webcam built into a PC or an external camera.

[0942] 2. Software

[0943] Speech recognition engine: Engines such as Google Cloud Speech-to-Text and IBM Watson Speech to Text are used to convert speech into text.

[0944] Video analysis engine: OpenCV and Google Cloud Vision are used as the engine to analyze video data and identify the user's facial expressions, gaze, and gestures.

[0945] Generative AI models: GPT-3 and BERT are used as AI models to generate appropriate questions based on analyzed data.

[0946] System Operation

[0947] Start your presentation

[0948] The user starts the dedicated application and presses the "Start Presentation" button on their device, such as a PC or smartphone, which then uses the microphone and camera to capture audio and video and transmits the data to the server in real time.

[0949] Data analysis

[0950] The server converts the received voice data into text using a speech recognition engine (Google Cloud Speech-to-Text or IBM Watson Speech to Text), and simultaneously analyzes the video data using a video analysis engine (OpenCV or Google Cloud Vision). For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data, and the video analysis engine analyzes the user's facial expressions and eye direction as they speak.

[0951] question generation

[0952] Based on the analyzed data, a generative AI model (such as GPT-3 or BERT) on the server generates an appropriate question. For example, a question such as "Please tell me specifically about risk management for this project" is generated. The generated question is sent from the server to the device and displayed on the user's screen.

[0953] Answers to questions

[0954] The user checks the question displayed on the device screen and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The server again analyzes this data using its voice recognition engine and video analysis engine to evaluate the user's answer.

[0955] Generating and Providing Feedback

[0956] The server performs a comprehensive evaluation of the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. For example, it obtains a specific evaluation result such as "Your speaking speed is appropriate, but your voice volume is low." Based on this evaluation result, the server generates specific feedback and displays it on the user's screen. For example, it may provide feedback such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[0957] Specific examples

[0958] Prompt Sentence Examples

[0959] Example presentation: "The goal of this project is to increase sales by 20%."

[0960] Example question: "What strategies do you have in mind to achieve this goal?"

[0961] Through this system, users can improve their presentation skills while receiving specific feedback in real time, and by repeatedly practicing, they can improve the quality of their presentations.

[0962] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0963] Step 1:

[0964] The user starts the dedicated application and presses the "Start Presentation" button, which causes the device to capture the user's voice and video and send the data to the server.

[0965] Input: User's audio and video

[0966] Output: Audio and video data sent to the server

[0967] Specific operation: The device's microphone and camera capture audio and video in real time, encode the data, and send it to the server.

[0968] Step 2:

[0969] The server analyzes the received voice data using a speech recognition engine and converts it into text, for example, using Google Cloud Speech-to-Text.

[0970] Input: Audio data sent to the server

[0971] Output: Text data

[0972] Specific operation: The speech recognition engine analyzes the voice data and converts the user's speech into text. For example, it extracts the utterance "The goal of this project is..." as text data.

[0973] Step 3:

[0974] The server analyzes the video data using a video analysis engine, such as OpenCV.

[0975] Input: Video data sent to the server

[0976] Output: Analysis results of user's facial expressions, gaze, and gestures

[0977] Specific operation: The video analysis engine analyzes the video data frame by frame and identifies the user's facial expressions (e.g., smiling or surprised), gaze (e.g., looking toward the camera), and gestures (e.g., hand movements and gestures).

[0978] Step 4:

[0979] A generative AI model on the server generates appropriate questions based on the results of audio and video analysis, using, for example, GPT-3.

[0980] Input: Text data and video analysis results

[0981] Output: Generated question text

[0982] Specific behavior: The generative AI model generates relevant questions based on the analysis results, such as "Please tell me specifically about risk management for this project."

[0983] Step 5:

[0984] The server sends the generated question to the terminal and displays it on the user's screen.

[0985] Input: Generated question text

[0986] Output: The question displayed on the user's screen

[0987] Specific operation: The server sends the generated question text to the terminal, and the dedicated application displays the received question text on the user's screen.

[0988] Step 6:

[0989] The user checks the questions displayed on the terminal and begins to answer them. The terminal again captures audio and video and sends them to the server.

[0990] Input: User's answer audio and video

[0991] Output: Answer audio and video data sent to the server

[0992] Specific operation: When the user answers a question, the device's microphone and camera again capture audio and video and send the data to the server.

[0993] Step 7:

[0994] The server converts the response voice into text again using a voice recognition engine and analyzes the video again.

[0995] Input: Answer audio and video data sent to the server

[0996] Output: Answer text and video analysis results

[0997] Specific operation: The speech recognition engine converts the answer into text, and the video analysis engine reanalyzes the user's facial expressions, gaze, and gestures. For example, the speech "Risk management is..." is extracted as text, and the facial expressions and gaze are reanalyzed.

[0998] Step 8:

[0999] The server will provide an overall rating and generate feedback.

[1000] Input: Answer text and video analysis results

[1001] Output: Feedback text

[1002] Specific behavior: The server comprehensively evaluates the text data and video analysis results and generates specific feedback, such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[1003] Step 9:

[1004] The server sends the generated feedback to the terminal and displays it on the user's screen.

[1005] Input: Feedback text

[1006] Output: Feedback displayed on the user's screen

[1007] Specific operation: The server sends the generated feedback text to the terminal, and the dedicated application displays the received feedback on the user's screen.

[1008] (Application example 1)

[1009] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1010] In modern factories and production sites, workers are required to have the skills to effectively deliver various presentations, such as introducing new products and providing safety training. These skills contribute to worker growth and improved productivity, but traditional educational methods often struggle to provide real-time feedback, and one-way lectures often fail to provide effective instruction. The present invention aims to provide a system that allows workers to efficiently improve their presentation and education / training skills in the field.

[1011] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1012] In this invention, the server includes means for capturing audio and images to start a user's presentation or training, means for transmitting the captured audio and images to the server, means for converting the audio to text in real time in the server, means for analyzing the images in the server and identifying the user's facial expressions, line of sight, and gestures, means for an AI to generate appropriate prompts based on the analyzed data, means for transmitting the generated prompts to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or training, and means for generating feedback based on the evaluation results and displaying them to the user. This allows workers to improve their presentation and training skills while receiving feedback in real time.

[1013] "Users" are workers who use this system to give presentations or provide training.

[1014] "Voice capture" is the act of recording a user's speech as voice data.

[1015] "Video capture" is the act of recording a user's posture and movements as video data.

[1016] A "server" is a computer system that receives and analyzes audio and video data and provides feedback based on the results.

[1017] "Speech-to-text" is the process of converting captured voice data into text data.

[1018] "Image analysis" is the process of identifying a user's facial expressions, gaze, and gestures based on captured video data.

[1019] "Prompts" are appropriate questions or instructions generated by AI based on analyzed data.

[1020] "Feedback" refers to guidance and advice provided to users based on the results of evaluation of their presentations and training.

[1021] "Real-time" refers to the immediacy of time in which the results of a user action are provided as analysis and feedback immediately after the action is taken.

[1022] The present invention is an AI system that enables users to effectively deliver presentations or educational training, and includes functions for capturing audio and video, analyzing them in real time, and generating feedback.

[1023] System Configuration

[1024] 1. User Interface

[1025] The user starts a dedicated application on a device (smartphone, tablet, PC, etc.) and starts a presentation or training session. By pressing the start button, the device captures audio and video and transmits the data to the server in real time.

[1026] 2. Server Functions

[1027] Audio Analysis:

[1028] The server uses a speech recognition engine (e.g., Google Speech Recognition API) to convert the voice data into text. For example, if a user says, "The operating procedure for this machine is...", the server extracts this as text data.

[1029] Image analysis:

[1030] The server uses a video analysis engine (e.g., OpenCV) to analyze the user's facial expressions, gaze, and gestures, for example, to evaluate whether the user is smiling when speaking and in what direction their eyes are pointing.

[1031] Prompt generation:

[1032] Based on the analyzed data, the AI ​​generates appropriate prompts (questions or instructions), such as "Please explain the emergency shutdown procedure for this machine."

[1033] Question display:

[1034] The generated prompts are sent from the server to the terminal and displayed to the user, who then initiates an answer based on the prompts.

[1035] Response analysis and evaluation:

[1036] The user's answers are also captured as audio and video and sent to the server. The server converts the audio back into text and analyzes the video to evaluate the answer. For example, the server may evaluate the answer as "appropriate, but the user's gaze is not focused."

[1037] Feedback generation and display:

[1038] Based on the evaluation results, the server generates feedback, such as "It would be more effective if you directed your gaze more at the other person," and sends it to the device and displays it to the user.

[1039] Program processing explanation

[1040] The specific hardware and software used in this system are as follows:

[1041] Hardware:

[1042] Camera and microphone: Use the device's built-in or external camera and microphone.

[1043] Server: A dedicated server for high-performance data analysis.

[1044] software:

[1045] OpenCV: A library for video capture and analysis.

[1046] SpeechRecognition: An engine for converting speech to text.

[1047] Flask: A framework for server-side data processing.

[1048] Generative AI model: An AI algorithm for generating prompts (questions or instructions).

[1049] Examples of specific examples and prompts

[1050] Consider a case where a user is giving a presentation on how to operate a new machine in a factory. When the user presses the "start button," the camera and microphone are activated to capture audio and video, which are then sent to the server. The server analyzes this and generates a prompt such as "Please explain the emergency stop procedure for this machine," which is displayed on the terminal. When the user responds to the prompt and begins their explanation, their answer is captured again and sent to the server. The server evaluates the user's answer and provides feedback such as "It would be more effective if you paid more attention to the other person."

[1051] Specific prompt examples:

[1052] "Please explain the emergency shutdown procedure for this machine."

[1053] "Please elaborate on the content of the next slide and explain the risk factors that should be considered."

