system

Abstract

Provide a system. A method for receiving a video file from a user includes: means for extracting an audio track from the received video file; a means for converting the audio track into text data; a generative artificial intelligence means for analyzing text data and generating feedback based on preset criteria; means for notifying the user of the generated feedback; A system including:

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] Conventionally, feedback on presentations has often relied on the opinions of others, making it difficult to obtain objective and consistent feedback. This problem is particularly pronounced when face-to-face feedback is difficult or when time or space for feedback is limited. Furthermore, when many people need feedback at the same time, the quality of the feedback can vary. The present invention aims to solve these problems by providing a system that automatically and objectively analyzes the strengths and weaknesses of a user's presentation and provides specific feedback. [Means for solving the problem]

[0005] The present invention provides a system that receives a video file from a user, extracts an audio track from the received video file, and converts the audio track into text data. The system further includes a generative artificial intelligence (AI) that analyzes the converted text data and generates feedback based on preset criteria. The AI ​​analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content, and generates specific feedback on strengths and weaknesses. This feedback is communicated to the user in an easy-to-understand format, allowing the user to receive objective and specific feedback. Furthermore, the quality of the feedback is maintained consistently even when multiple users use the system simultaneously.

[0006] "User" refers to any individual or legal entity that uses the System to upload presentation videos and receive feedback.

[0007] "Video File" means a digital file containing the audio and visual content of a Presentation uploaded by a User.

[0008] "Means for receiving" refers to the function for electronically receiving video files from users and storing them within the system.

[0009] "Means for extracting audio tracks" refers to a function for separating and obtaining audio data from a video file.

[0010] "Means for converting into text data" refers to a function for analyzing the extracted audio track and converting its content into text format.

[0011] "Generative AI means" refers to a function that includes artificial intelligence techniques for analyzing text data and generating feedback based on pre-set criteria.

[0012] "Means of analyzing feedback" refers to the algorithms and processing methods that the generative AI uses to evaluate the content of the presentation and identify strengths and areas for improvement.

[0013] "Means for notifying" refers to a function for communicating the generated feedback to the user in the form of email, an in-system message, or the like.

[0014] "Video Processing Library" refers to a collection of software and tools for processing and extracting specific data (e.g., audio tracks) from video files. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI). This system consists of a server, a terminal, and a user.

[0037] Uploading and saving videos

[0038] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[0039] Video Analysis

[0040] The server extracts the audio track from the stored video file, and using a video processing library (e.g. FFmpeg) separates the audio from the video file and saves it as a temporary audio file for further processing.

[0041] Speech to text

[0042] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, it can analyze the audio data using a specific API (e.g., a speech recognition API) and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[0043] Feedback generation using generative AI

[0044] The server sends the acquired text data to a generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and then generates specific feedback including good points and areas for improvement.

[0045] Notification of results

[0046] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement.

[0047] Specific examples

[0048] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of the project." This text is sent to a generative AI, which analyzes it and generates feedback such as "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly." This feedback is then sent to the user via email, allowing them to learn specific improvements.

[0049] In this way, the system provides users with concrete feedback to improve the quality of their presentations. By receiving objective and consistent evaluations, users are provided with concrete guidelines to help improve their presentations.

[0050] The processing flow will be explained below.

[0051] Step 1:

[0052] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[0053] Step 2:

[0054] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[0055] Step 3:

[0056] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[0057] Step 4:

[0058] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[0059] Step 5:

[0060] The server sends the text data to the generative AI, which then creates and sends a request to the generative AI (e.g., GPT-4 (registered trademark)) to analyze the presentation text.

[0061] Step 6:

[0062] Generative AI analyzes text data and generates feedback. Based on the user's presentation text, the generative AI evaluates the presentation's structure, content consistency, tone of voice, speed, use of visual content, and other aspects, and generates feedback including good points and areas for improvement.

[0063] Step 7:

[0064] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[0065] Example 1

[0066] 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."

[0067] In order to obtain evaluations and feedback on presentation videos, third-party evaluations or self-feedback are often required. However, these methods are subjective and have problems with maintaining consistency in evaluation criteria. Furthermore, it is difficult to efficiently identify specific areas for improvement, and there is a lack of guidelines for quickly improving the quality of presentations.

[0068] 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.

[0069] In this invention, the server includes means for receiving a video file from a user, means for saving the received video file in a predetermined directory, means for extracting an audio track from the received video file, means for saving the audio track as a temporary audio file, means for transmitting the temporary audio file to a speech recognition service and converting the speech into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, and means for formatting the generated feedback and notifying the user. This allows users to easily obtain objective and consistent feedback, quickly identify specific areas for improvement in their presentations, and improve the quality of their presentations.

[0070] A "user" is a person or organization that accesses the system to upload presentation videos and receive feedback.

[0071] "Video file" refers to a digital file containing presentation content uploaded by a user, with an example file extension of .mp4.

[0072] The "server" is the central part of the system, and is a collection of hardware and software that performs various processes such as receiving, storing, analyzing, generating feedback, and notifying users of video files.

[0073] A "directory" refers to a folder on a file system that is used to organize and store data within a server.

[0074] "Audio Track" refers to the audio data extracted from a video file that contains the audio content of the presentation.

[0075] "Temporary audio file" refers to an audio data file that is generated to temporarily store an audio track and is used for subsequent speech recognition processing.

[0076] "Speech Recognition Service" refers to a cloud-based or local service or API for analyzing and converting voice data into text.

[0077] "Text data" refers to the text information generated by the speech recognition service, and represents the content of the presentation in text form.

[0078] "Generative AI means" refers to an AI model that has an algorithm that generates and analyzes information based on input data, and is responsible for generating feedback for presentations.

[0079] "Feedback" refers to information provided by generative artificial intelligence means that includes an evaluation of the presentation and specific areas for improvement.

[0080] "Notification means" refers to a method or system for communicating generated feedback to the user, including email, dashboard display, etc.

[0081] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides feedback. This system is composed of a server, a terminal, and a user. Specific embodiments are described in detail below.

[0082] Uploading and saving videos

[0083] Users upload presentation videos to the system from their devices using a browser or a dedicated application. Video files generally have the extension .mp4. The video file selected by the user is sent to the server with a name such as "presentation.mp4." After receiving this video file, the server saves it in a specified directory (e.g., " / videos / "). This saving process makes it possible to manage the video file path in a database.

[0084] Video Analysis

[0085] The server extracts the audio track from the received video file. Specifically, it uses a video processing library called FFmpeg to execute a command to extract only the audio data from the video file. For example, it uses the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary audio file (e.g., " / temp / presentation_audio.mp3").

[0086] Speech to text

[0087] The server sends the extracted audio file to a speech recognition service, which converts the content into text data. This can be done using a cloud-based or local speech recognition service, such as Google's Speech-to-Text API. For example, " / temp / presentation_audio.mp3" is uploaded, and the speech recognition service returns text data such as "Today, I will talk about the progress of the project." The server receives this text data and proceeds to the next step.

[0088] Feedback generation using generative AI

[0089] The server sends the acquired text data to a generative artificial intelligence (e.g., OpenAI's GPT-3). An example of a specific prompt sentence is shown below.

[0090] Example prompt sentence:

[0091] Please provide feedback on the structure, tone, pace, and consistency of the content of the following presentation text:

[0092] Today I'll be talking about the progress of the project.

[0093] Based on this text data, generative AI analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback such as, "The flow of the presentation is easy to understand and the tone is appropriate, but you speak too quickly."

[0094] Notification of results

[0095] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., HTML email or internal message). The user can receive this feedback and get specific guidance on how to improve the quality of their presentation.

[0096] Specific examples

[0097] For example, a case will be described where a user uploads a presentation video called "project_update.mp4."

[0098] 1. The user uses the terminal to upload the "project_update.mp4" file to the system.

[0099] 2. The server saves the video as " / videos / project_update.mp4".

[0100] 3. The server uses the video processing library FFmpeg to extract the audio track from "project_update.mp4" and save it as " / temp / project_update_audio.mp3".

[0101] 4. The server sends " / temp / project_update_audio.mp3" to the speech recognition API and receives the text data "Today I will talk about the progress of the project."

[0102] 5. The server sends this text data to a generative AI, which generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0103] 6. The server formats this feedback and notifies the user via email.

[0104] In this way, the system provides the user with specific feedback to improve the quality of their presentation.

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Program processing steps

[0107] Step 1:

[0108] User uploads video files

[0109] Input: Presentation video file (e.g. "presentation.mp4")

[0110] Process: A user logs in to the system using a browser or a dedicated application, selects a presentation video from the file upload interface, and then clicks the upload button to send the video to the server.

[0111] Output: The video file is received on the server.

[0112] Step 2:

[0113] Video file storage on the server

[0114] Input: The received video file (e.g., "presentation.mp4")

[0115] Processing: The server saves the received file in a specific directory (e.g., " / videos / "), recording the file metadata and save path in a database.

[0116] Output: The saved video file (e.g. " / videos / presentation.mp4") and its path information.

[0117] Step 3:

[0118] Extraction of audio tracks from video files by the server

[0119] Input: Saved video file (e.g. " / videos / presentation.mp4")

[0120] Processing: The server extracts the audio track using FFmpeg, using the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary file.

[0121] Output: The extracted audio file (e.g. " / temp / presentation_audio.mp3").

[0122] Step 4:

[0123] Conversion of voice data into text by the server

[0124] Input: Extracted audio file (e.g. " / temp / presentation_audio.mp3")

[0125] Processing: The server sends the audio file to the speech recognition service (e.g., by sending an HTTP POST request). The speech recognition service analyzes the audio data and generates corresponding text data. The server receives this text data.

[0126] Output: Text data converted from the speech data (e.g., "Today I will talk about the progress of the project").

[0127] Step 5:

[0128] The server sends text data to the AI ​​generator and generates feedback

[0129] Input: Text data generated from speech data (e.g., "Today, I'll talk about the progress of the project.")

[0130] Processing: The server sends the text data to a generative AI model (e.g., GPT-3). It uses prompts to instruct the AI ​​on how to analyze it. For example, the prompt might look like this: "Please provide feedback on the structure, tone, speed, and consistency of the following presentation text: 'Today I'll be talking about the progress of my project.'"

[0131] The generative AI model analyzes the input text data and generates feedback.

[0132] Output: Feedback generated by the generative AI model (e.g., "The flow of your presentation was clear and the tone was appropriate, but you speak too quickly").

[0133] Step 6:

[0134] Server-generated feedback and notification to the user

[0135] Input: Feedback from the generative AI model (e.g., "The flow of the presentation was clear and the tone was appropriate, but you spoke too quickly.")

[0136] Processing: The server formats the feedback and converts it into an appropriate format (e.g. HTML email, JSON for dashboard display, etc.), and then provides the formatted feedback to the user via email or system notification.

[0137] Output: Feedback given to the user.

[0138] In this way, the system provides the user with concrete and objective suggestions for improving their presentation.

[0139] (Application example 1)

[0140] 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."

[0141] Conventional feedback systems for presentation videos provide feedback only to a single user, and have issues with efficiency and scalability when widely used on content distribution platforms. Furthermore, the system does not adequately provide feedback tailored to the type and purpose of the content provided by the user. In particular, it lacks adaptability as a feedback system for creators in content distribution services.

[0142] 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.

[0143] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, means for notifying the user of the generated feedback, and means for providing the generated feedback to a content distribution service, thereby enabling content creators to receive specific and effective feedback to continuously improve the quality of their content.

[0144] "User" means any person or entity that uploads video files to the System and receives generated Feedback.

[0145] "Video File" means a media file in digital format containing audio and visual content uploaded by a User.

[0146] An "audio track" is the audio data contained within a video file.

[0147] "Text data" refers to text information that has been analyzed and converted from an audio track.

[0148] "Generative AI" refers to an AI technology that analyzes content and generates feedback based on pre-set criteria using text data.

[0149] "Feedback" refers to evaluations and improvement suggestions for users' content generated by generative artificial intelligence.

[0150] "Notification means" refers to the method or technique by which generated feedback is communicated to the user.

[0151] "Content distribution service" means an online platform for publishing and distributing content uploaded by users to an audience.

[0152] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides specific feedback. Detailed embodiments of this system are described below.

[0153] Uploading and saving videos

[0154] The user uploads their video file to the server. The server manages the file name and path of the video file to save it in the appropriate directory. This step is performed using Python and OS libraries operated by the server.

[0155] Extracting Audio Tracks

[0156] The server extracts the audio track from the uploaded video file by using a video processing library such as FFmpeg or moviepy to separate the audio from the video file and save it as an audio file.

[0157] Speech to text

[0158] The server sends the extracted audio file to a speech recognition service that converts the audio data into text data, for example, by using the Google Cloud Speech-to-Text API to extract the contents of the audio file as text information.

[0159] Feedback generation using generative AI

[0160] The server sends the acquired text data to a generative AI model, such as OpenAI's GPT-4, which analyzes the input text data and generates feedback based on the presentation's structure, content coherence, tone of voice, speed, and use of visual content. A specific example of a prompt might look like this:

[0161] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[0162] text:

[0163] Today I'd like to talk about the progress of the project. It's currently at 50%...

[0164] Notification of results

[0165] The generated feedback is formatted into an easy-to-understand format by the server and notified to the user or the content distribution service via email or a message within the system.

[0166] Collaboration with content distribution services

[0167] Furthermore, this system provides the generated feedback to the content distribution service, allowing users (creators) to receive specific guidance on how to improve the quality of their content. This collaboration allows creators to more objectively know how their content is being evaluated and what areas need improvement.

[0168] This system allows creators to efficiently improve their content and increase its value to viewers.

[0169] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0170] Step 1:

[0171] A user uploads a video file to a server using a terminal.

[0172] Input: User's video file (e.g. presentation.mp4)

[0173] Data processing: The video file is sent to the server, and the server saves the video file in the appropriate directory.

[0174] Output: Path to the video file saved on the server (e.g. / videos / presentation.mp4)

[0175] Step 2:

[0176] The server extracts the audio track from the received video file.

[0177] Input: Path to the video file stored on the server (e.g. / videos / presentation.mp4)

[0178] Data processing: Use FFmpeg or moviepy to separate the audio track from the video and generate a temporary audio file.

[0179] Output: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[0180] Step 3:

[0181] The server sends the extracted audio file to a speech recognition service, which converts the audio data into text data.

[0182] Input: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[0183] Data processing: Convert the audio data into text data using the Google Cloud Speech-to-Text API.

[0184] Output: Converted text data (e.g., "Today, I'll talk about the progress of the project.")

[0185] Step 4:

[0186] The server sends the text data to the generative AI, which generates feedback.

[0187] Input: Converted text data (e.g., "Today I'll talk about the progress of the project.")

[0188] Data processing: We use OpenAI's GPT-4 to analyze the text data and generate feedback, using the following as an example prompt:

[0189] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[0190] text:

[0191] Today I'd like to talk about the progress of the project. It's currently at 50%...

[0192] Output: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[0193] Step 5:

[0194] The server notifies the user and the content delivery service of the generated feedback.

