System

Abstract

A system is provided.SOLUTION: A system includes a means for recording audio data of a presentation, a means for acquiring a slide image of the presentation, a means for converting the audio data into text data, a means for analyzing the slide image and extracting a main text, a means for integrating the text data and the extracted main text and analyzing a presentation content, a means for generating a question based on the presentation content, a means for presenting the generated question to a terminal of an audience, a means for the audience to post a question, a means for voting on the question presented to the audience, and a means for aggregating vote results and determining a priority.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] During presentations, there are often no questions about the content of the presentation, and the atmosphere becomes dull. This situation is frustrating for the presenter and can undermine motivation. Furthermore, a decline in interest in the content of the presentation reduces the learning effect of the audience. Therefore, there is a need for a method to smoothly conduct the Q&A session so that both the presenter and the audience can be satisfied. [Means for solving the problem]

[0005] The system provides a means for recording audio data of a presentation and a means for acquiring slide images of the presentation. It also includes a means for converting the audio data into text data and a means for analyzing the slide images and extracting key text. It also uses a means for integrating the text data and the extracted key text to analyze the content of the presentation. It also includes a means for generating questions based on the content of the presentation and a means for presenting the generated questions to the viewer's terminal. Viewers can use a means for submitting questions and a means for voting on presented questions. By building a system that includes a means for tallying the voting results and determining priorities, the Q&A session can proceed smoothly and promote a lively atmosphere.

[0006] "Audio data" refers to digital data that records the presenter's voice.

[0007] "Slide images" are image data that capture visual materials used during a presentation.

[0008] A "speech recognition engine" is software or a system that converts voice data into text.

[0009] An "image recognition engine" is software or a system that extracts text and images from slide images.

[0010] "Text data" is character data converted by a voice recognition engine.

[0011] "Key text" is important character data extracted by the image recognition engine.

[0012] "Generative AI" is an artificial intelligence system that automatically generates questions based on specific input.

[0013] "Viewer device" refers to the user's device (e.g., PC, smartphone, tablet) used to view the presentation.

[0014] The "means for posting questions" is an interface that allows viewers to input their own questions and send them to the system.

[0015] The "means for voting on questions" is a voting function that allows viewers to indicate their support for a presented question.

[0016] The "means for aggregating voting results" is a system that calculates the number of votes for questions and determines their priority.

[0017] The "means for determining priority" is a system that determines the display order of questions based on the tallied voting results. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] System Overview

[0040] The invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to ensure that the Q&A session proceeds smoothly and satisfies both the presenter and the audience.

[0041] System configuration

[0042] 1. Terminal

[0043] During the presentation, the device uses a microphone to record the presenter's voice.

[0044] At the same time, capture an image of the slide using a camera or a screen capture tool.

[0045] 2. Server

[0046] Voice data is received and converted into text using a voice recognition engine.

[0047] The acquired slide images are received and text is extracted using an image recognition engine.

[0048] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[0049] Questions are generated using generative AI based on the content of the presentation.

[0050] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[0051] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[0052] Specific processing examples

[0053] Acquiring and processing audio data

[0054] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[0055] Slide image acquisition and processing

[0056] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0057] Analysis of presentation content

[0058] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[0059] Question Generation

[0060] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0061] Viewer questions and polls

[0062] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[0063] Prioritizing and presenting questions

[0064] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[0065] System Effects

[0066] This system allows the Q&A session to proceed smoothly and create a lively atmosphere. The audience can actively participate based on their interests, and the presenter can easily understand the audience's interests through questions. The result is high-quality presentations and interactive communication.

[0067] The processing flow will be explained below.

[0068] Step 1:

[0069] As soon as the presentation begins, the device uses a microphone to record the presenter's voice and also captures images of the slides using a camera or screen capture tool.

[0070] Step 2:

[0071] The device transmits the recorded audio data to the server at regular intervals, and simultaneously transmits the acquired slide image data to the server.

[0072] Step 3:

[0073] The server then runs the received voice data through a speech recognition engine and converts it into text data, such as "AI will become mainstream in many industries over the next decade."