[1054] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1055] Step 1:

[1056] The user launches the dedicated application on their device and presses the "Start Presentation" button. This activates the device's camera and microphone, and audio and video capture begins. The input is the user's audio and video, and the output is the captured audio and video data. Specifically, the device's application starts the capture function of the specified camera and microphone.

[1057] Step 2:

[1058] The device transmits the captured audio and video data to the server in real time. Specifically, the device uploads the data to the server via a network. The data is often compressed and encrypted. The input is the captured audio and video data, and the output is the data transmitted to the server.

[1059] Step 3:

[1060] The server converts the received voice data into text using a voice recognition engine (e.g., Google Speech Recognition API). The input is voice data, and the output is text data. Specifically, the server calls the voice recognition engine, analyzes the voice data, and converts it into text.

[1061] Step 4:

[1062] The server analyzes the received video data using a video analysis engine (e.g., OpenCV) to identify the user's facial expressions, gaze, and gestures. The input is video data, and the output is analyzed feature data of the user's facial expressions, gaze, and gestures. Specifically, the video analysis engine processes the video frame by frame, identifies facial features, and extracts their features.

[1063] Step 5:

[1064] The server uses a generative AI model to generate appropriate prompts based on the analyzed voice text and video data. The input is text data and characteristic data on the user's facial expressions, eye movements, and gestures, and the output is the generated prompt. Specifically, the generative AI model analyzes this data and generates appropriate questions and instructions.

[1065] Step 6:

[1066] The server sends the generated prompt to the terminal, and the terminal displays the prompt on the user's screen. The input is the prompt sent from the server, and the output is the prompt displayed on the user's screen. Specifically, the server sends the generated prompt to the terminal via the network, and the terminal receives it and displays it on the screen.

[1067] Step 7:

[1068] The user responds based on the prompts displayed on the device screen, and the device again captures audio and video and sends them to the server. The input is the user's new audio and video, and the output is the captured data sent to the server. Specifically, the device again captures audio and video using the camera and microphone and sends them to the server.

[1069] Step 8:

[1070] The server analyzes the received audio and video again, converts the user's audio response into text, and analyzes the video to evaluate it. The input is the newly received audio and video data, and the output is the analyzed text data and the evaluation results. Specifically, the server uses a voice recognition engine to convert the audio into text, and a video analysis engine to analyze the user's facial expressions, gaze, and gestures.

[1071] Step 9:

[1072] The server generates feedback based on the evaluation results and sends it to the terminal. The terminal displays the feedback to the user. The input is the evaluation results, and the output is the generated feedback sentence and its display to the user by the terminal that received it. Specifically, the server generates feedback and sends it to the terminal via the network, and the terminal displays the feedback on its screen.

[1073] Example prompt sentence:

[1074] "Please explain the emergency shutdown procedure for this machine."

[1075] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1076] The present invention relates to an AI presentation trainer system that helps users improve their presentation skills. The system combines audio and video capture with an emotion engine that recognizes the user's emotions. This provides more detailed feedback based on the user's emotions, which can be expected to improve presentation skills.

[1077] System Configuration

[1078] 1. User Interface

[1079] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[1080] 2. Server Functions

[1081] The server receives the captured audio and video data and performs the following processes:

[1082] 1. Audio analysis:

[1083] The server uses a speech recognition engine to convert the user's speech into text in real time.

[1084] 2. Video analysis:

[1085] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[1086] 3. Emotion recognition:

[1087] The server uses an emotion engine to recognize the user's emotions, for example, determining whether the user is nervous or relaxed during a presentation.

[1088] 4. Question generation:

[1089] The AI ​​on the server generates appropriate questions based on the analyzed data and emotion recognition results, and adjusts the content and difficulty of the questions depending on the emotional state.

[1090] 5. Question display:

[1091] The question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[1092] 6. Answer analysis:

[1093] The user's answers are also captured as audio and video and sent to the server. The server then converts the answers into text using a speech recognition engine, analyzes facial expressions, eye movements, and gestures using a video analysis engine, and analyzes the emotions expressed when answering using an emotion engine.

[1094] 7. Rating and Feedback:

[1095] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may evaluate the user's speaking speed as appropriate, but the user's eyes are not directed toward the camera and the user's emotional state is tense.

[1096] 8. Feedback Generation and Display:

[1097] The server generates feedback based on the evaluation results. The feedback includes specific areas for improvement and advice. For example, the server might generate advice such as, "You should look more closely at the camera. Also, take a deep breath and relax to reduce tension." The feedback is sent from the server to the device and displayed to the user.

[1098] Natural language explanation of program processing

[1099] User starts presenting

[1100] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[1101] The server receives and analyzes the data

[1102] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[1103] emotion recognition

[1104] The server's emotion engine analyzes the user's emotional state from the video data, determining whether they are tense, relaxed, enjoying themselves, etc.

[1105] AI-generated questions

[1106] The AI ​​on the server generates appropriate questions based on the results of voice and video analysis and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The question is then sent to the device and displayed to the user.

[1107] User answers the question

[1108] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[1109] The server analyzes and evaluates the answers

[1110] The server converts the voice response back into text and analyzes the video. It also uses an emotion engine to analyze the user's emotional state. For example, it may evaluate the user's speaking speed as appropriate, but their eyes are not looking at the camera and their emotional state seems tense.

[1111] Generating and displaying feedback

[1112] The server generates feedback based on the evaluation results. For example, it might say, "Your appeal to the audience will increase if you look more closely at the camera. Also, try taking deep breaths to relax." The feedback is sent to the device and displayed to the user.

[1113] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[1114] The processing flow will be explained below.

[1115] Step 1:

[1116] The user starts the dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[1117] Step 2:

[1118] The server inputs the received voice data into a speech recognition engine to convert the speech into text, which is used to identify what the user is saying.

[1119] Step 3:

[1120] The server inputs the received video data into a video analysis engine, which analyzes the user's facial expressions, gaze direction, and gestures. The video analysis engine identifies the user's facial expressions, gaze direction, hand and body movements, etc.

[1121] Step 4:

[1122] The server uses an emotion engine to recognize the user's emotions from the video data, for example, determining whether the user is nervous, relaxed, or having fun.

[1123] Step 5:

[1124] The AI ​​on the server generates appropriate questions based on the results of voice, video, and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The content and difficulty of the questions are adjusted according to the user's emotional state.

[1125] Step 6:

[1126] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[1127] Step 7:

[1128] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[1129] Step 8:

[1130] The server inputs the voice data of the response into a speech recognition engine and converts it into text. It also analyzes the video data of the response with a video analysis engine to identify the user's facial expressions, eye movements, and gestures. The server then uses an emotion engine to analyze the user's emotional state.

[1131] Step 9:

[1132] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may provide an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera. Also, your emotional state seems tense."

[1133] Step 10:

[1134] The server generates feedback based on the evaluation results. The feedback includes specific improvements and advice. For example, advice such as "Turn your eyes more closely into the camera. Also, try taking deep breaths to relax" is generated. The feedback is sent from the server to the device and displayed to the user.

[1135] Through this specific processing step, users can effectively learn and improve their presentation skills. Multifaceted feedback, including emotional state, is provided, allowing users to gain a deeper understanding of how their presentations are perceived.

[1136] Example 2

[1137] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1138] Conventional presentation training systems are limited to capturing and analyzing audio and video data, and are unable to provide feedback that takes into account the user's emotional state. This limits the extent to which users can improve their presentation skills. Specifically, the user's emotional state, such as tension or relaxation, is not reflected in the evaluation or feedback of the presentation, making it difficult to provide training that closely resembles a real presentation situation.

[1139] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1140] In this invention, the server includes means for capturing audio and video to start a presentation by a user, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for an AI model to generate appropriate questions based on the analyzed data, means for transmitting the generated questions to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation, means for generating feedback based on the evaluation results and displaying them to the user, means for recognizing and analyzing the user's emotions, and means for adjusting the content of the feedback based on the user's emotional state, thereby enabling multifaceted feedback that takes the user's emotional state into consideration.

[1141] "User" refers to an individual or group that makes a presentation using a dedicated application.

[1142] "Terminal" refers to a device such as a PC, smartphone, or tablet that a user uses when giving a presentation.

[1143] "Means for capturing audio and video" refers to the ability to record audio and video data of a user using the microphone and camera installed on the device.

[1144] "Means for transmitting audio and video to a server" refers to a technology for transmitting captured audio and video data to a server via a network in real time.

[1145] "Means for converting speech to text" refers to a function that utilizes a speech recognition engine to convert captured speech data into corresponding text data.

[1146] "Means for analyzing video and identifying a user's facial expressions, gaze direction, and gestures" refers to a technology that uses a video analysis engine to detect and analyze a user's facial expressions, gaze direction, and gestures.

[1147] "Means for an AI model to generate appropriate questions" refers to the function of automatically generating appropriate questions for a user using a generative AI model based on the results of audio and video analysis.

[1148] "Means for transmitting the generated question to the terminal and displaying it to the user" refers to a function for transmitting a question generated by the server to the terminal and displaying it on the screen of the terminal.

[1149] "Means for recapturing the user's answers and sending them to the server" refers to a function for recapturing audio and video data when the user gives an answer and sending that data to the server.

[1150] "Means for analyzing the user's answers and evaluating the presentation" refers to a technology for analyzing audio and video data and evaluating the content of the user's presentation.

[1151] "Means for generating feedback based on the evaluation results and displaying it to the user" refers to a function for generating useful feedback based on the evaluation results of the presentation, transmitting it to the terminal, and displaying it to the user.

[1152] "Means for recognizing and analyzing user emotions" refers to technology that recognizes and analyzes a user's emotional state from video data.

[1153] The "means for adjusting the feedback content based on the user's emotional state" refers to a function for appropriately adjusting and providing the feedback content based on the emotion recognition result.