[0195] Input: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[0196] Data processing: The feedback is formatted into an easy-to-understand format and sent via email or in-system message.

[0197] Output: Feedback communicated to the user and content delivery service

[0198] 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.

[0199] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI) and an emotion engine. This system consists of a server, a terminal, and a user.

[0200] Uploading and saving videos

[0201] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[0202] Video Analysis

[0203] The server extracts the audio track from the saved video file. Using a video processing library (e.g., FFmpeg), the server separates only the audio from the video file and saves the audio data as a temporary audio file, for example, named "presentation_audio.mp3."

[0204] Speech to text

[0205] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, a specific speech recognition API can be used to analyze the audio data and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[0206] Emotion recognition by emotion engine

[0207] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial expression recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[0208] Feedback generation using generative AI

[0209] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback including good points and areas for improvement. Furthermore, the AI ​​adjusts the content and tone of the feedback based on the recognized emotions.

[0210] Notification of results

[0211] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement. The feedback also includes advice based on the user's emotional state.

[0212] Specific examples

[0213] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0214] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[0215] In this way, the system provides users with specific feedback to improve the overall quality of their presentations. By receiving objective and consistent evaluations, users are provided with specific guidelines to help improve their presentations.

[0216] The processing flow will be explained below.

[0217] Step 1:

[0218] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[0219] Step 2:

[0220] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[0221] Step 3:

[0222] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[0223] Step 4:

[0224] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[0225] Step 5:

[0226] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[0227] Step 6:

[0228] The server sends the acquired text data and emotional data to a generative AI. Based on the input data, the generative AI analyzes the presentation's structure, content consistency, voice tone, speed, and visual content used, and generates specific feedback including good points and areas for improvement. It also adjusts the content and tone of the feedback based on the recognized emotions.

[0229] Step 7:

[0230] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[0231] Specific examples

[0232] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0233] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[0234] Example 2

[0235] 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."

[0236] In today's world, evaluating presentations and identifying areas for improvement have become important challenges. However, conventional methods have struggled to provide comprehensive feedback that includes not only the content of the presentation but also the speaker's emotional state. This has made it difficult for users to obtain specific guidelines for improving the overall quality of their presentations.

[0237] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, means for sending the audio track to an emotion recognition engine and analyzing the emotional state, generative artificial intelligence means for analyzing the text data and the emotional data and generating feedback based on preset criteria, and means for notifying the user of the generated feedback. This allows the user to receive comprehensive and specific feedback based on the content of their presentation and their emotional state.

[0238] "User" refers to a person who accesses the system to upload and evaluate presentation videos.

[0239] "Video File" means a video and audio file containing presentation content that a User uploads to the System.

[0240] "Server" refers to a computer device that has a central function for processing and managing data for the entire system.

[0241] "Means for receiving" refers to software or a network interface that allows the server to receive video files from users.

[0242] "Means for extracting the audio track" refers to the video processing software or libraries used by the server to extract only the audio portion of the video file.

[0243] "Audio track" refers only to the audio data contained in a video file.

[0244] "Means for converting to text data" refers to the processes and techniques that convert an audio track into written information using a speech recognition service.

[0245] An "emotion recognition engine" refers to software or algorithms that analyze and identify a speaker's emotional state from audio or video.

[0246] "Generative AI means" refers to artificial intelligence techniques and algorithms for automatically generating feedback based on text data and emotional data.

[0247] "Generating feedback" refers to creating information including an evaluation of the presentation and suggestions for improvement based on the analysis results.

[0248] "Means of notification" refers to email or in-system messaging functions that provide the generated feedback to the user in an easy-to-understand format.

[0249] "Visual content" refers to the visual elements of a presentation, such as charts, images, and slides.

[0250] The present invention provides a system that allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence and an emotion recognition engine. Specific embodiments of this system are described below.

[0251] Uploading and saving videos

[0252] A user accesses the system using a terminal and uploads a presentation video. To do this, the user selects a file, for example, "project_update.mp4." The server receives the video file and saves it in the appropriate directory. The server records the file path in a database so that the file can be accessed in subsequent processing steps.

[0253] Video Analysis

[0254] The server uses a video processing library such as FFmpeg to extract the audio track from the saved video file. Specifically, it executes an FFmpeg command to extract only the audio portion from the video file and saves it as a temporary file named "project_update_audio.mp3."

[0255] Speech to text

[0256] The server sends the "project_update_audio.mp3" file to a speech recognition service (e.g., Google Speech-to-Text API) and analyzes the content as text data. This process converts the audio data into text information, generating text data such as "Today, I'll be talking about the progress of the project."

[0257] Emotion recognition analysis

[0258] The server sends the "project_update_audio.mp3" file to the emotion recognition engine to analyze the speaker's emotional state. The emotion recognition engine identifies the user's emotional state (e.g., nervous, calm, excited, etc.) based on the tone, pitch, and rhythm of the voice. Optionally, facial expression recognition technology can also be used to analyze facial expressions from the video for more accurate emotion recognition.

[0259] Feedback generation using generative AI

[0260] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, voice tone, speed, visual content, and other factors to generate feedback. This feedback includes both good points and areas for improvement. The content and tone of the feedback are also adjusted based on the recognized emotions.

[0261] Notification of results

[0262] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal system message). This allows users to receive specific feedback based on the content and emotional state of their presentation, allowing them to objectively understand what needs to be improved.

[0263] Specific examples

[0264] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." Next, using this audio file, a speech recognition service generates text data such as "Today, I'll talk about the progress of the project." The generative AI analyzes this text data and the emotion data, such as "I'm nervous." As a result, it generates feedback such as "The flow of the presentation is easy to understand, and the tone is appropriate, but you speak too fast." It also includes advice, taking into account the user's nervousness, such as "Practice speaking calmly." This feedback is then sent to the user via email.

[0265] Example prompt sentence:

[0266] Presentation content: Today I will talk about the progress of the project.

[0267] Emotional data: Tension

[0268] Additional information for analysis: None

[0269] In this way, the system can provide users with specific feedback that helps them improve the quality of their presentations, providing them with objective and consistent evaluations and specific guidelines for improving their presentations.

[0270] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0271] Step 1: User uploads a video

[0272] A user accesses the system using their own terminal and uploads a presentation video. For example, they select a file called "project_update.mp4" and upload it.

[0273] Input: Presentation video file (e.g., project_update.mp4)

[0274] Output: The video file is sent to the server.

[0275] Step 2: Storing videos on the server

[0276] The server saves the video file (e.g., project_update.mp4) received from the user in the appropriate directory. The save path is recorded in the database.

[0277] Input: Video file received from user

[0278] Output: Path to the saved video file (e.g. / uploads / project_update.mp4)

[0279] Step 3: Extracting the audio track by the server

[0280] The server uses a video processing library such as FFmpeg to extract the audio track from the video file.

[0281] Specifically, run the FFmpeg command to extract the audio portion from the video file (e.g., / uploads / project_update.mp4) and save it as "project_update_audio.mp3".

[0282] Input: Path to the saved video file

[0283] Output: Extracted audio file (e.g. project_update_audio.mp3)

[0284] Step 4: The server converts the audio file to text.

[0285] The server sends the extracted audio file (e.g., project_update_audio.mp3) to a speech recognition service (e.g., Google Speech-to-Text API) and converts the content into text data.

[0286] Specifically, the voice data is passed to a voice recognition API, and the text data obtained as the API response (e.g., "Today I will talk about the progress of the project") is processed.

[0287] Input: Audio file (e.g., project_update_audio.mp3)

[0288] Output: Text data (e.g., "Today, I'll talk about the progress of the project.")

[0289] Step 5: Emotion recognition analysis by the server

[0290] The server sends the audio file (e.g., project_update_audio.mp3) to the emotion recognition engine to analyze the user's emotional state.

[0291] Specifically, the emotion recognition engine analyzes the tone, pitch, and rhythm of the voice to identify emotions such as tension, calmness, excitement, etc. It also performs facial expression recognition on video data as needed.

[0292] Input: Audio file (e.g., project_update_audio.mp3)

[0293] Output: Emotion data (e.g., tension)

[0294] Step 6: Server sends data to generative AI

[0295] The server sends the acquired text data and emotion data to the generative AI.

[0296] The prompt text should include:

[0297] Presentation content: Today I will talk about the progress of the project.

[0298] Emotional data: Tension

[0299] Additional information for analysis: None

[0300] Input: Text data (e.g., "Today I'll talk about the progress of the project"), emotion data (e.g., nervousness)

[0301] Output: Input data formatted as a prompt

[0302] Step 7: Generative AI feedback generation

[0303] Generative AI uses input data to analyze the presentation's structure, content consistency, tone of voice, speed, visual content, etc., and generates feedback.

[0304] Specifically, the system generates feedback such as, "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly," and offers advice such as, "Practice speaking calmly to reduce nervousness."

[0305] Input: Input data formatted as a prompt

[0306] Output: Generated feedback (e.g., "The flow of your presentation was clear and your tone was appropriate, but you spoke too quickly.")

[0307] Step 8: Server notifies user of results

[0308] The generated feedback is formatted by the server and notified to the user.

[0309] Feedback is provided in an easy-to-understand format (e.g., email or in-system message) so users can review it and understand areas for improvement.

[0310] Input: Generated feedback

[0311] Output: Feedback sent to the user (e.g. email)

[0312] (Application example 2)

[0313] 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."

[0314] Real-time feedback is necessary to improve the quality of training and education in factories. However, this is difficult to achieve with conventional training methods, making it difficult to provide specific guidance for improvement to individual employees. Furthermore, since feedback based on emotional changes is not provided, there is a possibility that employee motivation and technical improvement may be delayed.

[0315] 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.

[0316] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and the user's emotional state and generating feedback based on preset criteria, and means for notifying the user of the generated feedback, thereby enabling real-time, specific, and emotion-based feedback for in-factory training and education.

[0317] "User" means a person who uses the system to upload video files and receive feedback.

[0318] A "video file" is a digital file that contains audio and video data.

[0319] An "audio track" is a portion of the audio data contained within a video file, which is separate from the video data.

[0320] "Text data" refers to data that has been converted from an audio track into text information.

[0321] "Emotional state" refers to the type and intensity of emotions a user displays during a presentation, and is analyzed based on their voice and facial expressions.

[0322] "Generative AI" refers to an AI technology that analyzes input data and generates feedback.

[0323] A "video processing library" is a software tool or library used to extract audio tracks from video files.

[0324] "Feedback" refers to advice and information about improvements and positive actions generated by the system and communicated to the user.

[0325] This invention aims to improve the quality of training and education within a factory through a system installed in a factory robot. In order to implement this invention, the following hardware and software are used.

[0326] Hardware and software used

[0327] Server: High performance server

[0328] Device: Device equipped with a voice recognition microphone and camera

[0329] Video processing library: FFmpeg

[0330] Speech Recognition Software: Google's SpeechRecognition API

[0331] Generative AI software: Transformers (Hugging Face)

[0332] Emotion Recognition Software: EmotionRecognizer

[0333] System configuration

[0334] 1. Upload your video file:

[0335] Users use the terminal to upload training videos to factory robots.

[0336] The server receives this video file and stores it in the appropriate directory.

[0337] 2. Extract audio tracks:

[0338] The server extracts and separates the audio track from the stored video file using emotion recognition software, using the FFmpeg library.

[0339] 3. Transcribe the audio track:

[0340] The server sends the separated audio files to the speech recognition API and receives the content as text data.

[0341] 4. Emotional state analysis:

[0342] The server sends the voice data to emotion recognition software, which analyzes the tone, pitch, and rhythm to identify the user's emotional state.

[0343] 5. Feedback Generation:

[0344] The server sends the captured text and emotion data to the generative AI, which then generates feedback, taking into account factors such as structure, coherence of content, tone of voice, speed, and the user's emotional state.

[0345] 6. Feedback Notification:

[0346] The generated feedback is notified to the user by the server, and is provided in a user-friendly format and sent via email or internal system messages.

[0347] Specific examples

[0348] For example, a factory robot might give a presentation on "safety procedures." The user then uploads the video of the presentation to the system. The server saves the video file, extracts the audio track, and converts it into text. This text data is then sent to a generative AI system, which analyzes it and provides feedback such as, "The procedure explanation is clear, but you explain it too quickly." If emotion-recognition software detects through voice analysis that the user is nervous, the feedback might include specific advice such as, "Practice speaking more calmly."

[0349] Prompt Sentence Examples

[0350] In actual generative AI analysis, the following prompts are used:

[0351] Analyze a presentation on safety procedures within a factory, which includes the following:

[0352] ...

[0353] The user's emotional state is as follows:

[0354] ...

[0355] Use this information to generate feedback on how to improve your presentation.

[0356] In this way, users can obtain specific improvements and advice for training and educational presentations, which can be used to improve technology within their factories.

[0357] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0358] Step 1:

[0359] A user uploads a training video file to the system using a terminal, which then transmits the video file to the server.

[0360] Input: A video file uploaded by the user.

[0361] Data processing: Receiving and storing video files.

[0362] Output: Uploaded video file saved on server.

[0363] Step 2:

[0364] The server uses a video processing library (FFmpeg) to extract the audio track from the received video file.

[0365] Input: Video files stored on the server.

[0366] Data processing: Separating audio tracks from video files and generating audio files.

[0367] Output: Audio file saved on the server.

[0368] Step 3:

[0369] The server sends the extracted audio file to a speech recognition API (Google SpeechRecognition) and converts the audio into text data.

[0370] Input: Audio files stored on the server.

[0371] Data processing: Transcription of audio data.

[0372] Output: Text data extracted from the audio.

[0373] Step 4:

[0374] The server sends the voice data to emotion recognition software (EmotionRecognizer) to analyze the user's emotional state.

[0375] Input: Audio files stored on the server.

[0376] Data processing: Analyzing the tone, pitch, and rhythm of speech to identify emotional states.

[0377] Output: User's emotional state data.

[0378] Step 5:

[0379] The server sends the acquired text data and emotion data to generative artificial intelligence (Transformers), which generates feedback based on the data.

[0380] Input: Text data and emotional state data.

[0381] Data processing: analyzing text and sentiment data, generating feedback.

[0382] Output: Feedback data.

[0383] Step 6:

[0384] The server notifies the user of the generated feedback, which is sent via email or an internal message.

[0385] Input: Feedback data.

[0386] Data processing: Formatting feedback and preparing notifications.

[0387] Output: Feedback sent to the user.

[0388] 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.

[0389] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (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.

[0390] 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.

[0391] [Second embodiment]

[0392] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0393] 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.

[0394] 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).

[0395] 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.

[0396] 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.

[0397] 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).

[0398] 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.

[0399] 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.

[0400] 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.

[0401] 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.

[0402] 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.

[0403] 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."

[0404] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI). This system consists of a server, a terminal, and a user.

[0405] Uploading and saving videos

[0406] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[0407] Video Analysis

[0408] The server extracts the audio track from the stored video file, and using a video processing library (e.g. FFmpeg) separates the audio from the video file and saves it as a temporary audio file for further processing.