[0074] Step 4:

[0075] The server then runs the received slide image data through an image recognition engine to extract key text and figures from the slides. For example, text such as "AI applications: medical, education, and transportation" may be extracted.

[0076] Step 5:

[0077] The server combines the text data generated by the speech recognition engine with the key text extracted by the image recognition engine. Using this combined data, NLP technology is used to analyze the presentation content and extract key points and keywords.

[0078] Step 6:

[0079] The server uses generative AI to generate questions based on the analysis results, such as, "Which field do you think will see the most dramatic changes with the application of AI?"

[0080] Step 7:

[0081] The server presents the generated questions to the viewer's terminal in real time, and the terminal displays the received questions on the viewer's screen.

[0082] Step 8:

[0083] The user (viewer) uses the application's question submission form to enter and submit their question. For example, a question such as "Can this method be applied to other cases?" is submitted. The device then sends this question to the server.

[0084] Step 9:

[0085] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[0086] Step 10:

[0087] The server will tally all the questions and votes it receives and select the question with the most votes, for example, "How will AI revolutionize education?"

[0088] Step 11:

[0089] The server notifies the presenter of the selected high-priority questions, and the presenter can then conduct a Q&A session based on these questions.

[0090] These processing steps ensure that the Q&A session following the presentation proceeds smoothly, providing a satisfying environment for both the presenter and the audience.

[0091] Example 1

[0092] 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."

[0093] In conventional Q&A sessions, there were problems such as the complicated process for viewers to submit questions and the difficulty for presenters to grasp the audience's interests. Furthermore, when a large number of questions are submitted, it becomes difficult to select the most important questions. In such cases, the limited time for Q&A often results in situations where many viewers are unable to receive adequate answers.

[0094] 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.

[0095] In this invention, the server includes a means for generating questions using a generative model based on the content of the presentation, a means for presenting the generated questions to the viewer's terminal, and a means for aggregating voting results in real time and determining priorities. This allows viewers to quickly and easily post questions and select important questions through voting. As a result, the Q&A session proceeds efficiently, and two-way communication between the presenter and the audience is facilitated.

[0096] "Presentation audio data" refers to data that includes audio information of a speaker in a presentation, lecture, or other presentation.

[0097] A "slide image" is image data that includes visual materials and charts used by a presenter during a presentation.

[0098] "Text data" refers to data that includes text information extracted from audio data or image data.

[0099] "Main text" is text data that includes important sentence information and keywords extracted from the slide image.

[0100] "Integration" refers to the process of combining multiple pieces of data or information and organizing them into a single data set or piece of information.

[0101] "Analysis" is the process of examining data or information in detail to make its structure and meaning easier to understand.

[0102] A "generative model" is an algorithm that automatically creates new data and information using machine learning techniques such as neural networks.

[0103] "Means for generating questions" refers to technology for automatically generating questions based on the content of a presentation.

[0104] "Means for presenting on the viewer's device" refers to the technology for displaying the generated questions and information on the screen of the viewer's device.

[0105] "Means for viewers to submit questions" refers to the technology by which viewers submit their questions to the system.

[0106] "Means for viewers to vote on questions presented" refers to technology that allows viewers to rate or select questions displayed on the screen.

[0107] "Means for aggregating voting results in real time" refers to technology for instantly aggregating voting data from viewers and determining rankings and priorities.

[0108] "Means for determining priorities" refers to a technique for determining which questions are most important based on voting results and determining their order.

[0109] "Means of notification" refers to techniques for conveying specific information to a target audience, such as a presenter.

[0110] This invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to smoothly conduct Q&A sessions during presentations and to satisfy both the presenter and the audience.

[0111] System hardware and software configuration

[0112] Terminal

[0113] The terminal is a device for recording the presenter's voice and acquiring slide images. The terminal has the following functions:

[0114] Microphone: Records the presenter's voice in real time.

[0115] Camera or screen capture tool: Take an image of the slide.