[1154] The present invention is a system for improving users' presentation skills, and in particular, by combining emotion recognition technology, it is possible to provide feedback based on the user's emotional state, allowing users to acquire more effective presentation skills.

[1155] First, the user launches a dedicated application on a device such as a PC or smartphone. When the user presses the "Start Presentation" button, the device's built-in microphone and camera start up and begin capturing audio and video data. The captured audio and video data is then sent to a server in real time via the Internet.

[1156] The server converts the received voice data into text data in real time using the Google Cloud Speech-to-Text API. At the same time, the received video data is analyzed using OpenCV. The video analysis engine detects and analyzes the user's facial expressions, eye direction, and gestures. This analysis allows the system to understand the context in which the user is speaking.

[1157] Furthermore, the server uses the Microsoft Azure Emotion API to analyze the user's emotional state from the video data. Based on facial expressions, eye movements, and changes in facial muscles, it can determine whether the user is tense, relaxed, or enjoying themselves. For example, the level of tension and stress can also be calculated.

[1158] The server uses OpenAI GPT-4 to generate appropriate questions based on these analysis results. The generated questions are sent from the server to the device and displayed on the user's screen. For example, a question such as "Please tell me specifically about risk management for this project" may be displayed.

[1159] The user begins to answer the displayed question. The device again uses the microphone and camera to capture audio and video data of the user's answer. The captured data is sent to the server, which again performs speech recognition and video analysis. The answer is converted to text using the Google Cloud Speech-to-Text API, and the video analysis engine analyzes facial expressions, eye movements, and gestures. The emotion engine is also used to analyze the user's emotional state.

[1160] The server comprehensively evaluates this data and makes a rating based on the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, a specific rating may be given such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera and your emotional state is tense."

[1161] Based on the evaluation results, the server generates feedback, which includes specific areas for improvement and advice. For example, advice such as "You can appeal more to the audience by looking more closely at the camera. Also, it would be a good idea to take deep breaths to relieve tension" is provided. The generated feedback is sent from the server to the device and displayed to the user.

[1162] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[1163] Specific examples

[1164] Prompt Sentence Examples

[1165] "Please tell me more about the risk management for this project."

[1166] Example of user analysis results

[1167] Speech text: "The first step in managing the risks of this project is..."

[1168] Facial expression analysis result: nervous

[1169] Eye analysis results: Not facing the camera

[1170] Emotion recognition result: tension

[1171] Feedback example

[1172] "Your speech is clear, but it would be better if you looked directly into the camera. To relax, try taking a deep breath before you start speaking."

[1173] In this way, the system can improve users' presentation skills in a variety of ways.

[1174] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1175] Step 1:

[1176] User starts presenting

[1177] The user launches the dedicated app on a device such as a PC or smartphone and presses the "Start Presentation" button. The input is the user's operation (pressing the button), which triggers the device's microphone and camera to start up. The output is the device starting to capture audio and video data.

[1178] Step 2:

[1179] Sending captured data

[1180] The terminal transmits the captured audio and video data to the server in real time. The input is the captured audio and video data, which are transmitted to the server via the network. The output is the real-time audio and video streams that arrive at the server.

[1181] Step 3:

[1182] Converting audio data to text

[1183] The server converts the received audio stream into text data using the Google Cloud Speech-to-Text API. The input is the received audio data, which is then subjected to speech recognition to generate text data. The output is text data that shows what the user said.

[1184] Step 4:

[1185] Video data analysis

[1186] The server analyzes the received video stream using OpenCV. The input is the received video data, and video analysis is performed on this data to identify the user's facial expressions, gaze, and gestures. The output is the analysis result data on the user's facial expressions, gaze, and gestures.

[1187] Step 5:

[1188] emotion recognition

[1189] The server uses the Microsoft Azure Emotion API to analyze the user's emotional state from video data. The input is the video analysis results data, and emotion recognition is performed based on this data. The output is data on the user's emotional state, such as whether they are tense or relaxed.

[1190] Step 6:

[1191] Question Generation

[1192] The server uses OpenAI GPT-4 to generate appropriate questions based on the results of voice and video analysis and emotion recognition. The input is voice text data, analysis result data, and emotion recognition data, and the question is generated based on these. The output is the generated question.

[1193] Step 7:

[1194] Show Questions

[1195] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and this data is sent to the terminal via the network. The output is the question displayed on the user's application screen. Example: "Please tell me specifically about risk management for this project."

[1196] Step 8:

[1197] User response capture

[1198] The user checks the displayed question and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The input is the audio and video data of the user's answer, which is captured and sent to the server. The output is the audio and video data that reaches the server.

[1199] Step 9:

[1200] Analysis of responses

[1201] The server converts the retransmitted voice data into text using a voice recognition engine, and analyzes the video data using a video analysis engine. The input is the user's response voice and video data, which are then subjected to voice recognition and video analysis. The output is the text of the response and the analysis results of the user's facial expressions, gaze, and gestures.

[1202] Step 10:

[1203] Emotion recognition when answering

[1204] The server again uses the Microsoft Azure Emotion API to analyze the user's emotional state at the time of answering. The input is the video analysis result data, and emotion recognition is performed based on this. The output is data indicating the user's emotional state.

[1205] Step 11:

[1206] Evaluation of presentation content

[1207] The server evaluates the presentation based on the content of the answers, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. The input is audio-text data, video analysis data, and emotion recognition data, and a comprehensive evaluation is made based on these. The output is the evaluation result of the user's presentation.

[1208] Step 12:

[1209] Generate feedback

[1210] Based on the evaluation results, the server generates feedback including specific areas for improvement and advice. The input is the evaluation result data, and feedback statements are generated based on this. The output is the generated feedback statements. Example: "Your appeal to the audience will increase if you look more closely at the camera. Try taking deep breaths to relax."

[1211] Step 13:

[1212] View Feedback

[1213] The server sends the generated feedback to the terminal and displays it on the user's screen. The input is the generated feedback sentence, and this data is sent to the terminal via the network. The output is the feedback sentence displayed on the user's application screen.

[1214] The above is the flow of processing for this system program. We have explained the specific operations performed at each step, as well as the processing or calculation of input and output data.

[1215] (Application example 2)

[1216] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1217] To improve the customer service skills of store staff, practice in a realistic environment is necessary. However, conventional methods provide limited feedback, and areas for improvement are not clearly identified. Furthermore, training does not take into account the emotional state of staff, and emotional upset during customer service can affect customer satisfaction. To address these issues, a system is needed that analyzes staff's customer service skills from multiple angles and provides real-time feedback.

[1218] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1219] In this invention, the server includes means for capturing audio and video to allow a user to start presentation or customer service training, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for recognizing the user's emotions in real time and generating appropriate questions using AI based on the analyzed data, means for transmitting the generated questions to a terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or customer service, and means for generating feedback based on the evaluation results and displaying them to the user. This allows staff to improve their customer service skills in real time while receiving multifaceted feedback including facial expressions and emotional information at the customer service site.

[1220] An "audio capture means" is a device or software for capturing audio data.

[1221] "Video capture means" refers to a device or software for acquiring video data.

[1222] A "server" is a computer system or computer on a network that analyzes and stores data.

[1223] "Real-time conversion means" is a technology or device that converts speech into text almost instantly.

[1224] "Video analysis means" refers to a technique or device that analyzes video data and extracts features.

[1225] The "facial expression identification means" is a technique or device that identifies the facial expression of a user from video data.

[1226] The "gaze direction identification means" is a technique or device that identifies the direction of a user's gaze from video data.

[1227] The "gesture recognition means" is a technique or device that recognizes a user's gesture action from video data.

[1228] "Emotion recognition means" refers to a technique or device that recognizes the user's emotional state from video data and audio data.

[1229] A "question generator" is a technique or device that generates appropriate questions based on the analyzed data.

[1230] The "question display means" is a technique or device that visually displays the generated question to the user.

[1231] An "answer capture means" is a device or software for recapturing a user's answers.

[1232] The "evaluation means" is a technology or device that analyzes and evaluates the user's answers and the performance of customer service and presentation.

[1233] The "feedback generation means" is a technique or device that generates feedback to the user based on the evaluation results.

[1234] A "feedback display means" is a technique or device that visually displays the generated feedback to the user.

[1235] This invention relates to a system for improving customer service skills in brick-and-mortar stores, which has the function of capturing audio and video data and analyzing the user's facial expressions, gaze, gestures, and emotional state. Based on the analysis results, AI generates appropriate questions and provides feedback to the user.

[1236] The server includes an audio capture means, a video capture means, a real-time conversion means, a video analysis means, a facial expression identification means, a gaze identification means, a gesture identification means, an emotion recognition means, a question generation means, a question display means, an answer capture means, an evaluation means, and a feedback generation means.

[1237] The user launches a dedicated application on a smartphone or other device and presses the "Start Training" button. The device then captures audio and video data and sends it to the server in real time. The server then converts the audio into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text), analyzes the user's facial expressions, gaze, and gestures using a video analysis engine (e.g., OpenCV), and recognizes the user's emotional state using an emotion recognition engine (e.g., Affectiva).

[1238] Based on the analyzed data, the AI ​​on the server uses a question generation engine to generate appropriate questions. For example, a question such as "Please tell me specifically about risk management for this project" may be generated. The generated question is sent to the user's device and displayed on the screen.

[1239] The user answers the displayed questions. The device captures the answer again as audio and video data and sends it to the server. The server receives the answer data and analyzes it again using its voice recognition engine and video analysis engine.

[1240] The server evaluates the results of the analysis and generates feedback using a feedback generation engine, including specific suggestions for improvement such as, "Your speaking speed is appropriate, but you don't seem to be looking at the camera enough. You should be more aware of the camera."

[1241] The generated feedback is sent to the user's device and displayed in real time, allowing the user to improve their customer service skills in multiple ways.