[0409] Speech to text

[0410] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, it can analyze the audio data using a specific API (e.g., a speech recognition API) and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[0411] Feedback generation using generative AI

[0412] The server sends the acquired text data to a generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and then generates specific feedback including good points and areas for improvement.

[0413] Notification of results

[0414] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement.

[0415] Specific examples

[0416] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of the project." This text is sent to a generative AI, which analyzes it and generates feedback such as "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly." This feedback is then sent to the user via email, allowing them to learn specific improvements.

[0417] In this way, the system provides users with concrete feedback to improve the quality of their presentations. By receiving objective and consistent evaluations, users are provided with concrete guidelines to help improve their presentations.

[0418] The processing flow will be explained below.

[0419] Step 1:

[0420] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[0421] Step 2:

[0422] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[0423] Step 3:

[0424] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[0425] Step 4:

[0426] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[0427] Step 5:

[0428] The server sends the text data to the generative AI, which then creates and sends a request to the generative AI (e.g., GPT-4) to analyze the presentation text.

[0429] Step 6:

[0430] Generative AI analyzes text data and generates feedback. Based on the user's presentation text, the generative AI evaluates the presentation's structure, content consistency, tone of voice, speed, use of visual content, and other aspects, and generates feedback including good points and areas for improvement.

[0431] Step 7:

[0432] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[0433] Example 1

[0434] 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."

[0435] In order to obtain evaluations and feedback on presentation videos, third-party evaluations or self-feedback are often required. However, these methods are subjective and have problems with maintaining consistency in evaluation criteria. Furthermore, it is difficult to efficiently identify specific areas for improvement, and there is a lack of guidelines for quickly improving the quality of presentations.

[0436] 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.

[0437] In this invention, the server includes means for receiving a video file from a user, means for saving the received video file in a predetermined directory, means for extracting an audio track from the received video file, means for saving the audio track as a temporary audio file, means for transmitting the temporary audio file to a speech recognition service and converting the speech into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, and means for formatting the generated feedback and notifying the user. This allows users to easily obtain objective and consistent feedback, quickly identify specific areas for improvement in their presentations, and improve the quality of their presentations.

[0438] A "user" is a person or organization that accesses the system to upload presentation videos and receive feedback.

[0439] "Video file" refers to a digital file containing presentation content uploaded by a user, with an example file extension of .mp4.

[0440] The "server" is the central part of the system, and is a collection of hardware and software that performs various processes such as receiving, storing, analyzing, generating feedback, and notifying users of video files.

[0441] A "directory" refers to a folder on a file system that is used to organize and store data within a server.

[0442] "Audio Track" refers to the audio data extracted from a video file that contains the audio content of the presentation.

[0443] "Temporary audio file" refers to an audio data file that is generated to temporarily store an audio track and is used for subsequent speech recognition processing.

[0444] "Speech Recognition Service" refers to a cloud-based or local service or API for analyzing and converting voice data into text.

[0445] "Text data" refers to the text information generated by the speech recognition service, and represents the content of the presentation in text form.

[0446] "Generative AI means" refers to an AI model that has an algorithm that generates and analyzes information based on input data, and is responsible for generating feedback for presentations.

[0447] "Feedback" refers to information provided by generative artificial intelligence means that includes an evaluation of the presentation and specific areas for improvement.

[0448] "Notification means" refers to a method or system for communicating generated feedback to the user, including email, dashboard display, etc.

[0449] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides feedback. This system is composed of a server, a terminal, and a user. Specific embodiments are described in detail below.

[0450] Uploading and saving videos

[0451] Users upload presentation videos to the system from their devices using a browser or a dedicated application. Video files generally have the extension .mp4. The video file selected by the user is sent to the server with a name such as "presentation.mp4." After receiving this video file, the server saves it in a specified directory (e.g., " / videos / "). This saving process makes it possible to manage the video file path in a database.

[0452] Video Analysis

[0453] The server extracts the audio track from the received video file. Specifically, it uses a video processing library called FFmpeg to execute a command to extract only the audio data from the video file. For example, it uses the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary audio file (e.g., " / temp / presentation_audio.mp3").

[0454] Speech to text

[0455] The server sends the extracted audio file to a speech recognition service, which converts it into text. This can be done using a cloud-based or local speech recognition service, such as Google's Speech-to-Text API. For example, " / temp / presentation_audio.mp3" is uploaded, and the speech recognition service returns text data such as "Today, I'll talk about the progress of the project." The server receives this text data and proceeds to the next step.

[0456] Feedback generation using generative AI

[0457] The server sends the acquired text data to a generative artificial intelligence (e.g., OpenAI's GPT-3). An example of a specific prompt sentence is shown below.

[0458] Example prompt sentence:

[0459] Please provide feedback on the structure, tone, pace, and consistency of the content of the following presentation text:

[0460] Today I'll be talking about the progress of the project.

[0461] Based on this text data, generative AI analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback such as, "The flow of the presentation is easy to understand and the tone is appropriate, but you speak too quickly."

[0462] Notification of results

[0463] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., HTML email or internal message). The user can receive this feedback and get specific guidance on how to improve the quality of their presentation.

[0464] Specific examples

[0465] For example, a case will be described where a user uploads a presentation video called "project_update.mp4."

[0466] 1. The user uses the terminal to upload the "project_update.mp4" file to the system.

[0467] 2. The server saves the video as " / videos / project_update.mp4".

[0468] 3. The server uses the video processing library FFmpeg to extract the audio track from "project_update.mp4" and save it as " / temp / project_update_audio.mp3".

[0469] 4. The server sends " / temp / project_update_audio.mp3" to the speech recognition API and receives the text data "Today I will talk about the progress of the project."

[0470] 5. The server sends this text data to a generative AI, which generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0471] 6. The server formats this feedback and notifies the user via email.

[0472] In this way, the system provides the user with specific feedback to improve the quality of their presentation.

[0473] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0474] Program processing steps

[0475] Step 1:

[0476] User uploads video files

[0477] Input: Presentation video file (e.g. "presentation.mp4")

[0478] Process: A user logs in to the system using a browser or a dedicated application, selects a presentation video from the file upload interface, and then clicks the upload button to send the video to the server.

[0479] Output: The video file is received on the server.

[0480] Step 2:

[0481] Video file storage on the server

[0482] Input: The received video file (e.g., "presentation.mp4")

[0483] Processing: The server saves the received file in a specific directory (e.g., " / videos / "), recording the file metadata and save path in a database.

[0484] Output: The saved video file (e.g. " / videos / presentation.mp4") and its path information.

[0485] Step 3:

[0486] Extraction of audio tracks from video files by the server

[0487] Input: Saved video file (e.g. " / videos / presentation.mp4")

[0488] Processing: The server extracts the audio track using FFmpeg, using the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary file.

[0489] Output: The extracted audio file (e.g. " / temp / presentation_audio.mp3").

[0490] Step 4:

[0491] Conversion of voice data into text by the server

[0492] Input: Extracted audio file (e.g. " / temp / presentation_audio.mp3")

[0493] Processing: The server sends the audio file to the speech recognition service (e.g., by sending an HTTP POST request). The speech recognition service analyzes the audio data and generates corresponding text data. The server receives this text data.

[0494] Output: Text data converted from the speech data (e.g., "Today I will talk about the progress of the project").

[0495] Step 5:

[0496] The server sends text data to the AI ​​generator and generates feedback

[0497] Input: Text data generated from speech data (e.g., "Today, I'll talk about the progress of the project.")

[0498] Processing: The server sends the text data to a generative AI model (e.g., GPT-3). It uses prompts to instruct the AI ​​on how to analyze it. For example, the prompt might look like this: "Please provide feedback on the structure, tone, speed, and consistency of the following presentation text: 'Today I'll be talking about the progress of my project.'"

[0499] The generative AI model analyzes the input text data and generates feedback.

[0500] Output: Feedback generated by the generative AI model (e.g., "The flow of your presentation was clear and the tone was appropriate, but you speak too quickly").

[0501] Step 6:

[0502] Server-generated feedback and notification to the user

[0503] Input: Feedback from the generative AI model (e.g., "The flow of the presentation was clear and the tone was appropriate, but you spoke too quickly.")

[0504] Processing: The server formats the feedback and converts it into an appropriate format (e.g. HTML email, JSON for dashboard display, etc.), and then provides the formatted feedback to the user via email or system notification.

[0505] Output: Feedback given to the user.

[0506] In this way, the system provides the user with concrete and objective suggestions for improving their presentation.

[0507] (Application example 1)

[0508] 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."

[0509] Conventional feedback systems for presentation videos provide feedback only to a single user, and have issues with efficiency and scalability when widely used on content distribution platforms. Furthermore, the system does not adequately provide feedback tailored to the type and purpose of the content provided by the user. In particular, it lacks adaptability as a feedback system for creators in content distribution services.

[0510] 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.

[0511] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, means for notifying the user of the generated feedback, and means for providing the generated feedback to a content distribution service, thereby enabling content creators to receive specific and effective feedback to continuously improve the quality of their content.

[0512] "User" means any person or entity that uploads video files to the System and receives generated Feedback.

[0513] "Video File" means a media file in digital format containing audio and visual content uploaded by a User.

[0514] An "audio track" is the audio data contained within a video file.

[0515] "Text data" refers to text information that has been analyzed and converted from an audio track.

[0516] "Generative AI" refers to an AI technology that analyzes content and generates feedback based on pre-set criteria using text data.

[0517] "Feedback" refers to evaluations and improvement suggestions for users' content generated by generative artificial intelligence.

[0518] "Notification means" refers to the method or technique by which generated feedback is communicated to the user.

[0519] "Content distribution service" means an online platform for publishing and distributing content uploaded by users to an audience.

[0520] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides specific feedback. Detailed embodiments of this system are described below.

[0521] Uploading and saving videos

[0522] The user uploads their video file to the server. The server manages the file name and path of the video file to save it in the appropriate directory. This step is performed using Python and OS libraries operated by the server.

[0523] Extracting Audio Tracks

[0524] The server extracts the audio track from the uploaded video file by using a video processing library such as FFmpeg or moviepy to separate the audio from the video file and save it as an audio file.

[0525] Speech to text

[0526] The server sends the extracted audio file to a speech recognition service that converts the audio data into text data, for example, by using the Google Cloud Speech-to-Text API to extract the contents of the audio file as text information.

[0527] Feedback generation using generative AI

[0528] The server sends the acquired text data to a generative AI model, such as OpenAI's GPT-4, which analyzes the input text data and generates feedback based on the presentation's structure, content coherence, tone of voice, speed, and use of visual content. A specific example of a prompt might look like this:

[0529] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[0530] text:

[0531] Today I'd like to talk about the progress of the project. It's currently at 50%...

[0532] Notification of results

[0533] The generated feedback is formatted into an easy-to-understand format by the server and notified to the user or the content distribution service via email or a message within the system.

[0534] Collaboration with content distribution services

[0535] Furthermore, this system provides the generated feedback to the content distribution service, allowing users (creators) to receive specific guidance on how to improve the quality of their content. This collaboration allows creators to more objectively know how their content is being evaluated and what areas need improvement.

[0536] This system allows creators to efficiently improve their content and increase its value to viewers.

[0537] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0538] Step 1:

[0539] A user uploads a video file to a server using a terminal.

[0540] Input: User's video file (e.g. presentation.mp4)

[0541] Data processing: The video file is sent to the server, and the server saves the video file in the appropriate directory.

[0542] Output: Path to the video file saved on the server (e.g. / videos / presentation.mp4)

[0543] Step 2:

[0544] The server extracts the audio track from the received video file.

[0545] Input: Path to the video file stored on the server (e.g. / videos / presentation.mp4)

[0546] Data processing: Use FFmpeg or moviepy to separate the audio track from the video and generate a temporary audio file.

[0547] Output: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[0548] Step 3:

[0549] The server sends the extracted audio file to a speech recognition service, which converts the audio data into text data.

[0550] Input: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[0551] Data processing: Convert the audio data into text data using the Google Cloud Speech-to-Text API.

[0552] Output: Converted text data (e.g., "Today, I'll talk about the progress of the project.")

[0553] Step 4:

[0554] The server sends the text data to the generative AI, which generates feedback.

[0555] Input: Converted text data (e.g., "Today I'll talk about the progress of the project.")

[0556] Data processing: We use OpenAI's GPT-4 to analyze the text data and generate feedback, using the following as an example prompt:

[0557] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[0558] text:

[0559] Today I'd like to talk about the progress of the project. It's currently at 50%...

[0560] Output: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[0561] Step 5:

[0562] The server notifies the user and the content delivery service of the generated feedback.

[0563] Input: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[0564] Data processing: The feedback is formatted into an easy-to-understand format and sent via email or in-system message.

[0565] Output: Feedback communicated to the user and content delivery service

[0566] 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.

[0567] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI) and an emotion engine. This system consists of a server, a terminal, and a user.

[0568] Uploading and saving videos

[0569] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[0570] Video Analysis

[0571] The server extracts the audio track from the saved video file. Using a video processing library (e.g., FFmpeg), the server separates only the audio from the video file and saves the audio data as a temporary audio file, for example, named "presentation_audio.mp3."

[0572] Speech to text

[0573] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, a specific speech recognition API can be used to analyze the audio data and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[0574] Emotion recognition by emotion engine

[0575] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial expression recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[0576] Feedback generation using generative AI

[0577] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback including good points and areas for improvement. Furthermore, the AI ​​adjusts the content and tone of the feedback based on the recognized emotions.

[0578] Notification of results

[0579] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement. The feedback also includes advice based on the user's emotional state.

[0580] Specific examples

[0581] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0582] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[0583] In this way, the system provides users with specific feedback to improve the overall quality of their presentations. By receiving objective and consistent evaluations, users are provided with specific guidelines to help improve their presentations.

[0584] The processing flow will be explained below.

[0585] Step 1:

[0586] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[0587] Step 2:

[0588] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[0589] Step 3:

[0590] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[0591] Step 4:

[0592] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[0593] Step 5:

[0594] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[0595] Step 6:

[0596] The server sends the acquired text data and emotional data to a generative AI. Based on the input data, the generative AI analyzes the presentation's structure, content consistency, voice tone, speed, and visual content used, and generates specific feedback including good points and areas for improvement. It also adjusts the content and tone of the feedback based on the recognized emotions.

[0597] Step 7:

[0598] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[0599] Specific examples

[0600] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0601] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[0602] Example 2

[0603] 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."

[0604] In today's world, evaluating presentations and identifying areas for improvement have become important challenges. However, conventional methods have struggled to provide comprehensive feedback that includes not only the content of the presentation but also the speaker's emotional state. This has made it difficult for users to obtain specific guidelines for improving the overall quality of their presentations.

[0605] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, means for sending the audio track to an emotion recognition engine and analyzing the emotional state, generative artificial intelligence means for analyzing the text data and the emotional data and generating feedback based on preset criteria, and means for notifying the user of the generated feedback. This allows the user to receive comprehensive and specific feedback based on the content of their presentation and their emotional state.

[0606] "User" refers to a person who accesses the system to upload and evaluate presentation videos.