[0116] The terminal may be a general-purpose computing device such as a laptop, tablet, or smartphone.

[0117] server

[0118] The server is the core component that analyzes the audio data and slide image data, and generates and presents questions. The specific software and functions used are as follows:

[0119] Speech recognition engine: Converts voice data into text data using Google Cloud Speech-to-Text or similar.

[0120] Image recognition engine: Extract text from slide images using Google Cloud Vision or similar.

[0121] NLP technology: Using natural language processing technology, voice recognition and image recognition results are analyzed to extract key points and keywords.

[0122] Generative AI model: Using OpenAI's GPT-3 and other models, questions are generated based on the analysis results.

[0123] Real-time tallying function: Tally the votes from the audience and notify the presenter of the highest priority questions.

[0124] User

[0125] Users (viewers) can post questions and vote on questions provided from their own devices.

[0126] Specific examples of program processing

[0127] Acquiring and processing audio data

[0128] The device records the speaker's voice and sends the audio data to a server. The server then converts the audio data into text using Google Cloud Speech-to-Text. For example, a statement such as "AI has a wide range of applications" in a presentation can be converted into text.

[0129] Slide image acquisition and processing

[0130] The device uses a screen capture tool to capture an image of the slide and sends it to the server, which then uses Google Cloud Vision to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0131] Analysis of presentation content

[0132] The server combines the received voice and image recognition results and uses NLP technology to analyze the content of the presentation, providing key points such as, "This presentation explains different areas in which AI can be applied."

[0133] Question Generation

[0134] Based on the analysis results, the server uses a generative AI model to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0135] Viewer questions and polls

[0136] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Viewers can also vote on the questions presented, and the results of these votes are tallied in real time by the server.

[0137] Prioritizing and presenting questions

[0138] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[0139] Prompt Sentence Examples

[0140] Generate questions based on the given presentation using the following criteria:

[0141] Announcement: "AI has a wide range of applications. It is being used in a wide variety of fields, from medicine to education and even entertainment."

[0142] Generating conditions: Ask for specific examples and numerical questions.

[0143] This system facilitates smooth Q&A sessions following presentations and allows for two-way communication between presenters and audience members.

[0144] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0145] Step 1:

[0146] The device records the presenter's voice. When the presentation begins, the device's microphone automatically starts up and collects the voice data. The input here is the voice captured in real time through the microphone, and the output is the voice data that is temporarily saved in local storage. Specifically, the voice data is saved periodically every few seconds, and this saved data is put on hold for further processing.

[0147] Step 2:

[0148] The device uses a screen capture tool to obtain image data of the slides. Each time a slide changes, the device automatically performs a screen capture and obtains image data. The input here is the slide screen obtained through the screen capture tool, and the output is image data that is temporarily saved in local storage. Specifically, each time the slide changes to a new page, an image is captured and saved.

[0149] Step 3:

[0150] The device sends the recorded audio data to the server. The input here is the audio data stored in the local storage, and the output is the audio data sent to the server. Specifically, the audio files are divided into fixed sizes and sent to the server one by one, and the server waits to receive them.

[0151] Step 4:

[0152] The server converts the received voice data into text data using a speech recognition engine such as Google Cloud Speech-to-Text. The input here is the transmitted voice data, and the output is text data. Specifically, the voice file is input into the engine, and within a few seconds, the corresponding text is generated and stored in a database.

[0153] Step 5:

[0154] The device sends the acquired slide image data to the server. The input here is the image data stored in the local storage, and the output is the image data sent to the server. Specifically, the captured image files are sent to the server one by one, and the server enters a waiting state to receive them.

[0155] Step 6:

[0156] The server analyzes the received slide image data using an image recognition engine such as Google Cloud Vision to extract key text. The input here is the transmitted slide image data, and the output is the extracted text data. Specifically, the image file is input into the engine, and the text is analyzed, extracted, and stored in a database.