[1242] Prompt Sentence Examples

[1243] "We have a new product in stock. Please explain in detail how it differs from similar products we have available so far, and what you recommend it for customers."

[1244] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1245] Step 1:

[1246] The user launches the dedicated application on a device such as a smartphone and presses the "Start Training" button. In this step, the device starts capturing audio and video data and sends this data to the server in real time. The input is the audio and video made by the user, and the output is the captured data sent to the server.

[1247] Step 2:

[1248] The server inputs the received voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into text in real time. The input is the captured voice data, and the output is the converted text data. The specific operation involves signal processing of the voice data and generation of text data.

[1249] Step 3:

[1250] The server inputs the received video data into a video analysis engine (e.g., OpenCV) to analyze facial expressions, gaze, and gestures. The input is the captured video data, and the output is the analyzed facial expression data, gaze data, and gesture data. The specific operation is to analyze the video frames and extract features.

[1251] Step 4:

[1252] The server uses an emotion recognition engine (e.g., Affectiva) to recognize the user's emotional state from video and audio data. The input is video and audio data, and the output is the recognized emotional state data. The specific operation is to classify and label the emotional state.

[1253] Step 5:

[1254] The AI ​​on the server generates appropriate questions based on the analyzed data and emotional state. The input is the analyzed text data, facial expression data, eye gaze data, gesture data, and emotional state data, and the output is the generated question. The specific operations are natural language generation and text generation.

[1255] Step 6:

[1256] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and the output is the question displayed on the terminal screen. The specific operations are data transmission and UI update.

[1257] Step 7:

[1258] The user answers the displayed questions. The device recaptures the answers as audio and video data and sends them to the server. The input is the user's audio and video answers, and the output is the captured data sent to the server.

[1259] Step 8:

[1260] The server then inputs the received voice data into a voice recognition engine and converts it into text. The input is the captured response voice, and the output is the converted text data. The specific operations are voice recognition and text conversion.

[1261] Step 9:

[1262] The server inputs the received video data into the video analysis engine and analyzes facial expressions, gaze, and gestures. The input is the captured response video, and the output is analyzed facial expression data, gaze data, and gesture data. The specific operations are video analysis and feature extraction.

[1263] Step 10:

[1264] The server evaluates the answer data and generates feedback using a feedback generation engine. The input is the analyzed answer data, and the output is the generated feedback. The specific operations are evaluation analysis and feedback generation.

[1265] Step 11:

[1266] The server sends the generated feedback to the device and displays it to the user in real time. The input is the generated feedback, and the output is the feedback displayed on the device screen. The specific operations are data transmission and UI update.

[1267] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1268] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1269] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1270] [Fourth embodiment]

[1271] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1272] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1273] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1274] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1275] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1276] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1277] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1278] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1279] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1280] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1281] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1282] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1283] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1284] This invention relates to an AI presentation training system that allows users to improve their presentation skills. This system can capture audio and video, and provide real-time analysis and feedback, enabling effective practice in a realistic environment.

[1285] System Configuration

[1286] 1. User Interface

[1287] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[1288] 2. Server Functions

[1289] The server receives the captured audio and video data and performs the following processes:

[1290] 1. Audio analysis:

[1291] The server uses a speech recognition engine to convert the user's speech into text in real time.

[1292] 2. Video analysis:

[1293] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[1294] 3. Question generation:

[1295] The AI ​​on the server generates appropriate questions based on the analyzed data. The questions are generated to match the content of the user's presentation, and the AI's settings take into account personality and expertise.

[1296] 4. Question display:

[1297] The question is sent from the server to the terminal and displayed on the user's screen. The user checks the question and begins answering.

[1298] 5. Answer analysis:

[1299] The user's responses are also captured as audio and video and sent to the server, where they are converted into text using a speech recognition engine and analyzed using a video analysis engine for facial expressions, eye movements, and gestures.

[1300] 6. Rating and Feedback:

[1301] The server comprehensively evaluates the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. Feedback is generated based on the evaluation results and sent to the terminal to be displayed to the user.

[1302] Natural language explanation of program processing

[1303] User starts presenting

[1304] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[1305] The server receives and analyzes the data

[1306] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[1307] AI-generated questions

[1308] The AI ​​on the server generates appropriate questions based on the analyzed data. For example, it might ask, "Please tell me specifically about risk management for this project." The questions are then sent to the device and displayed to the user.

[1309] User answers the question

[1310] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[1311] The server analyzes and evaluates the answers

[1312] The server converts the audio responses back into text and analyzes the video. The server evaluates the user's speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, it may evaluate the user's speaking speed as appropriate, but their voice volume as low.

[1313] Generating and displaying feedback

[1314] The server generates feedback based on the evaluation results, such as "Your speaking speed is appropriate, but your voice volume is low. You should speak a little louder." The feedback is sent to the device and displayed to the user.

[1315] Through this system, users can effectively improve their presentation skills while receiving continuous feedback. Furthermore, repeated practice allows for training in a more realistic environment.

[1316] The processing flow will be explained below.

[1317] Step 1:

[1318] The user starts a dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[1319] Step 2:

[1320] The server inputs the received voice data into a speech recognition engine and converts the voice into text, which is then used to analyze the content of the presentation.

[1321] Step 3:

[1322] The server inputs the received video data into a video analysis engine to analyze the user's facial expressions, gaze, and gestures, for example, to determine whether the user is smiling, looking at the camera, or moving their hands.

[1323] Step 4:

[1324] The AI ​​on the server generates appropriate questions based on the results of audio and video analysis. The questions are related to the content of the presentation, and the difficulty and perspective are adjusted according to the AI ​​settings.

[1325] Step 5:

[1326] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[1327] Step 6:

[1328] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[1329] Step 7:

[1330] The server inputs the voice data of the response into a voice recognition engine and converts it into text. It also analyzes the video data of the response using a video analysis engine. The server also analyzes facial expressions, eye movements, and gestures when responding.

[1331] Step 8:

[1332] The server evaluates the presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, and gestures. For example, the server may give an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not facing the camera."

[1333] Step 9:

[1334] The server generates feedback based on the evaluation results, including specific areas for improvement and advice, such as "If you look more closely at the camera, you will be more appealing to viewers."

[1335] Step 10:

[1336] The generated feedback is sent from the server to the device and displayed on the device screen. The user can review the feedback and use it to practice their next presentation to improve the points pointed out.

[1337] Through this specific processing step, users can effectively learn and improve their presentation skills.

[1338] Example 1

[1339] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1340] Conventional systems for improving presentation skills have the problem that it is difficult for users to receive immediate, specific feedback even when they practice, and because they do not perform real-time analysis, it is difficult to practice in a realistic environment.Furthermore, conventional systems are unable to comprehensively evaluate the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. during a presentation, and therefore can only provide limited feedback.

[1341] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1342] In this invention, the server includes means for converting voice to text, means for analyzing video to identify the user's facial expressions, eye movements, and gestures, means for generating questions using artificial intelligence based on the analyzed data, means for comprehensively evaluating data related to the user's presentation, means for providing sequential feedback based on the evaluation results, and means for enabling the user to practice in a manner close to reality to improve their presentation skills. This allows the user to receive specific feedback in real time and receive advice based on a comprehensive evaluation.

[1343] "Audio capturing means" refers to a microphone or other audio input device that captures the user's speech.

[1344] "Video capture means" refers to a camera or other video input device that captures the user's visual movements and expressions.

[1345] "Server" refers to a computer system for receiving and analyzing captured audio and video data.

[1346] "Means for converting voice to text" refers to technology that uses a voice recognition engine to convert voice data into text data.

[1347] "Means for analyzing video" refers to technology that processes and analyzes video data to identify a user's facial expressions, gaze, gestures, etc.

[1348] "Artificial intelligence" refers to a computer program or system that uses machine learning and natural language processing techniques to make inferences and judgments.

[1349] "Means for generating questions" refers to artificial intelligence technology for creating appropriate questions based on analyzed data.

[1350] "Means for comprehensively evaluating data related to a user's presentation" refers to technology for simultaneously analyzing and evaluating multiple elements, such as the user's speaking speed, voice volume, choice of words, eye contact, facial expressions, and gestures.

[1351] "Means for providing feedback" refers to technology that notifies users of specific improvements and advice in real time based on the analysis and evaluation results.

[1352] "Realistic practice tools" refers to systems that allow users to effectively improve their skills under conditions similar to those experienced in a real presentation.

[1353] This invention relates to a system for enabling users to improve their presentation skills. The system begins when a user launches a dedicated application on a device such as a PC or smartphone and starts a presentation.

[1354] Hardware and Software Configuration

[1355] 1. Hardware

[1356] Voice capture microphone: A microphone is used to capture the user's voice. Examples include a typical condenser microphone or a headset microphone.

[1357] Video capture camera: A camera is used to capture the user's video. Examples include a webcam built into a PC or an external camera.

[1358] 2. Software

[1359] Speech recognition engine: Engines such as Google Cloud Speech-to-Text and IBM Watson Speech to Text are used to convert speech into text.

[1360] Video analysis engine: OpenCV and Google Cloud Vision are used as the engine to analyze video data and identify the user's facial expressions, gaze, and gestures.

[1361] Generative AI models: GPT-3 and BERT are used as AI models to generate appropriate questions based on analyzed data.

[1362] System Operation

[1363] Start your presentation

[1364] The user starts the dedicated application and presses the "Start Presentation" button on their device, such as a PC or smartphone, which then uses the microphone and camera to capture audio and video and transmits the data to the server in real time.

[1365] Data analysis

[1366] The server converts the received voice data into text using a speech recognition engine (Google Cloud Speech-to-Text or IBM Watson Speech to Text), and simultaneously analyzes the video data using a video analysis engine (OpenCV or Google Cloud Vision). For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data, and the video analysis engine analyzes the user's facial expressions and eye direction as they speak.