[0607] "Video File" means a video and audio file containing presentation content that a User uploads to the System.

[0608] "Server" refers to a computer device that has a central function for processing and managing data for the entire system.

[0609] "Means for receiving" refers to software or a network interface that allows the server to receive video files from users.

[0610] "Means for extracting the audio track" refers to the video processing software or libraries used by the server to extract only the audio portion of the video file.

[0611] "Audio track" refers only to the audio data contained in a video file.

[0612] "Means for converting to text data" refers to the processes and techniques that convert an audio track into written information using a speech recognition service.

[0613] An "emotion recognition engine" refers to software or algorithms that analyze and identify a speaker's emotional state from audio or video.

[0614] "Generative AI means" refers to artificial intelligence techniques and algorithms for automatically generating feedback based on text data and emotional data.

[0615] "Generating feedback" refers to creating information including an evaluation of the presentation and suggestions for improvement based on the analysis results.

[0616] "Means of notification" refers to email or in-system messaging functions that provide the generated feedback to the user in an easy-to-understand format.

[0617] "Visual content" refers to the visual elements of a presentation, such as charts, images, and slides.

[0618] The present invention provides a system that allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence and an emotion recognition engine. Specific embodiments of this system are described below.

[0619] Uploading and saving videos

[0620] A user accesses the system using a terminal and uploads a presentation video. To do this, the user selects a file, for example, "project_update.mp4." The server receives the video file and saves it in the appropriate directory. The server records the file path in a database so that the file can be accessed in subsequent processing steps.

[0621] Video Analysis

[0622] The server uses a video processing library such as FFmpeg to extract the audio track from the saved video file. Specifically, it executes an FFmpeg command to extract only the audio portion from the video file and saves it as a temporary file named "project_update_audio.mp3."

[0623] Speech to text

[0624] The server sends the "project_update_audio.mp3" file to a speech recognition service (e.g., Google Speech-to-Text API) and analyzes the content as text data. This process converts the audio data into text information, generating text data such as "Today, I'll be talking about the progress of the project."

[0625] Emotion recognition analysis

[0626] The server sends the "project_update_audio.mp3" file to the emotion recognition engine to analyze the speaker's emotional state. The emotion recognition engine identifies the user's emotional state (e.g., nervous, calm, excited, etc.) based on the tone, pitch, and rhythm of the voice. Optionally, facial expression recognition technology can also be used to analyze facial expressions from the video for more accurate emotion recognition.

[0627] Feedback generation using generative AI

[0628] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, voice tone, speed, visual content, and other factors to generate feedback. This feedback includes both good points and areas for improvement. The content and tone of the feedback are also adjusted based on the recognized emotions.

[0629] Notification of results

[0630] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal system message). This allows users to receive specific feedback based on the content and emotional state of their presentation, allowing them to objectively understand what needs to be improved.

[0631] Specific examples

[0632] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." Next, using this audio file, a speech recognition service generates text data such as "Today, I'll talk about the progress of the project." The generative AI analyzes this text data and the emotion data, such as "I'm nervous." As a result, it generates feedback such as "The flow of the presentation is easy to understand, and the tone is appropriate, but you speak too fast." It also includes advice, taking into account the user's nervousness, such as "Practice speaking calmly." This feedback is then sent to the user via email.

[0633] Example prompt sentence:

[0634] Presentation content: Today I will talk about the progress of the project.

[0635] Emotional data: Tension

[0636] Additional information for analysis: None

[0637] In this way, the system can provide users with specific feedback that helps them improve the quality of their presentations, providing them with objective and consistent evaluations and specific guidelines for improving their presentations.

[0638] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0639] Step 1: User uploads a video

[0640] A user accesses the system using their own terminal and uploads a presentation video. For example, they select a file called "project_update.mp4" and upload it.

[0641] Input: Presentation video file (e.g., project_update.mp4)

[0642] Output: The video file is sent to the server.

[0643] Step 2: Storing videos on the server

[0644] The server saves the video file (e.g., project_update.mp4) received from the user in the appropriate directory. The save path is recorded in the database.

[0645] Input: Video file received from user

[0646] Output: Path to the saved video file (e.g. / uploads / project_update.mp4)

[0647] Step 3: Extracting the audio track by the server

[0648] The server uses a video processing library such as FFmpeg to extract the audio track from the video file.

[0649] Specifically, run the FFmpeg command to extract the audio portion from the video file (e.g., / uploads / project_update.mp4) and save it as "project_update_audio.mp3".

[0650] Input: Path to the saved video file

[0651] Output: Extracted audio file (e.g. project_update_audio.mp3)

[0652] Step 4: The server converts the audio file to text.

[0653] The server sends the extracted audio file (e.g., project_update_audio.mp3) to a speech recognition service (e.g., Google Speech-to-Text API) and converts the content into text data.

[0654] Specifically, the voice data is passed to a voice recognition API, and the text data obtained as the API response (e.g., "Today I will talk about the progress of the project") is processed.

[0655] Input: Audio file (e.g., project_update_audio.mp3)

[0656] Output: Text data (e.g., "Today, I'll talk about the progress of the project.")

[0657] Step 5: Emotion recognition analysis by the server

[0658] The server sends the audio file (e.g., project_update_audio.mp3) to the emotion recognition engine to analyze the user's emotional state.

[0659] Specifically, the emotion recognition engine analyzes the tone, pitch, and rhythm of the voice to identify emotions such as tension, calmness, excitement, etc. It also performs facial expression recognition on video data as needed.

[0660] Input: Audio file (e.g., project_update_audio.mp3)

[0661] Output: Emotion data (e.g., tension)

[0662] Step 6: Server sends data to generative AI

[0663] The server sends the acquired text data and emotion data to the generative AI.

[0664] The prompt text should include:

[0665] Presentation content: Today I will talk about the progress of the project.

[0666] Emotional data: Tension

[0667] Additional information for analysis: None

[0668] Input: Text data (e.g., "Today I'll talk about the progress of the project"), emotion data (e.g., nervousness)

[0669] Output: Input data formatted as a prompt

[0670] Step 7: Generative AI feedback generation

[0671] Generative AI uses input data to analyze the presentation's structure, content consistency, tone of voice, speed, visual content, etc., and generates feedback.

[0672] Specifically, the system generates feedback such as, "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly," and offers advice such as, "Practice speaking calmly to reduce nervousness."

[0673] Input: Input data formatted as a prompt

[0674] Output: Generated feedback (e.g., "The flow of your presentation was clear and your tone was appropriate, but you spoke too quickly.")

[0675] Step 8: Server notifies user of results

[0676] The generated feedback is formatted by the server and notified to the user.

[0677] Feedback is provided in an easy-to-understand format (e.g., email or in-system message) so users can review it and understand areas for improvement.

[0678] Input: Generated feedback

[0679] Output: Feedback sent to the user (e.g. email)

[0680] (Application example 2)

[0681] 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."

[0682] Real-time feedback is necessary to improve the quality of training and education in factories. However, this is difficult to achieve with conventional training methods, making it difficult to provide specific guidance for improvement to individual employees. Furthermore, since feedback based on emotional changes is not provided, there is a possibility that employee motivation and technical improvement may be delayed.

[0683] 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.

[0684] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and the user's emotional state and generating feedback based on preset criteria, and means for notifying the user of the generated feedback, thereby enabling real-time, specific, and emotion-based feedback for in-factory training and education.

[0685] "User" means a person who uses the system to upload video files and receive feedback.

[0686] A "video file" is a digital file that contains audio and video data.

[0687] An "audio track" is a portion of the audio data contained within a video file, which is separate from the video data.

[0688] "Text data" refers to data that has been converted from an audio track into text information.

[0689] "Emotional state" refers to the type and intensity of emotions a user displays during a presentation, and is analyzed based on their voice and facial expressions.

[0690] "Generative AI" refers to an AI technology that analyzes input data and generates feedback.

[0691] A "video processing library" is a software tool or library used to extract audio tracks from video files.

[0692] "Feedback" refers to advice and information about improvements and positive actions generated by the system and communicated to the user.

[0693] This invention aims to improve the quality of training and education within a factory through a system installed in a factory robot. In order to implement this invention, the following hardware and software are used.

[0694] Hardware and software used

[0695] Server: High performance server

[0696] Device: Device equipped with a voice recognition microphone and camera

[0697] Video processing library: FFmpeg

[0698] Speech Recognition Software: Google's SpeechRecognition API

[0699] Generative AI software: Transformers (Hugging Face)

[0700] Emotion Recognition Software: EmotionRecognizer

[0701] System configuration

[0702] 1. Upload your video file:

[0703] Users use the terminal to upload training videos to factory robots.

[0704] The server receives this video file and stores it in the appropriate directory.

[0705] 2. Extract audio tracks:

[0706] The server extracts and separates the audio track from the stored video file using emotion recognition software, using the FFmpeg library.

[0707] 3. Transcribe the audio track:

[0708] The server sends the separated audio files to the speech recognition API and receives the content as text data.

[0709] 4. Emotional state analysis:

[0710] The server sends the voice data to emotion recognition software, which analyzes the tone, pitch, and rhythm to identify the user's emotional state.

[0711] 5. Feedback Generation:

[0712] The server sends the captured text and emotion data to the generative AI, which then generates feedback, taking into account factors such as structure, coherence of content, tone of voice, speed, and the user's emotional state.

[0713] 6. Feedback Notification:

[0714] The generated feedback is notified to the user by the server, and is provided in a user-friendly format and sent via email or internal system messages.

[0715] Specific examples

[0716] For example, a factory robot might give a presentation on "safety procedures." The user then uploads the video of the presentation to the system. The server saves the video file, extracts the audio track, and converts it into text. This text data is then sent to a generative AI system, which analyzes it and provides feedback such as, "The procedure explanation is clear, but you explain it too quickly." If emotion-recognition software detects through voice analysis that the user is nervous, the feedback might include specific advice such as, "Practice speaking more calmly."

[0717] Prompt Sentence Examples

[0718] In actual generative AI analysis, the following prompts are used:

[0719] Analyze a presentation on safety procedures within a factory, which includes the following:

[0720] ...

[0721] The user's emotional state is as follows:

[0722] ...

[0723] Use this information to generate feedback on how to improve your presentation.

[0724] In this way, users can obtain specific improvements and advice for training and educational presentations, which can be used to improve technology within their factories.

[0725] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0726] Step 1:

[0727] A user uploads a training video file to the system using a terminal, which then transmits the video file to the server.

[0728] Input: A video file uploaded by the user.

[0729] Data processing: Receiving and storing video files.

[0730] Output: Uploaded video file saved on server.

[0731] Step 2:

[0732] The server uses a video processing library (FFmpeg) to extract the audio track from the received video file.

[0733] Input: Video files stored on the server.

[0734] Data processing: Separating audio tracks from video files and generating audio files.

[0735] Output: Audio file saved on the server.

[0736] Step 3:

[0737] The server sends the extracted audio file to a speech recognition API (Google SpeechRecognition) and converts the audio into text data.

[0738] Input: Audio files stored on the server.

[0739] Data processing: Transcription of audio data.

[0740] Output: Text data extracted from the audio.

[0741] Step 4:

[0742] The server sends the voice data to emotion recognition software (EmotionRecognizer) to analyze the user's emotional state.

[0743] Input: Audio files stored on the server.

[0744] Data processing: Analyzing the tone, pitch, and rhythm of speech to identify emotional states.

[0745] Output: User's emotional state data.

[0746] Step 5:

[0747] The server sends the acquired text data and emotion data to generative artificial intelligence (Transformers), which generates feedback based on the data.

[0748] Input: Text data and emotional state data.

[0749] Data processing: analyzing text and sentiment data, generating feedback.

[0750] Output: Feedback data.

[0751] Step 6:

[0752] The server notifies the user of the generated feedback, which is sent via email or an internal message.

[0753] Input: Feedback data.

[0754] Data processing: Formatting feedback and preparing notifications.

[0755] Output: Feedback sent to the user.

[0756] 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.

[0757] 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.

[0758] 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.

[0759] [Third embodiment]

[0760] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0761] 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.

[0762] 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).

[0763] 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.

[0764] 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.

[0765] 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).

[0766] 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.

[0767] 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.

[0768] 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.

[0769] 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.

[0770] 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.

[0771] 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."

[0772] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI). This system consists of a server, a terminal, and a user.

[0773] Uploading and saving videos

[0774] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[0775] Video Analysis

[0776] The server extracts the audio track from the stored video file, and using a video processing library (e.g. FFmpeg) separates the audio from the video file and saves it as a temporary audio file for further processing.

[0777] Speech to text

[0778] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, it can analyze the audio data using a specific API (e.g., a speech recognition API) and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[0779] Feedback generation using generative AI

[0780] The server sends the acquired text data to a generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and then generates specific feedback including good points and areas for improvement.

[0781] Notification of results

[0782] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement.

[0783] Specific examples

[0784] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of the project." This text is sent to a generative AI, which analyzes it and generates feedback such as "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly." This feedback is then sent to the user via email, allowing them to learn specific improvements.

[0785] In this way, the system provides users with concrete feedback to improve the quality of their presentations. By receiving objective and consistent evaluations, users are provided with concrete guidelines to help improve their presentations.

[0786] The processing flow will be explained below.

[0787] Step 1:

[0788] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[0789] Step 2:

[0790] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[0791] Step 3:

[0792] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[0793] Step 4:

[0794] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[0795] Step 5:

[0796] The server sends the text data to the generative AI, which then creates and sends a request to the generative AI (e.g., GPT-4) to analyze the presentation text.

[0797] Step 6:

[0798] Generative AI analyzes text data and generates feedback. Based on the user's presentation text, the generative AI evaluates the presentation's structure, content consistency, tone of voice, speed, use of visual content, and other aspects, and generates feedback including good points and areas for improvement.

[0799] Step 7:

[0800] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[0801] Example 1

[0802] 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."

[0803] In order to obtain evaluations and feedback on presentation videos, third-party evaluations or self-feedback are often required. However, these methods are subjective and have problems with maintaining consistency in evaluation criteria. Furthermore, it is difficult to efficiently identify specific areas for improvement, and there is a lack of guidelines for quickly improving the quality of presentations.

[0804] 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.

[0805] In this invention, the server includes means for receiving a video file from a user, means for saving the received video file in a predetermined directory, means for extracting an audio track from the received video file, means for saving the audio track as a temporary audio file, means for transmitting the temporary audio file to a speech recognition service and converting the speech into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, and means for formatting the generated feedback and notifying the user. This allows users to easily obtain objective and consistent feedback, quickly identify specific areas for improvement in their presentations, and improve the quality of their presentations.

[0806] A "user" is a person or organization that accesses the system to upload presentation videos and receive feedback.

[0807] "Video file" refers to a digital file containing presentation content uploaded by a user, with an example file extension of .mp4.

[0808] The "server" is the central part of the system, and is a collection of hardware and software that performs various processes such as receiving, storing, analyzing, generating feedback, and notifying users of video files.

[0809] A "directory" refers to a folder on a file system that is used to organize and store data within a server.