[0157] Step 7:

[0158] The server integrates the results of speech recognition and image recognition and uses NLP (natural language processing) technology to perform a detailed analysis of the presentation content. The input here is text data from speech recognition and image recognition, and the output is a summary of the presentation content and key keywords. Specifically, the text data is input into the NLP model, and key points are extracted and summarized.

[0159] Step 8:

[0160] The server uses a generative AI model (e.g., GPT-3) based on the analysis results to generate questions. The input here is the text data of the analyzed presentation, and the output is the generated question text. Specifically, the analysis data is input into the AI ​​model as a prompt, and natural questions are generated.

[0161] Step 9:

[0162] The server presents the generated question to the viewer's device in real time. The input here is the generated question text, and the output is the question displayed on the viewer's device. Specifically, the question text is sent to the viewer and displayed on the viewer's screen.

[0163] Step 10:

[0164] Users (viewers) post their own questions. The input here is text data entered by the viewer, and the output is the question data sent to the server. Specifically, when the viewer enters a question in the input field on their device and presses the send button, the data is sent to the server.

[0165] Step 11:

[0166] Users (viewers) vote on the questions presented. The input here is the viewer's voting operation, and the output is the voting data sent to the server. Specifically, when a viewer presses the voting button on their device, the data is sent to the server in real time.

[0167] Step 12:

[0168] The server tally the votes from viewers in real time and determine the priorities. The input here is the voting data, and the output is a list of questions with determined priorities. Specifically, the voting results are tallyed in real time using a tallying algorithm, and priorities are assigned.

[0169] Step 13:

[0170] The server notifies the presenter of high-priority questions. The input is a prioritized list of questions, and the output is the questions to be displayed on the presenter's device. Specifically, the highest-priority question is sent to the presenter's device and displayed visually.

[0171] (Application example 1)

[0172] 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."

[0173] In conventional presentations and product demonstrations, questions are generated and presented manually, making it difficult to smoothly conduct a Q&A session. Especially in situations with a large number of viewers or customers, important questions can be overlooked, reducing viewer and customer satisfaction. Furthermore, there was no system for collecting and voting on questions from viewers and customers in real time, making it difficult to effectively convey questions of interest to the presenter or presenter. The present invention aims to solve these problems.

[0174] 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.

[0175] In this invention, the server includes means for recording audio data of the presentation, means for acquiring slide images of the presentation, means for converting the audio data into text data, means for analyzing the slide images and extracting key text, means for integrating the text data and the extracted key text to analyze the content of the presentation, means for generating questions based on the content of the presentation, means for presenting the generated questions to user terminals, means for users to submit questions, means for users to vote on the submitted questions, means for tallying up the votes and determining priorities, means for analyzing the audio and images of the product introduction demonstration and automatically generating and presenting questions from customers, and means for presenting the generated questions to user terminals in real time. This enables the Q&A session to proceed smoothly, important questions to be effectively conveyed to the presenter or introducer, and increases the interest and satisfaction of many viewers and customers.

[0176] "Means for recording audio data of a presentation" refers to a device or system for recording the presenter's voice using an audio input device.

[0177] "Means for capturing images of presentation slides" refers to a device or system for capturing images of slides during a presentation using a camera or screen capture tool.

[0178] The "means for converting the voice data into text data" refers to a system or method for converting recorded voice data into text using a voice recognition engine.

[0179] The "means for analyzing the slide images and extracting the main text" refers to a system or method for extracting text information from the acquired slide images using an image recognition engine.

[0180] "Means for integrating the above text data and the extracted main text to analyze the content of the presentation" refers to a system or method for integrating the results of speech recognition and image recognition and analyzing the content of the presentation using natural language processing technology.

[0181] A "means for generating questions based on presentation content" is a system or method for automatically generating questions from presentation content using a generative AI model.

[0182] The "means for presenting the generated question on the user's terminal" is a system or method for displaying the generated question on a device used by a viewer or customer.

[0183] "Means for users to post questions" refers to a device or function that allows viewers or customers to submit their questions to the system.

[0184] "Means for users to vote on questions presented" refers to a system or function that allows viewers or customers to vote on questions that are displayed.