[1367] question generation

[1368] Based on the analyzed data, a generative AI model (such as GPT-3 or BERT) on the server generates an appropriate question. For example, a question such as "Please tell me specifically about risk management for this project" is generated. The generated question is sent from the server to the device and displayed on the user's screen.

[1369] Answers to questions

[1370] The user checks the question displayed on the device screen and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The server again analyzes this data using its voice recognition engine and video analysis engine to evaluate the user's answer.

[1371] Generating and Providing Feedback

[1372] The server performs a comprehensive evaluation of the user's speaking speed, volume, choice of words, eye contact, facial expressions, gestures, etc. For example, it obtains a specific evaluation result such as "Your speaking speed is appropriate, but your voice volume is low." Based on this evaluation result, the server generates specific feedback and displays it on the user's screen. For example, it may provide feedback such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[1373] Specific examples

[1374] Prompt Sentence Examples

[1375] Example presentation: "The goal of this project is to increase sales by 20%."

[1376] Example question: "What strategies do you have in mind to achieve this goal?"

[1377] Through this system, users can improve their presentation skills while receiving specific feedback in real time, and by repeatedly practicing, they can improve the quality of their presentations.

[1378] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1379] Step 1:

[1380] The user starts the dedicated application and presses the "Start Presentation" button, which causes the device to capture the user's voice and video and send the data to the server.

[1381] Input: User's audio and video

[1382] Output: Audio and video data sent to the server

[1383] Specific operation: The device's microphone and camera capture audio and video in real time, encode the data, and send it to the server.

[1384] Step 2:

[1385] The server analyzes the received voice data using a speech recognition engine and converts it into text, for example, using Google Cloud Speech-to-Text.

[1386] Input: Audio data sent to the server

[1387] Output: Text data

[1388] Specific operation: The speech recognition engine analyzes the voice data and converts the user's speech into text. For example, it extracts the utterance "The goal of this project is..." as text data.

[1389] Step 3:

[1390] The server analyzes the video data using a video analysis engine, such as OpenCV.

[1391] Input: Video data sent to the server

[1392] Output: Analysis results of user's facial expressions, gaze, and gestures

[1393] Specific operation: The video analysis engine analyzes the video data frame by frame and identifies the user's facial expressions (e.g., smiling or surprised), gaze (e.g., looking toward the camera), and gestures (e.g., hand movements and gestures).

[1394] Step 4:

[1395] A generative AI model on the server generates appropriate questions based on the results of audio and video analysis, using, for example, GPT-3.

[1396] Input: Text data and video analysis results

[1397] Output: Generated question text

[1398] Specific behavior: The generative AI model generates relevant questions based on the analysis results, such as "Please tell me specifically about risk management for this project."

[1399] Step 5:

[1400] The server sends the generated question to the terminal and displays it on the user's screen.

[1401] Input: Generated question text

[1402] Output: The question displayed on the user's screen

[1403] Specific operation: The server sends the generated question text to the terminal, and the dedicated application displays the received question text on the user's screen.

[1404] Step 6:

[1405] The user checks the questions displayed on the terminal and begins to answer them. The terminal again captures audio and video and sends them to the server.

[1406] Input: User's answer audio and video

[1407] Output: Answer audio and video data sent to the server

[1408] Specific operation: When the user answers a question, the device's microphone and camera again capture audio and video and send the data to the server.

[1409] Step 7:

[1410] The server converts the response voice into text again using a voice recognition engine and analyzes the video again.

[1411] Input: Answer audio and video data sent to the server

[1412] Output: Answer text and video analysis results

[1413] Specific operation: The speech recognition engine converts the answer into text, and the video analysis engine reanalyzes the user's facial expressions, gaze, and gestures. For example, the speech "Risk management is..." is extracted as text, and the facial expressions and gaze are reanalyzed.

[1414] Step 8:

[1415] The server will provide an overall rating and generate feedback.

[1416] Input: Answer text and video analysis results

[1417] Output: Feedback text

[1418] Specific behavior: The server comprehensively evaluates the text data and video analysis results and generates specific feedback, such as "Your speaking speed is appropriate, but your voice volume is low. It would be better if you spoke a little louder."

[1419] Step 9:

[1420] The server sends the generated feedback to the terminal and displays it on the user's screen.

[1421] Input: Feedback text

[1422] Output: Feedback displayed on the user's screen

[1423] Specific operation: The server sends the generated feedback text to the terminal, and the dedicated application displays the received feedback on the user's screen.

[1424] (Application example 1)

[1425] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1426] In modern factories and production sites, workers are required to have the skills to effectively deliver various presentations, such as introducing new products and providing safety training. These skills contribute to worker growth and improved productivity, but traditional educational methods often struggle to provide real-time feedback, and one-way lectures often fail to provide effective instruction. The present invention aims to provide a system that allows workers to efficiently improve their presentation and education / training skills in the field.

[1427] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1428] In this invention, the server includes means for capturing audio and images to start a user's presentation or training, means for transmitting the captured audio and images to the server, means for converting the audio to text in real time in the server, means for analyzing the images in the server and identifying the user's facial expressions, line of sight, and gestures, means for an AI to generate appropriate prompts based on the analyzed data, means for transmitting the generated prompts to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or training, and means for generating feedback based on the evaluation results and displaying them to the user. This allows workers to improve their presentation and training skills while receiving feedback in real time.

[1429] "Users" are workers who use this system to give presentations or provide training.

[1430] "Voice capture" is the act of recording a user's speech as voice data.

[1431] "Video capture" is the act of recording a user's posture and movements as video data.

[1432] A "server" is a computer system that receives and analyzes audio and video data and provides feedback based on the results.

[1433] "Speech-to-text" is the process of converting captured voice data into text data.

[1434] "Image analysis" is the process of identifying a user's facial expressions, gaze, and gestures based on captured video data.

[1435] "Prompts" are appropriate questions or instructions generated by AI based on analyzed data.

[1436] "Feedback" refers to guidance and advice provided to users based on the results of evaluation of their presentations and training.

[1437] "Real-time" refers to the immediacy of time in which the results of a user action are provided as analysis and feedback immediately after the action is taken.

[1438] The present invention is an AI system that enables users to effectively deliver presentations or educational training, and includes functions for capturing audio and video, analyzing them in real time, and generating feedback.

[1439] System Configuration

[1440] 1. User Interface

[1441] The user starts a dedicated application on a device (smartphone, tablet, PC, etc.) and starts a presentation or training session. By pressing the start button, the device captures audio and video and transmits the data to the server in real time.

[1442] 2. Server Functions

[1443] Audio Analysis:

[1444] The server uses a speech recognition engine (e.g., Google Speech Recognition API) to convert the voice data into text. For example, if a user says, "The operating procedure for this machine is...", the server extracts this as text data.

[1445] Image analysis:

[1446] The server uses a video analysis engine (e.g., OpenCV) to analyze the user's facial expressions, gaze, and gestures, for example, to evaluate whether the user is smiling when speaking and in what direction their eyes are pointing.

[1447] Prompt generation:

[1448] Based on the analyzed data, the AI ​​generates appropriate prompts (questions or instructions), such as "Please explain the emergency shutdown procedure for this machine."

[1449] Question display:

[1450] The generated prompts are sent from the server to the terminal and displayed to the user, who then initiates an answer based on the prompts.

[1451] Response analysis and evaluation:

[1452] The user's answers are also captured as audio and video and sent to the server. The server converts the audio back into text and analyzes the video to evaluate the answer. For example, the server may evaluate the answer as "appropriate, but the user's gaze is not focused."

[1453] Feedback generation and display:

[1454] Based on the evaluation results, the server generates feedback, such as "It would be more effective if you directed your gaze more at the other person," and sends it to the device and displays it to the user.

[1455] Program processing explanation

[1456] The specific hardware and software used in this system are as follows:

[1457] Hardware:

[1458] Camera and microphone: Use the device's built-in or external camera and microphone.

[1459] Server: A dedicated server for high-performance data analysis.

[1460] software:

[1461] OpenCV: A library for video capture and analysis.

[1462] SpeechRecognition: An engine for converting speech to text.

[1463] Flask: A framework for server-side data processing.

[1464] Generative AI model: An AI algorithm for generating prompts (questions or instructions).

[1465] Examples of specific examples and prompts

[1466] Consider a case where a user is giving a presentation on how to operate a new machine in a factory. When the user presses the "start button," the camera and microphone are activated to capture audio and video, which are then sent to the server. The server analyzes this and generates a prompt such as "Please explain the emergency stop procedure for this machine," which is displayed on the terminal. When the user responds to the prompt and begins their explanation, their answer is captured again and sent to the server. The server evaluates the user's answer and provides feedback such as "It would be more effective if you paid more attention to the other person."

[1467] Specific prompt examples:

[1468] "Please explain the emergency shutdown procedure for this machine."

[1469] "Please elaborate on the content of the next slide and explain the risk factors that should be considered."

[1470] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1471] Step 1:

[1472] The user launches the dedicated application on their device and presses the "Start Presentation" button. This activates the device's camera and microphone, and audio and video capture begins. The input is the user's audio and video, and the output is the captured audio and video data. Specifically, the device's application starts the capture function of the specified camera and microphone.

[1473] Step 2:

[1474] The device transmits the captured audio and video data to the server in real time. Specifically, the device uploads the data to the server via a network. The data is often compressed and encrypted. The input is the captured audio and video data, and the output is the data transmitted to the server.

[1475] Step 3:

[1476] The server converts the received voice data into text using a voice recognition engine (e.g., Google Speech Recognition API). The input is voice data, and the output is text data. Specifically, the server calls the voice recognition engine, analyzes the voice data, and converts it into text.