[0810] "Audio Track" refers to the audio data extracted from a video file that contains the audio content of the presentation.

[0811] "Temporary audio file" refers to an audio data file that is generated to temporarily store an audio track and is used for subsequent speech recognition processing.

[0812] "Speech Recognition Service" refers to a cloud-based or local service or API for analyzing and converting voice data into text.

[0813] "Text data" refers to the text information generated by the speech recognition service, and represents the content of the presentation in text form.

[0814] "Generative AI means" refers to an AI model that has an algorithm that generates and analyzes information based on input data, and is responsible for generating feedback for presentations.

[0815] "Feedback" refers to information provided by generative artificial intelligence means that includes an evaluation of the presentation and specific areas for improvement.

[0816] "Notification means" refers to a method or system for communicating generated feedback to the user, including email, dashboard display, etc.

[0817] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides feedback. This system is composed of a server, a terminal, and a user. Specific embodiments are described in detail below.

[0818] Uploading and saving videos

[0819] Users upload presentation videos to the system from their devices using a browser or a dedicated application. Video files generally have the extension .mp4. The video file selected by the user is sent to the server with a name such as "presentation.mp4." After receiving this video file, the server saves it in a specified directory (e.g., " / videos / "). This saving process makes it possible to manage the video file path in a database.

[0820] Video Analysis

[0821] The server extracts the audio track from the received video file. Specifically, it uses a video processing library called FFmpeg to execute a command to extract only the audio data from the video file. For example, it uses the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary audio file (e.g., " / temp / presentation_audio.mp3").

[0822] Speech to text

[0823] The server sends the extracted audio file to a speech recognition service, which converts it into text. This can be done using a cloud-based or local speech recognition service, such as Google's Speech-to-Text API. For example, " / temp / presentation_audio.mp3" is uploaded, and the speech recognition service returns text data such as "Today, I'll talk about the progress of the project." The server receives this text data and proceeds to the next step.

[0824] Feedback generation using generative AI

[0825] The server sends the acquired text data to a generative artificial intelligence (e.g., OpenAI's GPT-3). An example of a specific prompt sentence is shown below.

[0826] Example prompt sentence:

[0827] Please provide feedback on the structure, tone, pace, and consistency of the content of the following presentation text:

[0828] Today I'll be talking about the progress of the project.

[0829] Based on this text data, generative AI analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback such as, "The flow of the presentation is easy to understand and the tone is appropriate, but you speak too quickly."

[0830] Notification of results

[0831] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., HTML email or internal message). The user can receive this feedback and get specific guidance on how to improve the quality of their presentation.

[0832] Specific examples

[0833] For example, a case will be described where a user uploads a presentation video called "project_update.mp4."

[0834] 1. The user uses the terminal to upload the "project_update.mp4" file to the system.

[0835] 2. The server saves the video as " / videos / project_update.mp4".

[0836] 3. The server uses the video processing library FFmpeg to extract the audio track from "project_update.mp4" and save it as " / temp / project_update_audio.mp3".

[0837] 4. The server sends " / temp / project_update_audio.mp3" to the speech recognition API and receives the text data "Today I will talk about the progress of the project."

[0838] 5. The server sends this text data to a generative AI, which generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0839] 6. The server formats this feedback and notifies the user via email.

[0840] In this way, the system provides the user with specific feedback to improve the quality of their presentation.

[0841] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0842] Program processing steps

[0843] Step 1:

[0844] User uploads video files

[0845] Input: Presentation video file (e.g. "presentation.mp4")

[0846] Process: A user logs in to the system using a browser or a dedicated application, selects a presentation video from the file upload interface, and then clicks the upload button to send the video to the server.

[0847] Output: The video file is received on the server.

[0848] Step 2:

[0849] Video file storage on the server

[0850] Input: The received video file (e.g., "presentation.mp4")

[0851] Processing: The server saves the received file in a specific directory (e.g., " / videos / "), recording the file metadata and save path in a database.

[0852] Output: The saved video file (e.g. " / videos / presentation.mp4") and its path information.

[0853] Step 3:

[0854] Extraction of audio tracks from video files by the server

[0855] Input: Saved video file (e.g. " / videos / presentation.mp4")

[0856] Processing: The server extracts the audio track using FFmpeg, using the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary file.

[0857] Output: The extracted audio file (e.g. " / temp / presentation_audio.mp3").

[0858] Step 4:

[0859] Conversion of voice data into text by the server

[0860] Input: Extracted audio file (e.g. " / temp / presentation_audio.mp3")

[0861] Processing: The server sends the audio file to the speech recognition service (e.g., by sending an HTTP POST request). The speech recognition service analyzes the audio data and generates corresponding text data. The server receives this text data.

[0862] Output: Text data converted from the speech data (e.g., "Today I will talk about the progress of the project").

[0863] Step 5:

[0864] The server sends text data to the AI ​​generator and generates feedback

[0865] Input: Text data generated from speech data (e.g., "Today, I'll talk about the progress of the project.")

[0866] Processing: The server sends the text data to a generative AI model (e.g., GPT-3). It uses prompts to instruct the AI ​​on how to analyze it. For example, the prompt might look like this: "Please provide feedback on the structure, tone, speed, and consistency of the following presentation text: 'Today I'll be talking about the progress of my project.'"

[0867] The generative AI model analyzes the input text data and generates feedback.

[0868] Output: Feedback generated by the generative AI model (e.g., "The flow of your presentation was clear and the tone was appropriate, but you speak too quickly").

[0869] Step 6:

[0870] Server-generated feedback and notification to the user

[0871] Input: Feedback from the generative AI model (e.g., "The flow of the presentation was clear and the tone was appropriate, but you spoke too quickly.")

[0872] Processing: The server formats the feedback and converts it into an appropriate format (e.g. HTML email, JSON for dashboard display, etc.), and then provides the formatted feedback to the user via email or system notification.

[0873] Output: Feedback given to the user.

[0874] In this way, the system provides the user with concrete and objective suggestions for improving their presentation.

[0875] (Application example 1)

[0876] 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."

[0877] Conventional feedback systems for presentation videos provide feedback only to a single user, and have issues with efficiency and scalability when widely used on content distribution platforms. Furthermore, the system does not adequately provide feedback tailored to the type and purpose of the content provided by the user. In particular, it lacks adaptability as a feedback system for creators in content distribution services.

[0878] 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.

[0879] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, means for notifying the user of the generated feedback, and means for providing the generated feedback to a content distribution service, thereby enabling content creators to receive specific and effective feedback to continuously improve the quality of their content.

[0880] "User" means any person or entity that uploads video files to the System and receives generated Feedback.

[0881] "Video File" means a media file in digital format containing audio and visual content uploaded by a User.

[0882] An "audio track" is the audio data contained within a video file.

[0883] "Text data" refers to text information that has been analyzed and converted from an audio track.

[0884] "Generative AI" refers to an AI technology that analyzes content and generates feedback based on pre-set criteria using text data.

[0885] "Feedback" refers to evaluations and improvement suggestions for users' content generated by generative artificial intelligence.

[0886] "Notification means" refers to the method or technique by which generated feedback is communicated to the user.

[0887] "Content distribution service" means an online platform for publishing and distributing content uploaded by users to an audience.

[0888] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides specific feedback. Detailed embodiments of this system are described below.

[0889] Uploading and saving videos

[0890] The user uploads their video file to the server. The server manages the file name and path of the video file to save it in the appropriate directory. This step is performed using Python and OS libraries operated by the server.

[0891] Extracting Audio Tracks

[0892] The server extracts the audio track from the uploaded video file by using a video processing library such as FFmpeg or moviepy to separate the audio from the video file and save it as an audio file.

[0893] Speech to text

[0894] The server sends the extracted audio file to a speech recognition service that converts the audio data into text data, for example, by using the Google Cloud Speech-to-Text API to extract the contents of the audio file as text information.

[0895] Feedback generation using generative AI

[0896] The server sends the acquired text data to a generative AI model, such as OpenAI's GPT-4, which analyzes the input text data and generates feedback based on the presentation's structure, content coherence, tone of voice, speed, and use of visual content. A specific example of a prompt might look like this:

[0897] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[0898] text:

[0899] Today I'd like to talk about the progress of the project. It's currently at 50%...

[0900] Notification of results

[0901] The generated feedback is formatted into an easy-to-understand format by the server and notified to the user or the content distribution service via email or a message within the system.

[0902] Collaboration with content distribution services

[0903] Furthermore, this system provides the generated feedback to the content distribution service, allowing users (creators) to receive specific guidance on how to improve the quality of their content. This collaboration allows creators to more objectively know how their content is being evaluated and what areas need improvement.

[0904] This system allows creators to efficiently improve their content and increase its value to viewers.

[0905] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0906] Step 1:

[0907] A user uploads a video file to a server using a terminal.

[0908] Input: User's video file (e.g. presentation.mp4)

[0909] Data processing: The video file is sent to the server, and the server saves the video file in the appropriate directory.

[0910] Output: Path to the video file saved on the server (e.g. / videos / presentation.mp4)

[0911] Step 2:

[0912] The server extracts the audio track from the received video file.

[0913] Input: Path to the video file stored on the server (e.g. / videos / presentation.mp4)

[0914] Data processing: Use FFmpeg or moviepy to separate the audio track from the video and generate a temporary audio file.

[0915] Output: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[0916] Step 3:

[0917] The server sends the extracted audio file to a speech recognition service, which converts the audio data into text data.

[0918] Input: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[0919] Data processing: Convert the audio data into text data using the Google Cloud Speech-to-Text API.

[0920] Output: Converted text data (e.g., "Today, I'll talk about the progress of the project.")

[0921] Step 4:

[0922] The server sends the text data to the generative AI, which generates feedback.

[0923] Input: Converted text data (e.g., "Today I'll talk about the progress of the project.")

[0924] Data processing: We use OpenAI's GPT-4 to analyze the text data and generate feedback, using the following as an example prompt:

[0925] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[0926] text:

[0927] Today I'd like to talk about the progress of the project. It's currently at 50%...

[0928] Output: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[0929] Step 5:

[0930] The server notifies the user and the content delivery service of the generated feedback.

[0931] Input: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[0932] Data processing: The feedback is formatted into an easy-to-understand format and sent via email or in-system message.

[0933] Output: Feedback communicated to the user and content delivery service

[0934] 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.

[0935] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI) and an emotion engine. This system consists of a server, a terminal, and a user.

[0936] Uploading and saving videos

[0937] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[0938] Video Analysis

[0939] The server extracts the audio track from the saved video file. Using a video processing library (e.g., FFmpeg), the server separates only the audio from the video file and saves the audio data as a temporary audio file, for example, named "presentation_audio.mp3."

[0940] Speech to text

[0941] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, a specific speech recognition API can be used to analyze the audio data and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[0942] Emotion recognition by emotion engine

[0943] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial expression recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[0944] Feedback generation using generative AI

[0945] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback including good points and areas for improvement. Furthermore, the AI ​​adjusts the content and tone of the feedback based on the recognized emotions.

[0946] Notification of results

[0947] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement. The feedback also includes advice based on the user's emotional state.

[0948] Specific examples

[0949] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0950] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[0951] In this way, the system provides users with specific feedback to improve the overall quality of their presentations. By receiving objective and consistent evaluations, users are provided with specific guidelines to help improve their presentations.

[0952] The processing flow will be explained below.

[0953] Step 1:

[0954] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[0955] Step 2:

[0956] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[0957] Step 3:

[0958] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[0959] Step 4:

[0960] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[0961] Step 5:

[0962] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[0963] Step 6:

[0964] The server sends the acquired text data and emotional data to a generative AI. Based on the input data, the generative AI analyzes the presentation's structure, content consistency, voice tone, speed, and visual content used, and generates specific feedback including good points and areas for improvement. It also adjusts the content and tone of the feedback based on the recognized emotions.

[0965] Step 7:

[0966] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[0967] Specific examples

[0968] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[0969] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[0970] Example 2

[0971] 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."

[0972] In today's world, evaluating presentations and identifying areas for improvement have become important challenges. However, conventional methods have struggled to provide comprehensive feedback that includes not only the content of the presentation but also the speaker's emotional state. This has made it difficult for users to obtain specific guidelines for improving the overall quality of their presentations.

[0973] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, means for sending the audio track to an emotion recognition engine and analyzing the emotional state, generative artificial intelligence means for analyzing the text data and the emotional data and generating feedback based on preset criteria, and means for notifying the user of the generated feedback. This allows the user to receive comprehensive and specific feedback based on the content of their presentation and their emotional state.

[0974] "User" refers to a person who accesses the system to upload and evaluate presentation videos.

[0975] "Video File" means a video and audio file containing presentation content that a User uploads to the System.

[0976] "Server" refers to a computer device that has a central function for processing and managing data for the entire system.

[0977] "Means for receiving" refers to software or a network interface that allows the server to receive video files from users.

[0978] "Means for extracting the audio track" refers to the video processing software or libraries used by the server to extract only the audio portion of the video file.

[0979] "Audio track" refers only to the audio data contained in a video file.

[0980] "Means for converting to text data" refers to the processes and techniques that convert an audio track into written information using a speech recognition service.

[0981] An "emotion recognition engine" refers to software or algorithms that analyze and identify a speaker's emotional state from audio or video.

[0982] "Generative AI means" refers to artificial intelligence techniques and algorithms for automatically generating feedback based on text data and emotional data.

[0983] "Generating feedback" refers to creating information including an evaluation of the presentation and suggestions for improvement based on the analysis results.

[0984] "Means of notification" refers to email or in-system messaging functions that provide the generated feedback to the user in an easy-to-understand format.

[0985] "Visual content" refers to the visual elements of a presentation, such as charts, images, and slides.

[0986] The present invention provides a system that allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence and an emotion recognition engine. Specific embodiments of this system are described below.

[0987] Uploading and saving videos

[0988] A user accesses the system using a terminal and uploads a presentation video. To do this, the user selects a file, for example, "project_update.mp4." The server receives the video file and saves it in the appropriate directory. The server records the file path in a database so that the file can be accessed in subsequent processing steps.

[0989] Video Analysis

[0990] The server uses a video processing library such as FFmpeg to extract the audio track from the saved video file. Specifically, it executes an FFmpeg command to extract only the audio portion from the video file and saves it as a temporary file named "project_update_audio.mp3."

[0991] Speech to text

[0992] The server sends the "project_update_audio.mp3" file to a speech recognition service (e.g., Google Speech-to-Text API) and analyzes the content as text data. This process converts the audio data into text information, generating text data such as "Today, I'll be talking about the progress of the project."

[0993] Emotion recognition analysis

[0994] The server sends the "project_update_audio.mp3" file to the emotion recognition engine to analyze the speaker's emotional state. The emotion recognition engine identifies the user's emotional state (e.g., nervous, calm, excited, etc.) based on the tone, pitch, and rhythm of the voice. Optionally, facial expression recognition technology can also be used to analyze facial expressions from the video for more accurate emotion recognition.

[0995] Feedback generation using generative AI

[0996] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, voice tone, speed, visual content, and other factors to generate feedback. This feedback includes both good points and areas for improvement. The content and tone of the feedback are also adjusted based on the recognized emotions.