[0185] The "means for aggregating vote results and determining priority" refers to a system or method for collecting voting data from users and determining the question that receives the most votes as the highest priority.

[0186] "Means for analyzing audio and video of a product introduction demonstration and automatically generating and presenting questions from customers" refers to a system or method for analyzing audio and video during a product introduction and generating and presenting questions from customers based on that analysis.

[0187] "Means for presenting generated questions to a user's device in real time" refers to a system or method for displaying generated questions on the spot and instantly on a viewer's or customer's device.

[0188] System Overview

[0189] This invention is a system that analyzes the audio and video of a product introduction demonstration and automatically generates and presents questions to customers, aiming to ensure a smooth Q&A session and satisfy both the customer and the introducer.

[0190] System configuration

[0191] Terminal

[0192] The terminal performs the following functions during the product demonstration:

[0193] Audio recording: Record the introducer's voice using the device's microphone.

[0194] Image capture: Uses the device's camera to capture images during the demonstration.

[0195] server

[0196] The server does the following:

[0197] Speech recognition: Receives voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text).

[0198] Image Analysis: Receive the captured image and extract the key text using Tesseract OCR.

[0199] Content analysis: The results of voice recognition and image recognition are integrated, and the introduction content is analyzed using natural language processing technology.

[0200] Question generation: Based on the analysis results, questions are generated using a generative AI model (e.g., OpenAI's GPT-3).

[0201] Question presentation: The generated questions are presented to the user's device in real time.

[0202] Voting function: Provides a mechanism for users to vote on questions posted by users.

[0203] Prioritization: Tally the votes and present the highest priority questions to the introducer.

[0204] Specific processing examples

[0205] Acquiring and processing audio data

[0206] The device uses a microphone to record the introducer's voice and sends the voice data to the server, which then converts the voice data into text data using Google Speech-to-Text. For example, a statement such as "This new smartphone has an innovative camera function" can be recognized.

[0207] Slide image acquisition and processing

[0208] The device uses its camera to capture an image of the demonstration and sends it to a server, which uses Tesseract OCR to extract key text from the image, such as "Advanced night mode photography with AI."

[0209] Content analysis and question generation

[0210] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. Based on this analysis, a prompt is created to be provided to the generative AI model (GPT-3), which then generates a question. For example, the prompt is as follows:

[0211] Example of question generation

[0212] "Generate customer questions based on the following text: Voiceover: "This new smartphone comes with revolutionary camera features." Slide: "Advanced night mode photography powered by AI.""

[0213] An example of a question that can be generated is:

[0214] "Can you tell me more about the night mode photography feature on this phone? I'm especially interested to know how AI helps with that."

[0215] is.

[0216] Viewer questions and polls

[0217] Users (viewers or customers) can post their own questions from their devices. For example, a question such as, "I'm interested in how AI will affect the filming process." Similarly, viewers can also vote on questions presented. The results of these votes are tallied in real time by the server.

[0218] Prioritizing and presenting questions

[0219] The server tally the votes from the viewers and notify the introducer of the question that received the most votes. For example, if "Please explain the details of the night mode shooting function" received the most votes, this will be displayed to the introducer.

[0220] System Effects

[0221] This system facilitates Q&A sessions during product demonstrations, allowing customers to actively participate based on their interests. It also makes it easier for introducers to understand customer interests through questions, resulting in high-quality demonstrations and interactive communication.

[0222] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0223] Step 1:

[0224] The terminal uses a microphone to record the voice during the product introduction demonstration. This recorded voice data is sent to the server. The input is the recorded voice data, and the output is the voice data sent to the server.

[0225] Step 2:

[0226] The server converts the received voice data into text data using the Google Speech-to-Text API. The input is the voice data, and the output is the converted text data. As part of this conversion process, the server makes an API call and obtains the results.

[0227] Step 3:

[0228] The device uses a camera to capture images of slides during the demonstration. This captured image data is also sent to the server. The input is the captured image data, and the output is the image data sent to the server.