[1477] Step 4:

[1478] The server analyzes the received video data using a video analysis engine (e.g., OpenCV) to identify the user's facial expressions, gaze, and gestures. The input is video data, and the output is analyzed feature data of the user's facial expressions, gaze, and gestures. Specifically, the video analysis engine processes the video frame by frame, identifies facial features, and extracts their features.

[1479] Step 5:

[1480] The server uses a generative AI model to generate appropriate prompts based on the analyzed voice text and video data. The input is text data and characteristic data on the user's facial expressions, eye movements, and gestures, and the output is the generated prompt. Specifically, the generative AI model analyzes this data and generates appropriate questions and instructions.

[1481] Step 6:

[1482] The server sends the generated prompt to the terminal, and the terminal displays the prompt on the user's screen. The input is the prompt sent from the server, and the output is the prompt displayed on the user's screen. Specifically, the server sends the generated prompt to the terminal via the network, and the terminal receives it and displays it on the screen.

[1483] Step 7:

[1484] The user responds based on the prompts displayed on the device screen, and the device again captures audio and video and sends them to the server. The input is the user's new audio and video, and the output is the captured data sent to the server. Specifically, the device again captures audio and video using the camera and microphone and sends them to the server.

[1485] Step 8:

[1486] The server analyzes the received audio and video again, converts the user's audio response into text, and analyzes the video to evaluate it. The input is the newly received audio and video data, and the output is the analyzed text data and the evaluation results. Specifically, the server uses a voice recognition engine to convert the audio into text, and a video analysis engine to analyze the user's facial expressions, gaze, and gestures.

[1487] Step 9:

[1488] The server generates feedback based on the evaluation results and sends it to the terminal. The terminal displays the feedback to the user. The input is the evaluation results, and the output is the generated feedback sentence and its display to the user by the terminal that received it. Specifically, the server generates feedback and sends it to the terminal via the network, and the terminal displays the feedback on its screen.

[1489] Example prompt sentence:

[1490] "Please explain the emergency shutdown procedure for this machine."

[1491] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1492] The present invention relates to an AI presentation trainer system that helps users improve their presentation skills. The system combines audio and video capture with an emotion engine that recognizes the user's emotions. This provides more detailed feedback based on the user's emotions, which can be expected to improve presentation skills.

[1493] System Configuration

[1494] 1. User Interface

[1495] The user starts a presentation by launching a dedicated application on a device such as a PC or smartphone. By pressing the "Start Presentation" button, the device captures audio and video and transmits them to the server in real time.

[1496] 2. Server Functions

[1497] The server receives the captured audio and video data and performs the following processes:

[1498] 1. Audio analysis:

[1499] The server uses a speech recognition engine to convert the user's speech into text in real time.

[1500] 2. Video analysis:

[1501] The server uses a video analysis engine to analyze facial expressions, gaze, gestures, etc.

[1502] 3. Emotion recognition:

[1503] The server uses an emotion engine to recognize the user's emotions, for example, determining whether the user is nervous or relaxed during a presentation.

[1504] 4. Question generation:

[1505] The AI ​​on the server generates appropriate questions based on the analyzed data and emotion recognition results, and adjusts the content and difficulty of the questions depending on the emotional state.

[1506] 5. Question display:

[1507] The question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[1508] 6. Answer analysis:

[1509] The user's answers are also captured as audio and video and sent to the server. The server then converts the answers into text using a speech recognition engine, analyzes facial expressions, eye movements, and gestures using a video analysis engine, and analyzes the emotions expressed when answering using an emotion engine.

[1510] 7. Rating and Feedback:

[1511] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may evaluate the user's speaking speed as appropriate, but the user's eyes are not directed toward the camera and the user's emotional state is tense.

[1512] 8. Feedback Generation and Display:

[1513] The server generates feedback based on the evaluation results. The feedback includes specific areas for improvement and advice. For example, the server might generate advice such as, "You should look more closely at the camera. Also, take a deep breath and relax to reduce tension." The feedback is sent from the server to the device and displayed to the user.

[1514] Natural language explanation of program processing

[1515] User starts presenting

[1516] The user launches the dedicated app and presses the "Start Presentation" button. The device captures audio and video data and sends it to the server.

[1517] The server receives and analyzes the data

[1518] The server converts the audio stream into text using a speech recognition engine, and analyzes the video stream using a video analysis engine. For example, if a user says, "The goal of this project is...", the speech recognition engine extracts this as text data. The video analysis engine analyzes the user's facial expression as they speak, the direction of their eyes, and whether their gestures are appropriate.

[1519] emotion recognition

[1520] The server's emotion engine analyzes the user's emotional state from the video data, determining whether they are tense, relaxed, enjoying themselves, etc.

[1521] AI-generated questions

[1522] The AI ​​on the server generates appropriate questions based on the results of voice and video analysis and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The question is then sent to the device and displayed to the user.

[1523] User answers the question

[1524] The user confirms the question and begins to answer, and the device recaptures the audio and video of the answer and sends it to the server.

[1525] The server analyzes and evaluates the answers

[1526] The server converts the voice response back into text and analyzes the video. It also uses an emotion engine to analyze the user's emotional state. For example, it may evaluate the user's speaking speed as appropriate, but their eyes are not looking at the camera and their emotional state seems tense.

[1527] Generating and displaying feedback

[1528] The server generates feedback based on the evaluation results. For example, it might say, "Your appeal to the audience will increase if you look more closely at the camera. Also, try taking deep breaths to relax." The feedback is sent to the device and displayed to the user.

[1529] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[1530] The processing flow will be explained below.

[1531] Step 1:

[1532] The user starts the dedicated application on the device and presses the "Start Presentation" button. The device captures the user's voice and video and transmits this data to the server in real time.

[1533] Step 2:

[1534] The server inputs the received voice data into a speech recognition engine to convert the speech into text, which is used to identify what the user is saying.

[1535] Step 3:

[1536] The server inputs the received video data into a video analysis engine, which analyzes the user's facial expressions, gaze direction, and gestures. The video analysis engine identifies the user's facial expressions, gaze direction, hand and body movements, etc.

[1537] Step 4:

[1538] The server uses an emotion engine to recognize the user's emotions from the video data, for example, determining whether the user is nervous, relaxed, or having fun.

[1539] Step 5:

[1540] The AI ​​on the server generates appropriate questions based on the results of voice, video, and emotion recognition. For example, a question might be generated such as, "Please tell me specifically about risk management for this project." The content and difficulty of the questions are adjusted according to the user's emotional state.

[1541] Step 6:

[1542] The generated question is sent from the server to the terminal and displayed on the terminal screen. The user checks the displayed question and prepares an answer.

[1543] Step 7:

[1544] The user answers the questions by voice, and the device captures the user's answers again as audio and video and sends this data to the server.

[1545] Step 8:

[1546] The server inputs the voice data of the response into a speech recognition engine and converts it into text. It also analyzes the video data of the response with a video analysis engine to identify the user's facial expressions, eye movements, and gestures. The server then uses an emotion engine to analyze the user's emotional state.

[1547] Step 9:

[1548] The server evaluates the user's presentation based on the content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, the server may provide an evaluation result such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera. Also, your emotional state seems tense."

[1549] Step 10:

[1550] The server generates feedback based on the evaluation results. The feedback includes specific improvements and advice. For example, advice such as "Turn your eyes more closely into the camera. Also, try taking deep breaths to relax" is generated. The feedback is sent from the server to the device and displayed to the user.

[1551] Through this specific processing step, users can effectively learn and improve their presentation skills. Multifaceted feedback, including emotional state, is provided, allowing users to gain a deeper understanding of how their presentations are perceived.

[1552] Example 2

[1553] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1554] Conventional presentation training systems are limited to capturing and analyzing audio and video data, and are unable to provide feedback that takes into account the user's emotional state. This limits the extent to which users can improve their presentation skills. Specifically, the user's emotional state, such as tension or relaxation, is not reflected in the evaluation or feedback of the presentation, making it difficult to provide training that closely resembles a real presentation situation.

[1555] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1556] In this invention, the server includes means for capturing audio and video to start a presentation by a user, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for an AI model to generate appropriate questions based on the analyzed data, means for transmitting the generated questions to the terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation, means for generating feedback based on the evaluation results and displaying them to the user, means for recognizing and analyzing the user's emotions, and means for adjusting the content of the feedback based on the user's emotional state, thereby enabling multifaceted feedback that takes the user's emotional state into consideration.

[1557] "User" refers to an individual or group that makes a presentation using a dedicated application.

[1558] "Terminal" refers to a device such as a PC, smartphone, or tablet that a user uses when giving a presentation.

[1559] "Means for capturing audio and video" refers to the ability to record audio and video data of a user using the microphone and camera installed on the device.

[1560] "Means for transmitting audio and video to a server" refers to a technology for transmitting captured audio and video data to a server via a network in real time.

[1561] "Means for converting speech to text" refers to a function that utilizes a speech recognition engine to convert captured speech data into corresponding text data.

[1562] "Means for analyzing video and identifying a user's facial expressions, gaze direction, and gestures" refers to a technology that uses a video analysis engine to detect and analyze a user's facial expressions, gaze direction, and gestures.

[1563] "Means for an AI model to generate appropriate questions" refers to the function of automatically generating appropriate questions for a user using a generative AI model based on the results of audio and video analysis.

[1564] "Means for transmitting the generated question to the terminal and displaying it to the user" refers to a function for transmitting a question generated by the server to the terminal and displaying it on the screen of the terminal.

[1565] "Means for recapturing the user's answers and sending them to the server" refers to a function for recapturing audio and video data when the user gives an answer and sending that data to the server.

[1566] "Means for analyzing the user's answers and evaluating the presentation" refers to a technology for analyzing audio and video data and evaluating the content of the user's presentation.