[0997] Notification of results

[0998] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal system message). This allows users to receive specific feedback based on the content and emotional state of their presentation, allowing them to objectively understand what needs to be improved.

[0999] Specific examples

[1000] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." Next, using this audio file, a speech recognition service generates text data such as "Today, I'll talk about the progress of the project." The generative AI analyzes this text data and the emotion data, such as "I'm nervous." As a result, it generates feedback such as "The flow of the presentation is easy to understand, and the tone is appropriate, but you speak too fast." It also includes advice, taking into account the user's nervousness, such as "Practice speaking calmly." This feedback is then sent to the user via email.

[1001] Example prompt sentence:

[1002] Presentation content: Today I will talk about the progress of the project.

[1003] Emotional data: Tension

[1004] Additional information for analysis: None

[1005] In this way, the system can provide users with specific feedback that helps them improve the quality of their presentations, providing them with objective and consistent evaluations and specific guidelines for improving their presentations.

[1006] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1007] Step 1: User uploads a video

[1008] A user accesses the system using their own terminal and uploads a presentation video. For example, they select a file called "project_update.mp4" and upload it.

[1009] Input: Presentation video file (e.g., project_update.mp4)

[1010] Output: The video file is sent to the server.

[1011] Step 2: Storing videos on the server

[1012] The server saves the video file (e.g., project_update.mp4) received from the user in the appropriate directory. The save path is recorded in the database.

[1013] Input: Video file received from user

[1014] Output: Path to the saved video file (e.g. / uploads / project_update.mp4)

[1015] Step 3: Extracting the audio track by the server

[1016] The server uses a video processing library such as FFmpeg to extract the audio track from the video file.

[1017] Specifically, run the FFmpeg command to extract the audio portion from the video file (e.g., / uploads / project_update.mp4) and save it as "project_update_audio.mp3".

[1018] Input: Path to the saved video file

[1019] Output: Extracted audio file (e.g. project_update_audio.mp3)

[1020] Step 4: The server converts the audio file to text.

[1021] The server sends the extracted audio file (e.g., project_update_audio.mp3) to a speech recognition service (e.g., Google Speech-to-Text API) and converts the content into text data.

[1022] Specifically, the voice data is passed to a voice recognition API, and the text data obtained as the API response (e.g., "Today I will talk about the progress of the project") is processed.

[1023] Input: Audio file (e.g., project_update_audio.mp3)

[1024] Output: Text data (e.g., "Today, I'll talk about the progress of the project.")

[1025] Step 5: Emotion recognition analysis by the server

[1026] The server sends the audio file (e.g., project_update_audio.mp3) to the emotion recognition engine to analyze the user's emotional state.

[1027] Specifically, the emotion recognition engine analyzes the tone, pitch, and rhythm of the voice to identify emotions such as tension, calmness, excitement, etc. It also performs facial expression recognition on video data as needed.

[1028] Input: Audio file (e.g., project_update_audio.mp3)

[1029] Output: Emotion data (e.g., tension)

[1030] Step 6: Server sends data to generative AI

[1031] The server sends the acquired text data and emotion data to the generative AI.

[1032] The prompt text should include:

[1033] Presentation content: Today I will talk about the progress of the project.

[1034] Emotional data: Tension

[1035] Additional information for analysis: None

[1036] Input: Text data (e.g., "Today I'll talk about the progress of the project"), emotion data (e.g., nervousness)

[1037] Output: Input data formatted as a prompt

[1038] Step 7: Generative AI feedback generation

[1039] Generative AI uses input data to analyze the presentation's structure, content consistency, tone of voice, speed, visual content, etc., and generates feedback.

[1040] Specifically, the system generates feedback such as, "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly," and offers advice such as, "Practice speaking calmly to reduce nervousness."

[1041] Input: Input data formatted as a prompt

[1042] Output: Generated feedback (e.g., "The flow of your presentation was clear and your tone was appropriate, but you spoke too quickly.")

[1043] Step 8: Server notifies user of results

[1044] The generated feedback is formatted by the server and notified to the user.

[1045] Feedback is provided in an easy-to-understand format (e.g., email or in-system message) so users can review it and understand areas for improvement.

[1046] Input: Generated feedback

[1047] Output: Feedback sent to the user (e.g. email)

[1048] (Application example 2)

[1049] 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."

[1050] Real-time feedback is necessary to improve the quality of training and education in factories. However, this is difficult to achieve with conventional training methods, making it difficult to provide specific guidance for improvement to individual employees. Furthermore, since feedback based on emotional changes is not provided, there is a possibility that employee motivation and technical improvement may be delayed.

[1051] 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.

[1052] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and the user's emotional state and generating feedback based on preset criteria, and means for notifying the user of the generated feedback, thereby enabling real-time, specific, and emotion-based feedback for in-factory training and education.

[1053] "User" means a person who uses the system to upload video files and receive feedback.

[1054] A "video file" is a digital file that contains audio and video data.

[1055] An "audio track" is a portion of the audio data contained within a video file, which is separate from the video data.

[1056] "Text data" refers to data that has been converted from an audio track into text information.

[1057] "Emotional state" refers to the type and intensity of emotions a user displays during a presentation, and is analyzed based on their voice and facial expressions.

[1058] "Generative AI" refers to an AI technology that analyzes input data and generates feedback.

[1059] A "video processing library" is a software tool or library used to extract audio tracks from video files.

[1060] "Feedback" refers to advice and information about improvements and positive actions generated by the system and communicated to the user.

[1061] This invention aims to improve the quality of training and education within a factory through a system installed in a factory robot. In order to implement this invention, the following hardware and software are used.

[1062] Hardware and software used

[1063] Server: High performance server

[1064] Device: Device equipped with a voice recognition microphone and camera

[1065] Video processing library: FFmpeg

[1066] Speech Recognition Software: Google's SpeechRecognition API

[1067] Generative AI software: Transformers (Hugging Face)

[1068] Emotion Recognition Software: EmotionRecognizer

[1069] System configuration

[1070] 1. Upload your video file:

[1071] Users use the terminal to upload training videos to factory robots.

[1072] The server receives this video file and stores it in the appropriate directory.

[1073] 2. Extract audio tracks:

[1074] The server extracts and separates the audio track from the stored video file using emotion recognition software, using the FFmpeg library.

[1075] 3. Transcribe the audio track:

[1076] The server sends the separated audio files to the speech recognition API and receives the content as text data.

[1077] 4. Emotional state analysis:

[1078] The server sends the voice data to emotion recognition software, which analyzes the tone, pitch, and rhythm to identify the user's emotional state.

[1079] 5. Feedback Generation:

[1080] The server sends the captured text and emotion data to the generative AI, which then generates feedback, taking into account factors such as structure, coherence of content, tone of voice, speed, and the user's emotional state.

[1081] 6. Feedback Notification:

[1082] The generated feedback is notified to the user by the server, and is provided in a user-friendly format and sent via email or internal system messages.

[1083] Specific examples

[1084] For example, a factory robot might give a presentation on "safety procedures." The user then uploads the video of the presentation to the system. The server saves the video file, extracts the audio track, and converts it into text. This text data is then sent to a generative AI system, which analyzes it and provides feedback such as, "The procedure explanation is clear, but you explain it too quickly." If emotion-recognition software detects through voice analysis that the user is nervous, the feedback might include specific advice such as, "Practice speaking more calmly."

[1085] Prompt Sentence Examples

[1086] In actual generative AI analysis, the following prompts are used:

[1087] Analyze a presentation on safety procedures within a factory, which includes the following:

[1088] ...

[1089] The user's emotional state is as follows:

[1090] ...

[1091] Use this information to generate feedback on how to improve your presentation.

[1092] In this way, users can obtain specific improvements and advice for training and educational presentations, which can be used to improve technology within their factories.

[1093] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1094] Step 1:

[1095] A user uploads a training video file to the system using a terminal, which then transmits the video file to the server.

[1096] Input: A video file uploaded by the user.

[1097] Data processing: Receiving and storing video files.

[1098] Output: Uploaded video file saved on server.

[1099] Step 2:

[1100] The server uses a video processing library (FFmpeg) to extract the audio track from the received video file.

[1101] Input: Video files stored on the server.

[1102] Data processing: Separating audio tracks from video files and generating audio files.

[1103] Output: Audio file saved on the server.

[1104] Step 3:

[1105] The server sends the extracted audio file to a speech recognition API (Google SpeechRecognition) and converts the audio into text data.

[1106] Input: Audio files stored on the server.

[1107] Data processing: Transcription of audio data.

[1108] Output: Text data extracted from the audio.

[1109] Step 4:

[1110] The server sends the voice data to emotion recognition software (EmotionRecognizer) to analyze the user's emotional state.

[1111] Input: Audio files stored on the server.

[1112] Data processing: Analyzing the tone, pitch, and rhythm of speech to identify emotional states.

[1113] Output: User's emotional state data.

[1114] Step 5:

[1115] The server sends the acquired text data and emotion data to generative artificial intelligence (Transformers), which generates feedback based on the data.

[1116] Input: Text data and emotional state data.

[1117] Data processing: analyzing text and sentiment data, generating feedback.

[1118] Output: Feedback data.

[1119] Step 6:

[1120] The server notifies the user of the generated feedback, which is sent via email or an internal message.

[1121] Input: Feedback data.

[1122] Data processing: Formatting feedback and preparing notifications.

[1123] Output: Feedback sent to the user.

[1124] 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.

[1125] 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.

[1126] 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.

[1127] [Fourth embodiment]

[1128] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1129] 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.

[1130] 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).

[1131] 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.

[1132] 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.

[1133] 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).

[1134] 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.

[1135] 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.

[1136] 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.

[1137] 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.

[1138] 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.

[1139] 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.

[1140] 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."

[1141] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI). This system consists of a server, a terminal, and a user.

[1142] Uploading and saving videos

[1143] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[1144] Video Analysis

[1145] The server extracts the audio track from the stored video file, and using a video processing library (e.g. FFmpeg) separates the audio from the video file and saves it as a temporary audio file for further processing.

[1146] Speech to text

[1147] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, it can analyze the audio data using a specific API (e.g., a speech recognition API) and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[1148] Feedback generation using generative AI

[1149] The server sends the acquired text data to a generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and then generates specific feedback including good points and areas for improvement.

[1150] Notification of results

[1151] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement.

[1152] Specific examples

[1153] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of the project." This text is sent to a generative AI, which analyzes it and generates feedback such as "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly." This feedback is then sent to the user via email, allowing them to learn specific improvements.

[1154] In this way, the system provides users with concrete feedback to improve the quality of their presentations. By receiving objective and consistent evaluations, users are provided with concrete guidelines to help improve their presentations.

[1155] The processing flow will be explained below.

[1156] Step 1:

[1157] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[1158] Step 2:

[1159] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[1160] Step 3:

[1161] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[1162] Step 4:

[1163] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[1164] Step 5:

[1165] The server sends the text data to the generative AI, which then creates and sends a request to the generative AI (e.g., GPT-4) to analyze the presentation text.

[1166] Step 6:

[1167] Generative AI analyzes text data and generates feedback. Based on the user's presentation text, the generative AI evaluates the presentation's structure, content consistency, tone of voice, speed, use of visual content, and other aspects, and generates feedback including good points and areas for improvement.

[1168] Step 7:

[1169] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[1170] Example 1

[1171] 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."

[1172] In order to obtain evaluations and feedback on presentation videos, third-party evaluations or self-feedback are often required. However, these methods are subjective and have problems with maintaining consistency in evaluation criteria. Furthermore, it is difficult to efficiently identify specific areas for improvement, and there is a lack of guidelines for quickly improving the quality of presentations.

[1173] 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.

[1174] In this invention, the server includes means for receiving a video file from a user, means for saving the received video file in a predetermined directory, means for extracting an audio track from the received video file, means for saving the audio track as a temporary audio file, means for transmitting the temporary audio file to a speech recognition service and converting the speech into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, and means for formatting the generated feedback and notifying the user. This allows users to easily obtain objective and consistent feedback, quickly identify specific areas for improvement in their presentations, and improve the quality of their presentations.

[1175] A "user" is a person or organization that accesses the system to upload presentation videos and receive feedback.

[1176] "Video file" refers to a digital file containing presentation content uploaded by a user, with an example file extension of .mp4.

[1177] The "server" is the central part of the system, and is a collection of hardware and software that performs various processes such as receiving, storing, analyzing, generating feedback, and notifying users of video files.

[1178] A "directory" refers to a folder on a file system that is used to organize and store data within a server.

[1179] "Audio Track" refers to the audio data extracted from a video file that contains the audio content of the presentation.

[1180] "Temporary audio file" refers to an audio data file that is generated to temporarily store an audio track and is used for subsequent speech recognition processing.

[1181] "Speech Recognition Service" refers to a cloud-based or local service or API for analyzing and converting voice data into text.

[1182] "Text data" refers to the text information generated by the speech recognition service, and represents the content of the presentation in text form.

[1183] "Generative AI means" refers to an AI model that has an algorithm that generates and analyzes information based on input data, and is responsible for generating feedback for presentations.

[1184] "Feedback" refers to information provided by generative artificial intelligence means that includes an evaluation of the presentation and specific areas for improvement.

[1185] "Notification means" refers to a method or system for communicating generated feedback to the user, including email, dashboard display, etc.

[1186] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides feedback. This system is composed of a server, a terminal, and a user. Specific embodiments are described in detail below.

[1187] Uploading and saving videos

[1188] Users upload presentation videos to the system from their devices using a browser or a dedicated application. Video files generally have the extension .mp4. The video file selected by the user is sent to the server with a name such as "presentation.mp4." After receiving this video file, the server saves it in a specified directory (e.g., " / videos / "). This saving process makes it possible to manage the video file path in a database.

[1189] Video Analysis

[1190] The server extracts the audio track from the received video file. Specifically, it uses a video processing library called FFmpeg to execute a command to extract only the audio data from the video file. For example, it uses the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary audio file (e.g., " / temp / presentation_audio.mp3").

[1191] Speech to text

[1192] The server sends the extracted audio file to a speech recognition service, which converts it into text. This can be done using a cloud-based or local speech recognition service, such as Google's Speech-to-Text API. For example, " / temp / presentation_audio.mp3" is uploaded, and the speech recognition service returns text data such as "Today, I'll talk about the progress of the project." The server receives this text data and proceeds to the next step.

[1193] Feedback generation using generative AI

[1194] The server sends the acquired text data to a generative artificial intelligence (e.g., OpenAI's GPT-3). An example of a specific prompt sentence is shown below.

[1195] Example prompt sentence:

[1196] Please provide feedback on the structure, tone, pace, and consistency of the content of the following presentation text:

[1197] Today I'll be talking about the progress of the project.

[1198] Based on this text data, generative AI analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback such as, "The flow of the presentation is easy to understand and the tone is appropriate, but you speak too quickly."

[1199] Notification of results

[1200] The generated feedback is formatted by the server and notified to the user. The feedback is provided in an easy-to-understand format (e.g., HTML email or internal message). The user can receive this feedback and get specific guidance on how to improve the quality of their presentation.