[0229] Step 4:

[0230] The server analyzes the received image data using Tesseract OCR and extracts key character information. The input is image data, and the output is extracted text information. As part of this image analysis process, the server passes the image data to the OCR engine and returns the analysis results as text.

[0231] Step 5:

[0232] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. The input is text data of the speech and text information of the image, and the output is the analyzed introduction. The server combines the text data to extract important keywords and context.

[0233] Step 6:

[0234] The server generates a question using a generative AI model (GPT-3) based on the analysis results. The input is the analyzed introduction content, and the output is the generated question. In this procedure, the server sends a prompt to the generative AI model to generate a question.

[0235] Step 7:

[0236] The server presents the generated questions to the user's terminal in real time. The input is the generated question, and the output is the question displayed on the user's terminal. The server sends data to each terminal and displays it on the screen.

[0237] Step 8:

[0238] Users can post their own questions from their terminals. The input is the question entered by the user, and the output is the question sent to the server. The terminal receives the user's input and sends it to the server.

[0239] Step 9:

[0240] Users can vote on presented questions. The input is the vote selected by the user, and the output is the voting data sent to the server. The user's terminal sends the voting information for the selected question.

[0241] Step 10:

[0242] The server tally the votes and decides the priority questions. The input is the voting data and the output is the priority questions. The server tally each vote and selects the question with the most votes.

[0243] Step 11:

[0244] The server notifies the introducer's terminal of high-priority questions. The input is the high-priority questions, and the output is the questions displayed on the introducer's terminal. The server sends the selected questions to the introducer in real time and displays them on the screen.

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

[0246] System Overview

[0247] This invention provides a presentation support system that combines an emotion engine that recognizes the user's emotions. The system analyzes the presentation's audio data and slide images, automatically generates and presents questions, and combines this with a means to adjust the priority of questions using emotion data, making the Q&A session more effective and engaging.

[0248] System configuration

[0249] 1. Terminal

[0250] During the presentation, the device will record the presenter's voice using a microphone and capture images of the slides using a camera or screen capture tool.

[0251] 2. Server

[0252] Voice data is received and converted into text using a voice recognition engine.

[0253] The acquired slide images are received and text is extracted using an image recognition engine.

[0254] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[0255] Questions are generated using generative AI based on the content of the presentation.

[0256] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[0257] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[0258] 3. Emotion Engine

[0259] It recognizes emotions from viewers' facial expressions and tone of voice, and sends the data to the server in real time.

[0260] Dynamically adjust question priorities based on sentiment data.

[0261] Specific processing examples

[0262] Acquiring and processing audio data

[0263] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[0264] Slide image acquisition and processing

[0265] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0266] Analysis of presentation content

[0267] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[0268] Question Generation

[0269] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0270] Audience emotion recognition

[0271] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. For example, if the viewer has an interested look on their face, that data is used to prioritize questions.

[0272] Viewer questions and polls

[0273] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[0274] Prioritizing and presenting questions

[0275] The server then selects the highest-priority questions based on audience votes and sentiment data and notifies the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, it will be displayed to the presenter.

[0276] System Effects

[0277] This system makes Q&A sessions during presentations more effective and engaging. Audience members can actively participate based on their interests, and presenters can answer questions while understanding the audience's emotions, enabling deeper communication. The result is higher-quality presentations and more interactive engagement.

[0278] The processing flow will be explained below.

[0279] Step 1:

[0280] As soon as the presentation begins, the device will use the microphone to record the presenter's voice and the camera or screen capture tool to capture images of the slides.

[0281] Step 2:

[0282] The terminal transmits the recorded audio data and the acquired slide images to the server at regular intervals.

[0283] Step 3:

[0284] The server passes the received voice data to a speech recognition engine, which converts the voice data into text data. For example, a statement such as "AI will become mainstream in many industries over the next decade" is converted into text data.

[0285] Step 4:

[0286] The server passes the received slide image data to an image recognition engine, which extracts key text and figures from the slide. For example, text such as "AI applications: medical, education, and transportation" is extracted.