[1567] "Means for generating feedback based on the evaluation results and displaying it to the user" refers to a function for generating useful feedback based on the evaluation results of the presentation, transmitting it to the terminal, and displaying it to the user.

[1568] "Means for recognizing and analyzing user emotions" refers to technology that recognizes and analyzes a user's emotional state from video data.

[1569] The "means for adjusting the feedback content based on the user's emotional state" refers to a function for appropriately adjusting and providing the feedback content based on the emotion recognition result.

[1570] The present invention is a system for improving users' presentation skills, and in particular, by combining emotion recognition technology, it is possible to provide feedback based on the user's emotional state, allowing users to acquire more effective presentation skills.

[1571] First, the user launches a dedicated application on a device such as a PC or smartphone. When the user presses the "Start Presentation" button, the device's built-in microphone and camera start up and begin capturing audio and video data. The captured audio and video data is then sent to a server in real time via the Internet.

[1572] The server converts the received voice data into text data in real time using the Google Cloud Speech-to-Text API. At the same time, the received video data is analyzed using OpenCV. The video analysis engine detects and analyzes the user's facial expressions, eye direction, and gestures. This analysis allows the system to understand the context in which the user is speaking.

[1573] Furthermore, the server uses the Microsoft Azure Emotion API to analyze the user's emotional state from the video data. Based on facial expressions, eye movements, and changes in facial muscles, it can determine whether the user is tense, relaxed, or enjoying themselves. For example, the level of tension and stress can also be calculated.

[1574] The server uses OpenAI GPT-4 to generate appropriate questions based on these analysis results. The generated questions are sent from the server to the device and displayed on the user's screen. For example, a question such as "Please tell me specifically about risk management for this project" may be displayed.

[1575] The user begins to answer the displayed question. The device again uses the microphone and camera to capture audio and video data of the user's answer. The captured data is sent to the server, which again performs speech recognition and video analysis. The answer is converted to text using the Google Cloud Speech-to-Text API, and the video analysis engine analyzes facial expressions, eye movements, and gestures. The emotion engine is also used to analyze the user's emotional state.

[1576] The server comprehensively evaluates this data and makes a rating based on the presentation content, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. For example, a specific rating may be given such as, "Your speaking speed is appropriate, but your eyes are not directed toward the camera and your emotional state is tense."

[1577] Based on the evaluation results, the server generates feedback, which includes specific areas for improvement and advice. For example, advice such as "You can appeal more to the audience by looking more closely at the camera. Also, it would be a good idea to take deep breaths to relieve tension" is provided. The generated feedback is sent from the server to the device and displayed to the user.

[1578] Through this system, users can effectively improve their presentation skills while receiving multifaceted feedback, including their emotional state. Furthermore, repeated practice allows for training in a more realistic environment.

[1579] Specific examples

[1580] Prompt Sentence Examples

[1581] "Please tell me more about the risk management for this project."

[1582] Example of user analysis results

[1583] Speech text: "The first step in managing the risks of this project is..."

[1584] Facial expression analysis result: nervous

[1585] Eye analysis results: Not facing the camera

[1586] Emotion recognition result: tension

[1587] Feedback example

[1588] "Your speech is clear, but it would be better if you looked directly into the camera. To relax, try taking a deep breath before you start speaking."

[1589] In this way, the system can improve users' presentation skills in a variety of ways.

[1590] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1591] Step 1:

[1592] User starts presenting

[1593] The user launches the dedicated app on a device such as a PC or smartphone and presses the "Start Presentation" button. The input is the user's operation (pressing the button), which triggers the device's microphone and camera to start up. The output is the device starting to capture audio and video data.

[1594] Step 2:

[1595] Sending captured data

[1596] The terminal transmits the captured audio and video data to the server in real time. The input is the captured audio and video data, which are transmitted to the server via the network. The output is the real-time audio and video streams that arrive at the server.

[1597] Step 3:

[1598] Converting audio data to text

[1599] The server converts the received audio stream into text data using the Google Cloud Speech-to-Text API. The input is the received audio data, which is then subjected to speech recognition to generate text data. The output is text data that shows what the user said.

[1600] Step 4:

[1601] Video data analysis

[1602] The server analyzes the received video stream using OpenCV. The input is the received video data, and video analysis is performed on this data to identify the user's facial expressions, gaze, and gestures. The output is the analysis result data on the user's facial expressions, gaze, and gestures.

[1603] Step 5:

[1604] emotion recognition

[1605] The server uses the Microsoft Azure Emotion API to analyze the user's emotional state from video data. The input is the video analysis results data, and emotion recognition is performed based on this data. The output is data on the user's emotional state, such as whether they are tense or relaxed.

[1606] Step 6:

[1607] Question Generation

[1608] The server uses OpenAI GPT-4 to generate appropriate questions based on the results of voice and video analysis and emotion recognition. The input is voice text data, analysis result data, and emotion recognition data, and the question is generated based on these. The output is the generated question.

[1609] Step 7:

[1610] Show Questions

[1611] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and this data is sent to the terminal via the network. The output is the question displayed on the user's application screen. Example: "Please tell me specifically about risk management for this project."

[1612] Step 8:

[1613] User response capture

[1614] The user checks the displayed question and begins to answer. The device again captures the audio and video of the answer and sends it to the server. The input is the audio and video data of the user's answer, which is captured and sent to the server. The output is the audio and video data that reaches the server.

[1615] Step 9:

[1616] Analysis of responses

[1617] The server converts the retransmitted voice data into text using a voice recognition engine, and analyzes the video data using a video analysis engine. The input is the user's response voice and video data, which are then subjected to voice recognition and video analysis. The output is the text of the response and the analysis results of the user's facial expressions, gaze, and gestures.

[1618] Step 10:

[1619] Emotion recognition when answering

[1620] The server again uses the Microsoft Azure Emotion API to analyze the user's emotional state at the time of answering. The input is the video analysis result data, and emotion recognition is performed based on this. The output is data indicating the user's emotional state.

[1621] Step 11:

[1622] Evaluation of presentation content

[1623] The server evaluates the presentation based on the content of the answers, speaking speed, volume, choice of words, eye contact, facial expressions, gestures, and emotional state. The input is audio-text data, video analysis data, and emotion recognition data, and a comprehensive evaluation is made based on these. The output is the evaluation result of the user's presentation.

[1624] Step 12:

[1625] Generate feedback

[1626] Based on the evaluation results, the server generates feedback including specific areas for improvement and advice. The input is the evaluation result data, and feedback statements are generated based on this. The output is the generated feedback statements. Example: "Your appeal to the audience will increase if you look more closely at the camera. Try taking deep breaths to relax."

[1627] Step 13:

[1628] View Feedback

[1629] The server sends the generated feedback to the terminal and displays it on the user's screen. The input is the generated feedback sentence, and this data is sent to the terminal via the network. The output is the feedback sentence displayed on the user's application screen.

[1630] The above is the flow of processing for this system program. We have explained the specific operations performed at each step, as well as the processing or calculation of input and output data.

[1631] (Application example 2)

[1632] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1633] To improve the customer service skills of store staff, practice in a realistic environment is necessary. However, conventional methods provide limited feedback, and areas for improvement are not clearly identified. Furthermore, training does not take into account the emotional state of staff, and emotional upset during customer service can affect customer satisfaction. To address these issues, a system is needed that analyzes staff's customer service skills from multiple angles and provides real-time feedback.

[1634] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1635] In this invention, the server includes means for capturing audio and video to allow a user to start presentation or customer service training, means for transmitting the captured audio and video to the server, means for converting the audio to text in real time in the server, means for analyzing the video in the server and identifying the user's facial expressions, eye movements, and gestures, means for recognizing the user's emotions in real time and generating appropriate questions using AI based on the analyzed data, means for transmitting the generated questions to a terminal and displaying them to the user, means for re-capturing the user's answers and transmitting them to the server, means for analyzing the user's answers and evaluating the presentation or customer service, and means for generating feedback based on the evaluation results and displaying them to the user. This allows staff to improve their customer service skills in real time while receiving multifaceted feedback including facial expressions and emotional information at the customer service site.

[1636] An "audio capture means" is a device or software for capturing audio data.

[1637] "Video capture means" refers to a device or software for acquiring video data.

[1638] A "server" is a computer system or computer on a network that analyzes and stores data.

[1639] "Real-time conversion means" is a technology or device that converts speech into text almost instantly.

[1640] "Video analysis means" refers to a technique or device that analyzes video data and extracts features.

[1641] The "facial expression identification means" is a technique or device that identifies the facial expression of a user from video data.

[1642] The "gaze direction identification means" is a technique or device that identifies the direction of a user's gaze from video data.

[1643] The "gesture recognition means" is a technique or device that recognizes a user's gesture action from video data.

[1644] "Emotion recognition means" refers to a technique or device that recognizes the user's emotional state from video data and audio data.

[1645] A "question generator" is a technique or device that generates appropriate questions based on the analyzed data.

[1646] The "question display means" is a technique or device that visually displays the generated question to the user.

[1647] An "answer capture means" is a device or software for recapturing a user's answers.

[1648] The "evaluation means" is a technology or device that analyzes and evaluates the user's answers and the performance of customer service and presentation.

[1649] The "feedback generation means" is a technique or device that generates feedback to the user based on the evaluation results.

[1650] A "feedback display means" is a technique or device that visually displays the generated feedback to the user.

[1651] This invention relates to a system for improving customer service skills in brick-and-mortar stores, which has the function of capturing audio and video data and analyzing the user's facial expressions, gaze, gestures, and emotional state. Based on the analysis results, AI generates appropriate questions and provides feedback to the user.

[1652] The server includes an audio capture means, a video capture means, a real-time conversion means, a video analysis means, a facial expression identification means, a gaze identification means, a gesture identification means, an emotion recognition means, a question generation means, a question display means, an answer capture means, an evaluation means, and a feedback generation means.