[1201] Specific examples

[1202] For example, a case will be described where a user uploads a presentation video called "project_update.mp4."

[1203] 1. The user uses the terminal to upload the "project_update.mp4" file to the system.

[1204] 2. The server saves the video as " / videos / project_update.mp4".

[1205] 3. The server uses the video processing library FFmpeg to extract the audio track from "project_update.mp4" and save it as " / temp / project_update_audio.mp3".

[1206] 4. The server sends " / temp / project_update_audio.mp3" to the speech recognition API and receives the text data "Today I will talk about the progress of the project."

[1207] 5. The server sends this text data to a generative AI, which generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[1208] 6. The server formats this feedback and notifies the user via email.

[1209] In this way, the system provides the user with specific feedback to improve the quality of their presentation.

[1210] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1211] Program processing steps

[1212] Step 1:

[1213] User uploads video files

[1214] Input: Presentation video file (e.g. "presentation.mp4")

[1215] Process: A user logs in to the system using a browser or a dedicated application, selects a presentation video from the file upload interface, and then clicks the upload button to send the video to the server.

[1216] Output: The video file is received on the server.

[1217] Step 2:

[1218] Video file storage on the server

[1219] Input: The received video file (e.g., "presentation.mp4")

[1220] Processing: The server saves the received file in a specific directory (e.g., " / videos / "), recording the file metadata and save path in a database.

[1221] Output: The saved video file (e.g. " / videos / presentation.mp4") and its path information.

[1222] Step 3:

[1223] Extraction of audio tracks from video files by the server

[1224] Input: Saved video file (e.g. " / videos / presentation.mp4")

[1225] Processing: The server extracts the audio track using FFmpeg, using the command ffmpeg -i / videos / presentation.mp4 -q:a 0 -map a / temp / presentation_audio.mp3 to save the audio track as a temporary file.

[1226] Output: The extracted audio file (e.g. " / temp / presentation_audio.mp3").

[1227] Step 4:

[1228] Conversion of voice data into text by the server

[1229] Input: Extracted audio file (e.g. " / temp / presentation_audio.mp3")

[1230] Processing: The server sends the audio file to the speech recognition service (e.g., by sending an HTTP POST request). The speech recognition service analyzes the audio data and generates corresponding text data. The server receives this text data.

[1231] Output: Text data converted from the speech data (e.g., "Today I will talk about the progress of the project").

[1232] Step 5:

[1233] The server sends text data to the AI ​​generator and generates feedback

[1234] Input: Text data generated from speech data (e.g., "Today, I'll talk about the progress of the project.")

[1235] Processing: The server sends the text data to a generative AI model (e.g., GPT-3). It uses prompts to instruct the AI ​​on how to analyze it. For example, the prompt might look like this: "Please provide feedback on the structure, tone, speed, and consistency of the following presentation text: 'Today I'll be talking about the progress of my project.'"

[1236] The generative AI model analyzes the input text data and generates feedback.

[1237] Output: Feedback generated by the generative AI model (e.g., "The flow of your presentation was clear and the tone was appropriate, but you speak too quickly").

[1238] Step 6:

[1239] Server-generated feedback and notification to the user

[1240] Input: Feedback from the generative AI model (e.g., "The flow of the presentation was clear and the tone was appropriate, but you spoke too quickly.")

[1241] Processing: The server formats the feedback and converts it into an appropriate format (e.g. HTML email, JSON for dashboard display, etc.), and then provides the formatted feedback to the user via email or system notification.

[1242] Output: Feedback given to the user.

[1243] In this way, the system provides the user with concrete and objective suggestions for improving their presentation.

[1244] (Application example 1)

[1245] 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."

[1246] Conventional feedback systems for presentation videos provide feedback only to a single user, and have issues with efficiency and scalability when widely used on content distribution platforms. Furthermore, the system does not adequately provide feedback tailored to the type and purpose of the content provided by the user. In particular, it lacks adaptability as a feedback system for creators in content distribution services.

[1247] 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.

[1248] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and generating feedback based on preset criteria, means for notifying the user of the generated feedback, and means for providing the generated feedback to a content distribution service, thereby enabling content creators to receive specific and effective feedback to continuously improve the quality of their content.

[1249] "User" means any person or entity that uploads video files to the System and receives generated Feedback.

[1250] "Video File" means a media file in digital format containing audio and visual content uploaded by a User.

[1251] An "audio track" is the audio data contained within a video file.

[1252] "Text data" refers to text information that has been analyzed and converted from an audio track.

[1253] "Generative AI" refers to an AI technology that analyzes content and generates feedback based on pre-set criteria using text data.

[1254] "Feedback" refers to evaluations and improvement suggestions for users' content generated by generative artificial intelligence.

[1255] "Notification means" refers to the method or technique by which generated feedback is communicated to the user.

[1256] "Content distribution service" means an online platform for publishing and distributing content uploaded by users to an audience.

[1257] This invention is a system that allows users to upload presentation videos, analyzes the content using generative artificial intelligence (AI), and provides specific feedback. Detailed embodiments of this system are described below.

[1258] Uploading and saving videos

[1259] The user uploads their video file to the server. The server manages the file name and path of the video file to save it in the appropriate directory. This step is performed using Python and OS libraries operated by the server.

[1260] Extracting Audio Tracks

[1261] The server extracts the audio track from the uploaded video file by using a video processing library such as FFmpeg or moviepy to separate the audio from the video file and save it as an audio file.

[1262] Speech to text

[1263] The server sends the extracted audio file to a speech recognition service that converts the audio data into text data, for example, by using the Google Cloud Speech-to-Text API to extract the contents of the audio file as text information.

[1264] Feedback generation using generative AI

[1265] The server sends the acquired text data to a generative AI model, such as OpenAI's GPT-4, which analyzes the input text data and generates feedback based on the presentation's structure, content coherence, tone of voice, speed, and use of visual content. A specific example of a prompt might look like this:

[1266] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[1267] text:

[1268] Today I'd like to talk about the progress of the project. It's currently at 50%...

[1269] Notification of results

[1270] The generated feedback is formatted into an easy-to-understand format by the server and notified to the user or the content distribution service via email or a message within the system.

[1271] Collaboration with content distribution services

[1272] Furthermore, this system provides the generated feedback to the content distribution service, allowing users (creators) to receive specific guidance on how to improve the quality of their content. This collaboration allows creators to more objectively know how their content is being evaluated and what areas need improvement.

[1273] This system allows creators to efficiently improve their content and increase its value to viewers.

[1274] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1275] Step 1:

[1276] A user uploads a video file to a server using a terminal.

[1277] Input: User's video file (e.g. presentation.mp4)

[1278] Data processing: The video file is sent to the server, and the server saves the video file in the appropriate directory.

[1279] Output: Path to the video file saved on the server (e.g. / videos / presentation.mp4)

[1280] Step 2:

[1281] The server extracts the audio track from the received video file.

[1282] Input: Path to the video file stored on the server (e.g. / videos / presentation.mp4)

[1283] Data processing: Use FFmpeg or moviepy to separate the audio track from the video and generate a temporary audio file.

[1284] Output: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[1285] Step 3:

[1286] The server sends the extracted audio file to a speech recognition service, which converts the audio data into text data.

[1287] Input: Path to the audio file saved on the server (e.g., / audios / presentation_audio.wav)

[1288] Data processing: Convert the audio data into text data using the Google Cloud Speech-to-Text API.

[1289] Output: Converted text data (e.g., "Today, I'll talk about the progress of the project.")

[1290] Step 4:

[1291] The server sends the text data to the generative AI, which generates feedback.

[1292] Input: Converted text data (e.g., "Today I'll talk about the progress of the project.")

[1293] Data processing: We use OpenAI's GPT-4 to analyze the text data and generate feedback, using the following as an example prompt:

[1294] Analyze the presentation text below and provide feedback on coherence, tone, pace, and use of visuals.

[1295] text:

[1296] Today I'd like to talk about the progress of the project. It's currently at 50%...

[1297] Output: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[1298] Step 5:

[1299] The server notifies the user and the content delivery service of the generated feedback.

[1300] Input: Generated feedback (e.g., "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly.")

[1301] Data processing: The feedback is formatted into an easy-to-understand format and sent via email or in-system message.

[1302] Output: Feedback communicated to the user and content delivery service

[1303] 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.

[1304] This system allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence (AI) and an emotion engine. This system consists of a server, a terminal, and a user.

[1305] Uploading and saving videos

[1306] A user accesses the system using a terminal and uploads a presentation video. The server receives the video file and saves it in the appropriate directory. For example, if a user uploads a video file called "presentation.mp4", the server saves it and manages its path.

[1307] Video Analysis

[1308] The server extracts the audio track from the saved video file. Using a video processing library (e.g., FFmpeg), the server separates only the audio from the video file and saves the audio data as a temporary audio file, for example, named "presentation_audio.mp3."

[1309] Speech to text

[1310] The server sends the extracted audio file to a speech recognition service and receives the content as text data. For example, a specific speech recognition API can be used to analyze the audio data and extract the presentation content as text information. This text data becomes the basic input data for analysis by generative AI.

[1311] Emotion recognition by emotion engine

[1312] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial expression recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[1313] Feedback generation using generative AI

[1314] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, tone of voice, speed, and visual content used, and generates specific feedback including good points and areas for improvement. Furthermore, the AI ​​adjusts the content and tone of the feedback based on the recognized emotions.

[1315] Notification of results

[1316] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal message), allowing the user to objectively understand which parts of their presentation went well and which parts need improvement. The feedback also includes advice based on the user's emotional state.

[1317] Specific examples

[1318] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[1319] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[1320] In this way, the system provides users with specific feedback to improve the overall quality of their presentations. By receiving objective and consistent evaluations, users are provided with specific guidelines to help improve their presentations.

[1321] The processing flow will be explained below.

[1322] Step 1:

[1323] A user uploads a presentation video file to the system using a terminal. The user selects the video file through a web browser and clicks the send button to send a request to upload the video to the server.

[1324] Step 2:

[1325] The server receives the uploaded video file. The server parses the HTTP request and saves the video file in the specified directory. For example, if the file name is "presentation.mp4," the server saves the file in / path / to / save / presentation.mp4.

[1326] Step 3:

[1327] The server extracts the audio track from the saved video file. The server uses a video processing library (e.g., FFmpeg) to separate only the audio from the video file and saves the audio data as a temporary audio file, for example, "presentation_audio.mp3."

[1328] Step 4:

[1329] The server converts the extracted audio file into text format. The server then sends the audio file to a speech recognition service (e.g., a speech recognition API) and receives text data in response. This speech recognition process provides the presentation content as text information.

[1330] Step 5:

[1331] The server sends the voice data to the emotion engine, which analyzes the user's emotional state. The emotion engine analyzes the tone, pitch, and rhythm of the voice to identify the user's emotional state (e.g., excitement, calm, tension, etc.). Furthermore, by using a camera equipped with facial recognition technology, it is possible to analyze the user's facial expressions from the video and recognize emotions more accurately.

[1332] Step 6:

[1333] The server sends the acquired text data and emotional data to a generative AI. Based on the input data, the generative AI analyzes the presentation's structure, content consistency, voice tone, speed, and visual content used, and generates specific feedback including good points and areas for improvement. It also adjusts the content and tone of the feedback based on the recognized emotions.

[1334] Step 7:

[1335] The server formats the generated feedback and notifies the user. The feedback is formatted into a report format that is easy for the user to understand and sent as an email or an internal message. The user can receive this feedback through their device and use it to improve their presentation.

[1336] Specific examples

[1337] For example, say a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." A speech recognition service converts this audio file into text such as "Today, I'll talk about the progress of this project." This text is then sent to a generative AI, which analyzes it and generates feedback such as, "The flow of the presentation is clear and the tone is appropriate, but you speak too quickly."

[1338] In addition, if the emotion engine recognizes through voice analysis that the user is nervous during a presentation, the feedback will include specific advice such as "Practice speaking calmly." This feedback will be sent to the user via email, allowing them to identify specific areas for improvement.

[1339] Example 2

[1340] 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."

[1341] In today's world, evaluating presentations and identifying areas for improvement have become important challenges. However, conventional methods have struggled to provide comprehensive feedback that includes not only the content of the presentation but also the speaker's emotional state. This has made it difficult for users to obtain specific guidelines for improving the overall quality of their presentations.

[1342] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, means for sending the audio track to an emotion recognition engine and analyzing the emotional state, generative artificial intelligence means for analyzing the text data and the emotional data and generating feedback based on preset criteria, and means for notifying the user of the generated feedback. This allows the user to receive comprehensive and specific feedback based on the content of their presentation and their emotional state.

[1343] "User" refers to a person who accesses the system to upload and evaluate presentation videos.

[1344] "Video File" means a video and audio file containing presentation content that a User uploads to the System.

[1345] "Server" refers to a computer device that has a central function for processing and managing data for the entire system.

[1346] "Means for receiving" refers to software or a network interface that allows the server to receive video files from users.

[1347] "Means for extracting the audio track" refers to the video processing software or libraries used by the server to extract only the audio portion of the video file.

[1348] "Audio track" refers only to the audio data contained in a video file.

[1349] "Means for converting to text data" refers to the processes and techniques that convert an audio track into written information using a speech recognition service.

[1350] An "emotion recognition engine" refers to software or algorithms that analyze and identify a speaker's emotional state from audio or video.

[1351] "Generative AI means" refers to artificial intelligence techniques and algorithms for automatically generating feedback based on text data and emotional data.

[1352] "Generating feedback" refers to creating information including an evaluation of the presentation and suggestions for improvement based on the analysis results.

[1353] "Means of notification" refers to email or in-system messaging functions that provide the generated feedback to the user in an easy-to-understand format.

[1354] "Visual content" refers to the visual elements of a presentation, such as charts, images, and slides.

[1355] The present invention provides a system that allows users to upload presentation videos and provides feedback by analyzing the content using generative artificial intelligence and an emotion recognition engine. Specific embodiments of this system are described below.

[1356] Uploading and saving videos

[1357] A user accesses the system using a terminal and uploads a presentation video. To do this, the user selects a file, for example, "project_update.mp4." The server receives the video file and saves it in the appropriate directory. The server records the file path in a database so that the file can be accessed in subsequent processing steps.

[1358] Video Analysis

[1359] The server uses a video processing library such as FFmpeg to extract the audio track from the saved video file. Specifically, it executes an FFmpeg command to extract only the audio portion from the video file and saves it as a temporary file named "project_update_audio.mp3."

[1360] Speech to text

[1361] The server sends the "project_update_audio.mp3" file to a speech recognition service (e.g., Google Speech-to-Text API) and analyzes the content as text data. This process converts the audio data into text information, generating text data such as "Today, I'll be talking about the progress of the project."

[1362] Emotion recognition analysis

[1363] The server sends the "project_update_audio.mp3" file to the emotion recognition engine to analyze the speaker's emotional state. The emotion recognition engine identifies the user's emotional state (e.g., nervous, calm, excited, etc.) based on the tone, pitch, and rhythm of the voice. Optionally, facial expression recognition technology can also be used to analyze facial expressions from the video for more accurate emotion recognition.