[0287] Step 5:

[0288] The server combines the text data generated by speech recognition with the key text extracted by image recognition, and uses NLP technology to analyze the presentation and extract key points and keywords.

[0289] Step 6:

[0290] Based on the analysis results, the server uses generative AI to generate questions such as, "Which field do you think will see the most dramatic changes with the application of AI?" These questions are then displayed in real time on the viewer's device.

[0291] Step 7:

[0292] The emotion engine analyzes viewers' facial expressions and vocal tone in real time, using cameras and microphones to capture viewers' emotional data (e.g., interest, surprise, joy) while they watch.

[0293] Step 8:

[0294] The emotion engine sends the acquired emotion data to the server, and the priority of questions is dynamically adjusted based on that data. For example, questions that show a lot of interest to viewers can be given higher priority.

[0295] Step 9:

[0296] Users (viewers) can post their own questions from their devices. For example, they can post a question such as, "I would like to know more about the risks of AI." The device then sends this question to the server.

[0297] Step 10:

[0298] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[0299] Step 11:

[0300] The server aggregates all questions and votes and uses audience sentiment data to determine priority. For example, if the question "How will AI revolutionize education?" receives a lot of votes and emotional responses from viewers, it will be prioritized.

[0301] Step 12:

[0302] The server then notifies the presenter of the selected high-priority questions, allowing the presenter to conduct a Q&A session based on these questions. For example, by prioritizing answers to questions that evoke strong emotional responses, the presenter can increase audience interest.

[0303] Through these processing steps, the Q&A session during the presentation can be conducted more effectively and attractively, providing a satisfying environment for both the presenter and the audience.

[0304] Example 2

[0305] 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."

[0306] Conventional presentation support systems have the problem that the priority of questions received by the presenter is not optimized based on the audience's interests and emotions, making it difficult to effectively conduct a Q&A session. In addition, there is a lack of a way to present questions without distinguishing between questions from the audience and questions generated by the AI ​​generation system, making it difficult to realize a Q&A session that reflects the audience's interests.

[0307] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for converting audio data into text data, means for analyzing slide images and extracting key text, means for generating questions using a generative AI model, means for using an emotion engine that analyzes viewers' facial expressions and tone of voice, and means for dynamically adjusting the priority of questions based on emotion data. This enables the progress of an effective Q&A session that reflects viewers' interests and emotions.

[0308] "Means for recording audio data of a presentation" refers to a device or method for collecting the presenter's remarks as digital data using audio equipment such as a microphone.

[0309] "Means for acquiring slide images of presentations" refers to a device or method for capturing images of slides used during presentations and collecting them as digital image data.

[0310] "Means for converting voice data into text data" refers to software or a system that uses voice recognition technology to convert recorded voice data into text data.

[0311] "Means for analyzing slide images and extracting key text" refers to software or a system that extracts text information from slide images acquired using image recognition technology.

[0312] "Means for integrating text data and extracted key texts to analyze presentation content" refers to software or systems that integrate collected text data and use natural language processing technology to understand and analyze presentation content.

[0313] "Means using a generative AI model that generates questions based on the content of a presentation" refers to an artificial intelligence model and method that automatically generates questions based on the content of a presentation analyzed using natural language generation technology.

[0314] The "means for presenting the generated questions to the viewer terminal" refers to software or communication technology for displaying the questions generated by the server on the device used by the viewer.

[0315] "Means for viewers to post questions and vote on submitted questions" means an interface or system through which viewers can submit their own questions and rate other submitted questions.

[0316] The "means for aggregating vote results and determining question priority" refers to an algorithm or system that aggregates vote data from viewers and determines the importance of each question based on the results.

[0317] "Means using an emotion engine that analyzes the viewer's facial expressions and tone of voice" refers to software or equipment for analyzing the viewer's facial expressions and tone of voice and extracting emotion data.

[0318] "Means for dynamically adjusting question priority based on emotional data" refers to an algorithm or system that changes the display order and importance of questions in real time, taking into account the acquired emotional data.