[1653] The user launches a dedicated application on a smartphone or other device and presses the "Start Training" button. The device then captures audio and video data and sends it to the server in real time. The server then converts the audio into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text), analyzes the user's facial expressions, gaze, and gestures using a video analysis engine (e.g., OpenCV), and recognizes the user's emotional state using an emotion recognition engine (e.g., Affectiva).

[1654] Based on the analyzed data, the AI ​​on the server uses a question generation engine to generate appropriate questions. For example, a question such as "Please tell me specifically about risk management for this project" may be generated. The generated question is sent to the user's device and displayed on the screen.

[1655] The user answers the displayed questions. The device captures the answer again as audio and video data and sends it to the server. The server receives the answer data and analyzes it again using its voice recognition engine and video analysis engine.

[1656] The server evaluates the results of the analysis and generates feedback using a feedback generation engine, including specific suggestions for improvement such as, "Your speaking speed is appropriate, but you don't seem to be looking at the camera enough. You should be more aware of the camera."

[1657] The generated feedback is sent to the user's device and displayed in real time, allowing the user to improve their customer service skills in multiple ways.

[1658] Prompt Sentence Examples

[1659] "We have a new product in stock. Please explain in detail how it differs from similar products we have available so far, and what you recommend it for customers."

[1660] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1661] Step 1:

[1662] The user launches the dedicated application on a device such as a smartphone and presses the "Start Training" button. In this step, the device starts capturing audio and video data and sends this data to the server in real time. The input is the audio and video made by the user, and the output is the captured data sent to the server.

[1663] Step 2:

[1664] The server inputs the received voice data into a speech recognition engine (e.g., Google Cloud Speech-to-Text) and converts it into text in real time. The input is the captured voice data, and the output is the converted text data. The specific operation involves signal processing of the voice data and generation of text data.

[1665] Step 3:

[1666] The server inputs the received video data into a video analysis engine (e.g., OpenCV) to analyze facial expressions, gaze, and gestures. The input is the captured video data, and the output is the analyzed facial expression data, gaze data, and gesture data. The specific operation is to analyze the video frames and extract features.

[1667] Step 4:

[1668] The server uses an emotion recognition engine (e.g., Affectiva) to recognize the user's emotional state from video and audio data. The input is video and audio data, and the output is the recognized emotional state data. The specific operation is to classify and label the emotional state.

[1669] Step 5:

[1670] The AI ​​on the server generates appropriate questions based on the analyzed data and emotional state. The input is the analyzed text data, facial expression data, eye gaze data, gesture data, and emotional state data, and the output is the generated question. The specific operations are natural language generation and text generation.

[1671] Step 6:

[1672] The server sends the generated question to the terminal and displays it on the user's screen. The input is the generated question, and the output is the question displayed on the terminal screen. The specific operations are data transmission and UI update.

[1673] Step 7:

[1674] The user answers the displayed questions. The device recaptures the answers as audio and video data and sends them to the server. The input is the user's audio and video answers, and the output is the captured data sent to the server.

[1675] Step 8:

[1676] The server then inputs the received voice data into a voice recognition engine and converts it into text. The input is the captured response voice, and the output is the converted text data. The specific operations are voice recognition and text conversion.

[1677] Step 9:

[1678] The server inputs the received video data into the video analysis engine and analyzes facial expressions, gaze, and gestures. The input is the captured response video, and the output is analyzed facial expression data, gaze data, and gesture data. The specific operations are video analysis and feature extraction.

[1679] Step 10:

[1680] The server evaluates the answer data and generates feedback using a feedback generation engine. The input is the analyzed answer data, and the output is the generated feedback. The specific operations are evaluation analysis and feedback generation.

[1681] Step 11:

[1682] The server sends the generated feedback to the device and displays it to the user in real time. The input is the generated feedback, and the output is the feedback displayed on the device screen. The specific operations are data transmission and UI update.

[1683] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1684] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1685] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1686] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1687] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1688] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1689] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1690] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1691] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1692] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1693] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1694] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1695] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1696] 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.

[1697] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1698] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1699] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1700] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1701] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1702] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1703] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1704] The following is further disclosed regarding the above embodiment.

[1705] (Claim 1)

[1706] means for capturing audio and video for a user to initiate a presentation;

[1707] means for transmitting the captured audio and video to a server;

[1708] a means for converting speech to text in real time on a server;

[1709] A means for analyzing the video in the server and identifying the user's facial expression, gaze, and gesture;

[1710] A means for AI to generate appropriate questions based on the analyzed data; and

[1711] means for transmitting the generated question to a terminal and displaying it to a user;

[1712] a means for recapturing the user's answers and transmitting them to the server;

[1713] means for analyzing the user's responses and evaluating the presentation;

[1714] The system includes a means for generating and displaying feedback to the user based on the evaluation results.

[1715] (Claim 2)

[1716] 10. The system of claim 1, wherein the system evaluates the user's speaking speed, volume, language, eye contact, facial expressions, and gestures.

[1717] (Claim 3)

[1718] 10. The system of claim 1, wherein feedback is provided to the user in real time based on the evaluation results.

[1719] "Example 1"

[1720] (Claim 1)

[1721] means for capturing audio and video for a user to initiate a presentation;

[1722] means for transmitting the captured audio and video to a server;

[1723] a means for converting speech to text in real time on a server;

[1724] A means for analyzing the video in the server and identifying the user's facial expression, gaze, and gesture;

[1725] A means for the artificial intelligence to generate appropriate questions based on the analyzed data; and

[1726] means for transmitting the generated question to a terminal and displaying it to a user;

[1727] a means for recapturing the user's answers and transmitting them to the server;

[1728] means for analyzing the user's responses and evaluating the presentation;

[1729] means for generating and displaying feedback to the user based on the evaluation results;

[1730] a means for comprehensively evaluating data relating to a user's presentation;

[1731] a means for providing sequential feedback based on the evaluation results;

[1732] A realistic way to practice and improve users' presentation skills

[1733] A system including:

[1734] (Claim 2)

[1735] 10. The system of claim 1, wherein the system evaluates the user's speaking speed, volume, language, eye contact, facial expressions, and gestures.

[1736] (Claim 3)

[1737] 10. The system of claim 1, wherein the system provides sequential feedback to the user based on the evaluation results.

[1738] "Application Example 1"

[1739] (Claim 1)

[1740] a means for capturing audio and images for a user to begin a presentation or teaching;

[1741] means for transmitting the captured audio and images to a server;

[1742] a means for converting speech to text in real time on a server;

[1743] means for analyzing the image at the server and identifying the user's facial expression, gaze, and gestures;

[1744] A means for AI to generate appropriate prompts based on the analyzed data; and

[1745] means for sending the generated prompt to a terminal for display to a user;

[1746] a means for recapturing the user's answers and transmitting them to the server;

[1747] A means for analyzing the user's responses and evaluating the presentation and the education;

[1748] The system includes a means for generating and displaying feedback to the user based on the evaluation results.

[1749] (Claim 2)

[1750] 10. The system of claim 1, wherein the system evaluates the user's speaking rate, volume, language, eye contact, facial expressions, and gestures.

[1751] (Claim 3)

[1752] 10. The system of claim 1, wherein the system provides immediate feedback to the user based on the evaluation results.

[1753] "Example 2: Combining Emotion Engines"

[1754] (Claim 1)

[1755] means for capturing audio and video for a user to initiate a presentation;

[1756] means for transmitting the captured audio and video to a server;

[1757] a means for converting speech to text in real time on a server;

[1758] A means for analyzing the video in the server and identifying the user's facial expression, gaze, and gesture;

[1759] A means for the AI ​​model to generate appropriate questions based on the analyzed data; and

[1760] means for transmitting the generated question to a terminal and displaying it to a user;

[1761] a means for recapturing the user's answers and transmitting them to the server;

[1762] means for analyzing the user's responses and evaluating the presentation;

[1763] means for generating and displaying feedback to the user based on the evaluation results;

[1764] means for recognizing and analyzing user emotions;

[1765] The system includes means for adjusting feedback content based on the user's emotional state.

[1766] (Claim 2)

[1767] 10. The system of claim 1, wherein the system evaluates the user's speaking speed, volume, language, eye contact, facial expressions, and gestures.

[1768] (Claim 3)

[1769] 10. The system of claim 1, wherein feedback is provided to the user in real time based on the evaluation results.

[1770] "Application example 2 when combining emotion engines"

[1771] (Claim 1)

[1772] means for capturing audio and video for a user to initiate a presentation or customer service training;

[1773] means for transmitting the captured audio and video to a server;

[1774] a means for converting speech to text in real time on a server;

[1775] A means for analyzing the video in the server and identifying the user's facial expression, gaze, and gesture;

[1776] A means for AI to recognize user emotions in real time and generate appropriate questions based on the analyzed data,

[1777] means for transmitting the generated question to a terminal and displaying it to a user;

[1778] a means for recapturing the user's answers and transmitting them to the server;

[1779] A means for analyzing the user's answers and evaluating the presentation or custome...

Claims

1. means for capturing audio and video for a user to initiate a presentation; means for transmitting the captured audio and video to a server; a means for converting speech to text in real time on a server; A means for analyzing the video in the server and identifying the user's facial expression, gaze, and gesture; A means for AI to generate appropriate questions based on the analyzed data; and means for transmitting the generated question to a terminal and displaying it to a user; a means for recapturing the user's answers and transmitting them to the server; means for analyzing the user's responses and evaluating the presentation; The system includes a means for generating and displaying feedback to the user based on the evaluation results.

2. The system of claim 1 , wherein the system evaluates the user's speaking rate, volume, language, eye contact, facial expressions, and gestures.

3. The system of claim 1 , wherein feedback is provided to the user in real time based on the evaluation results.

Images