[1364] Feedback generation using generative AI

[1365] The server sends the acquired text data and emotional data to the generative AI, which analyzes the presentation's structure, content consistency, voice tone, speed, visual content, and other factors to generate feedback. This feedback includes both good points and areas for improvement. The content and tone of the feedback are also adjusted based on the recognized emotions.

[1366] Notification of results

[1367] The generated feedback is formatted by the server and notified to the user. The feedback is provided in a user-friendly format (e.g., email or internal system message). This allows users to receive specific feedback based on the content and emotional state of their presentation, allowing them to objectively understand what needs to be improved.

[1368] Specific examples

[1369] For example, suppose a user uploads a presentation video called "project_update.mp4." The server saves this video, extracts the audio track, and saves it as "project_update_audio.mp3." Next, using this audio file, a speech recognition service generates text data such as "Today, I'll talk about the progress of the project." The generative AI analyzes this text data and the emotion data, such as "I'm nervous." As a result, it generates feedback such as "The flow of the presentation is easy to understand, and the tone is appropriate, but you speak too fast." It also includes advice, taking into account the user's nervousness, such as "Practice speaking calmly." This feedback is then sent to the user via email.

[1370] Example prompt sentence:

[1371] Presentation content: Today I will talk about the progress of the project.

[1372] Emotional data: Tension

[1373] Additional information for analysis: None

[1374] In this way, the system can provide users with specific feedback that helps them improve the quality of their presentations, providing them with objective and consistent evaluations and specific guidelines for improving their presentations.

[1375] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1376] Step 1: User uploads a video

[1377] A user accesses the system using their own terminal and uploads a presentation video. For example, they select a file called "project_update.mp4" and upload it.

[1378] Input: Presentation video file (e.g., project_update.mp4)

[1379] Output: The video file is sent to the server.

[1380] Step 2: Storing videos on the server

[1381] The server saves the video file (e.g., project_update.mp4) received from the user in the appropriate directory. The save path is recorded in the database.

[1382] Input: Video file received from user

[1383] Output: Path to the saved video file (e.g. / uploads / project_update.mp4)

[1384] Step 3: Extracting the audio track by the server

[1385] The server uses a video processing library such as FFmpeg to extract the audio track from the video file.

[1386] Specifically, run the FFmpeg command to extract the audio portion from the video file (e.g., / uploads / project_update.mp4) and save it as "project_update_audio.mp3".

[1387] Input: Path to the saved video file

[1388] Output: Extracted audio file (e.g. project_update_audio.mp3)

[1389] Step 4: The server converts the audio file to text.

[1390] The server sends the extracted audio file (e.g., project_update_audio.mp3) to a speech recognition service (e.g., Google Speech-to-Text API) and converts the content into text data.

[1391] Specifically, the voice data is passed to a voice recognition API, and the text data obtained as the API response (e.g., "Today I will talk about the progress of the project") is processed.

[1392] Input: Audio file (e.g., project_update_audio.mp3)

[1393] Output: Text data (e.g., "Today, I'll talk about the progress of the project.")

[1394] Step 5: Emotion recognition analysis by the server

[1395] The server sends the audio file (e.g., project_update_audio.mp3) to the emotion recognition engine to analyze the user's emotional state.

[1396] Specifically, the emotion recognition engine analyzes the tone, pitch, and rhythm of the voice to identify emotions such as tension, calmness, excitement, etc. It also performs facial expression recognition on video data as needed.

[1397] Input: Audio file (e.g., project_update_audio.mp3)

[1398] Output: Emotion data (e.g., tension)

[1399] Step 6: Server sends data to generative AI

[1400] The server sends the acquired text data and emotion data to the generative AI.

[1401] The prompt text should include:

[1402] Presentation content: Today I will talk about the progress of the project.

[1403] Emotional data: Tension

[1404] Additional information for analysis: None

[1405] Input: Text data (e.g., "Today I'll talk about the progress of the project"), emotion data (e.g., nervousness)

[1406] Output: Input data formatted as a prompt

[1407] Step 7: Generative AI feedback generation

[1408] Generative AI uses input data to analyze the presentation's structure, content consistency, tone of voice, speed, visual content, etc., and generates feedback.

[1409] Specifically, the system generates feedback such as, "The flow of your presentation is clear and the tone is appropriate, but you speak too quickly," and offers advice such as, "Practice speaking calmly to reduce nervousness."

[1410] Input: Input data formatted as a prompt

[1411] Output: Generated feedback (e.g., "The flow of your presentation was clear and your tone was appropriate, but you spoke too quickly.")

[1412] Step 8: Server notifies user of results

[1413] The generated feedback is formatted by the server and notified to the user.

[1414] Feedback is provided in an easy-to-understand format (e.g., email or in-system message) so users can review it and understand areas for improvement.

[1415] Input: Generated feedback

[1416] Output: Feedback sent to the user (e.g. email)

[1417] (Application example 2)

[1418] 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."

[1419] Real-time feedback is necessary to improve the quality of training and education in factories. However, this is difficult to achieve with conventional training methods, making it difficult to provide specific guidance for improvement to individual employees. Furthermore, since feedback based on emotional changes is not provided, there is a possibility that employee motivation and technical improvement may be delayed.

[1420] 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.

[1421] In this invention, the server includes means for receiving a video file from a user, means for extracting an audio track from the received video file, means for converting the audio track into text data, generative artificial intelligence means for analyzing the text data and the user's emotional state and generating feedback based on preset criteria, and means for notifying the user of the generated feedback, thereby enabling real-time, specific, and emotion-based feedback for in-factory training and education.

[1422] "User" means a person who uses the system to upload video files and receive feedback.

[1423] A "video file" is a digital file that contains audio and video data.

[1424] An "audio track" is a portion of the audio data contained within a video file, which is separate from the video data.

[1425] "Text data" refers to data that has been converted from an audio track into text information.

[1426] "Emotional state" refers to the type and intensity of emotions a user displays during a presentation, and is analyzed based on their voice and facial expressions.

[1427] "Generative AI" refers to an AI technology that analyzes input data and generates feedback.

[1428] A "video processing library" is a software tool or library used to extract audio tracks from video files.

[1429] "Feedback" refers to advice and information about improvements and positive actions generated by the system and communicated to the user.

[1430] This invention aims to improve the quality of training and education within a factory through a system installed in a factory robot. In order to implement this invention, the following hardware and software are used.

[1431] Hardware and software used

[1432] Server: High performance server

[1433] Device: Device equipped with a voice recognition microphone and camera

[1434] Video processing library: FFmpeg

[1435] Speech Recognition Software: Google's SpeechRecognition API

[1436] Generative AI software: Transformers (Hugging Face)

[1437] Emotion Recognition Software: EmotionRecognizer

[1438] System configuration

[1439] 1. Upload your video file:

[1440] Users use the terminal to upload training videos to factory robots.

[1441] The server receives this video file and stores it in the appropriate directory.

[1442] 2. Extract audio tracks:

[1443] The server extracts and separates the audio track from the stored video file using emotion recognition software, using the FFmpeg library.

[1444] 3. Transcribe the audio track:

[1445] The server sends the separated audio files to the speech recognition API and receives the content as text data.

[1446] 4. Emotional state analysis:

[1447] The server sends the voice data to emotion recognition software, which analyzes the tone, pitch, and rhythm to identify the user's emotional state.

[1448] 5. Feedback Generation:

[1449] The server sends the captured text and emotion data to the generative AI, which then generates feedback, taking into account factors such as structure, coherence of content, tone of voice, speed, and the user's emotional state.

[1450] 6. Feedback Notification:

[1451] The generated feedback is notified to the user by the server, and is provided in a user-friendly format and sent via email or internal system messages.

[1452] Specific examples

[1453] For example, a factory robot might give a presentation on "safety procedures." The user then uploads the video of the presentation to the system. The server saves the video file, extracts the audio track, and converts it into text. This text data is then sent to a generative AI system, which analyzes it and provides feedback such as, "The procedure explanation is clear, but you explain it too quickly." If emotion-recognition software detects through voice analysis that the user is nervous, the feedback might include specific advice such as, "Practice speaking more calmly."

[1454] Prompt Sentence Examples

[1455] In actual generative AI analysis, the following prompts are used:

[1456] Analyze a presentation on safety procedures within a factory, which includes the following:

[1457] ...

[1458] The user's emotional state is as follows:

[1459] ...

[1460] Use this information to generate feedback on how to improve your presentation.

[1461] In this way, users can obtain specific improvements and advice for training and educational presentations, which can be used to improve technology within their factories.

[1462] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1463] Step 1:

[1464] A user uploads a training video file to the system using a terminal, which then transmits the video file to the server.

[1465] Input: A video file uploaded by the user.

[1466] Data processing: Receiving and storing video files.

[1467] Output: Uploaded video file saved on server.

[1468] Step 2:

[1469] The server uses a video processing library (FFmpeg) to extract the audio track from the received video file.

[1470] Input: Video files stored on the server.

[1471] Data processing: Separating audio tracks from video files and generating audio files.

[1472] Output: Audio file saved on the server.

[1473] Step 3:

[1474] The server sends the extracted audio file to a speech recognition API (Google SpeechRecognition) and converts the audio into text data.

[1475] Input: Audio files stored on the server.

[1476] Data processing: Transcription of audio data.

[1477] Output: Text data extracted from the audio.

[1478] Step 4:

[1479] The server sends the voice data to emotion recognition software (EmotionRecognizer) to analyze the user's emotional state.

[1480] Input: Audio files stored on the server.

[1481] Data processing: Analyzing the tone, pitch, and rhythm of speech to identify emotional states.

[1482] Output: User's emotional state data.

[1483] Step 5:

[1484] The server sends the acquired text data and emotion data to generative artificial intelligence (Transformers), which generates feedback based on the data.

[1485] Input: Text data and emotional state data.

[1486] Data processing: analyzing text and sentiment data, generating feedback.

[1487] Output: Feedback data.

[1488] Step 6:

[1489] The server notifies the user of the generated feedback, which is sent via email or an internal message.

[1490] Input: Feedback data.

[1491] Data processing: Formatting feedback and preparing notifications.

[1492] Output: Feedback sent to the user.

[1493] 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.

[1494] 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.

[1495] 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.

[1496] 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.

[1497] 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.

[1498] 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.

[1499] 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).

[1500] 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.

[1501] 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."

[1502] 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.

[1503] 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).

[1504] 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.

[1505] 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.

[1506] 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.

[1507] 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.

[1508] 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.

[1509] 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.

[1510] 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.

[1511] 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.

[1512] 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.

[1513] 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.

[1514] The following is further disclosed regarding the above embodiment.

[1515] (Claim 1)

[1516] means for receiving a video file from a user;

[1517] means for extracting an audio track from the received video file;

[1518] a means for converting the audio track into text data;

[1519] a generative artificial intelligence means for analyzing text data and generating feedback based on preset criteria;

[1520] means for notifying the user of the generated feedback;

[1521] A system including:

[1522] (Claim 2)

[1523] 10. The system of claim 1, wherein said generative artificial intelligence means includes means for analyzing the presentation's structure, content coherence, tone of voice, pace, and visual content.

[1524] (Claim 3)

[1525] 10. The system of claim 1, wherein the means for extracting the audio track includes means for generating the audio file using a video processing library.

[1526] "Example 1"

[1527] (Claim 1)

[1528] means for receiving a video file from a user;

[1529] A means for saving the received video file in a predetermined directory;

[1530] means for extracting an audio track from the received video file;

[1531] a means of saving the audio track as a temporary audio file;

[1532] a means for transmitting the temporary audio file to a speech recognition service and converting the audio into text data;

[1533] a generative artificial intelligence means for analyzing text data and generating feedback based on preset criteria;

[1534] a means for formatting the generated feedback and notifying the user;

[1535] A system including:

[1536] (Claim 2)

[1537] 10. The system of claim 1, wherein said generative artificial intelligence means includes means for analyzing the presentation's structure, content coherence, tone of voice, pace, and visual content.

[1538] (Claim 3)

[1539] 10. The system of claim 1, wherein the means for extracting the audio track includes means for generating the audio file using a video processing library.

[1540] "Application Example 1"

[1541] (Claim 1)

[1542] means for receiving a video file from a user;

[1543] means for extracting an audio track from the received video file;

[1544] a means for converting the audio track into text data;

[1545] a generative artificial intelligence means for analyzing text data and generating feedback based on preset criteria;

[1546] means for notifying the user of the generated feedback;

[1547] means for providing the generated feedback to a content delivery service;

[1548] A system including:

[1549] (Claim 2)

[1550] 10. The system of claim 1, wherein said generative artificial intelligence means includes means for analyzing the presentation for content coherence, tone of voice, speed, and use of visual content.

[1551] (Claim 3)

[1552] 10. The system of claim 1, wherein the means for extracting the audio track includes means for generating the audio file using a video processing library.

[1553] "Example 2: Combining Emotion Engines"

[1554] (Claim 1)

[1555] means for receiving a video file from a user;

[1556] means for extracting an audio track from the received video file;

[1557] a means for converting the audio track into text data;

[1558] means for transmitting the audio track to an emotion recognition engine for analyzing the emotional state;

[1559] a generative artificial intelligence means for analyzing the text data and the emotion data and generating feedback based on a preset criterion;

[1560] means for notifying the user of the generated feedback;

[1561] A system including:

[1562] (Claim 2)

[1563] 10. The system of claim 1, wherein said generative artificial intelligence means includes means for analyzing the presentation's structure, content coherence, tone of voice, pace, and visual content.

[1564] (Claim 3)

[1565] 10. The system of claim 1, wherein the means for extracting the audio track includes means for generating an audio file using video processing software.

[1566] "Application example 2 when combining emotion engines"

[1567] (Claim 1)

[1568] means for receiving a video file from a user;

[1569] means for extracting an audio track from the received video file;

[1570] a means for converting the audio track into text data;

[1571] a generative artificial intelligence means for analyzing the text data and the emotional state and generating feedback based on pre-set criteria;

[1572] means for notifying the user of the generated feedback;

[1573] A system including:

[1574] (Claim 2)

[1575] 10. The system of claim 1, wherein said generative artificial intelligence means includes means for analyzing presentation organization, content coherence, tone of voice, pace, visual content, and the user's emotional state.

[1576] (Claim 3)

[1577] 10. The system of claim 1, wherein the means for extracting the audio track includes means for generating the audio file using a video processing library. [Explanation of symbols]

[1578] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a video file from a user; means for extracting an audio track from the received video file; a means for converting the audio track into text data; a generative artificial intelligence means for analyzing text data and generating feedback based on preset criteria; means for notifying the user of the generated feedback; A system including:

2. 10. The system of claim 1, wherein said generative artificial intelligence means includes means for analyzing the presentation's structure, content coherence, tone of voice, pace, and visual content.

3. 10. The system of claim 1, wherein the means for extracting an audio track includes means for generating an audio file using a video processing library.

Images