[0319] The present invention relates to a presentation support system that optimizes a question and answer session using user emotion data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following describes in detail the preferred embodiments of the present invention.

[0320] First, we will describe the main embodiment of the system, which uses hardware and software to collect and analyze audio data, image data, and emotional data to effectively facilitate a question-and-answer session between the presenter and the audience.

[0321] Hardware and software used

[0322] The device includes a microphone, camera, and screen capture tool. These devices are used to collect the presenter's voice and slide images in real time. For example, when a presenter explains "Medical Applications of AI Technology," the device records the voice with the microphone and captures slide images with the screen capture tool.

[0323] The server is a central system for analyzing and integrating the collected data. It receives the audio data and slide image data sent from the devices via an internet connection and processes them as follows:

[0324] 1. Speech recognition: Using a speech recognition engine such as Google Cloud Speech-to-Text, the audio data is converted into text data. For example, a speaker's words such as "AI technology is being used in the medical field" are obtained as text data.

[0325] 2. Image Recognition: Using an image recognition engine such as Google Cloud Vision, key text is extracted from the slide image, specifically the slide title "Applications of AI in Medicine."

[0326] 3. Analysis of presentation content: The acquired text data is integrated and the presentation content is analyzed using natural language processing (NLP) technology. Here, the key points and key keywords of the presentation are identified. For example, key points such as "This presentation explains different application areas of AI" are extracted.

[0327] 4. Question generation: Based on the analysis results, questions are generated using generative AI (such as OpenAI's GPT-4). The generated questions are displayed in real time on the viewer's device. For example, a question such as, "What specific changes can we expect to see in the medical field through the use of AI?"

[0328] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotion data to the server. For example, if the viewer has an interested expression, the data will be collected by the emotion engine and sent to the server.

[0329] Prompt Sentence Examples

[0330] "Generate a question based on the following presentation: AI has achieved remarkable results in the field of medical imaging diagnosis. In particular, AI algorithms are now able to detect cellular abnormalities with high accuracy. This has significantly improved the speed and accuracy of diagnoses and reduced the burden on doctors."

[0331] This allows viewers to post questions based on their own interests and also vote on questions generated by other viewers. The server dynamically prioritizes questions based on the collected voting and sentiment data, and provides the presenter with the most relevant questions.

[0332] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0333] Processing flow and step details

[0334] Step 1: Acquiring and sending audio data

[0335] The device records the presenter's voice with a microphone and sends the audio data to the server. Specifically, the audio during the presentation is recorded in real time and converted into a digital audio file. This audio file is then sent to the server via the Internet.

[0336] Input: Speaker's voice

[0337] Data processing: audio recording and digital conversion

[0338] Output: Digital audio file (sent to server)

[0339] Step 2: Capture and send slide images

[0340] The device uses a screen capture tool to capture image data of the slides and sends it to the server. When the presenter switches slides, the device automatically performs a screen capture and sends the captured image to the server.

[0341] Input: Slide screen

[0342] Data processing: Screen capture

[0343] Output: Image data file (sent to server)

[0344] Step 3: Convert audio data to text

[0345] The server receives the voice data and converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The digital voice file sent to the server is analyzed and converted by the speech recognition engine and saved as a string of characters.

[0346] Input: Digital audio data

[0347] Data processing: Speech recognition (voice to text conversion)

[0348] Output: Text data

[0349] Step 4: Extract text from slide images

[0350] The server extracts key text from the slide images using an image recognition engine (e.g., Google Cloud Vision). It analyzes the received image data files to identify and extract text within the images.

[0351] Input: Slide image data

[0352] Data processing: Image recognition (text extraction)

[0353] Output: Text data

[0354] Step 5: Analyze the presentation

[0355] The server integrates the text data obtained from the speech and image recognition and analyzes the presentation using natural language processing (NLP) technology. Based on the integrated text data, the main points and key keywords of the presentation are identified.

[0356] Input: Text data extracted from audio and slides

[0357] Data Processing: Integration and NLP Analysis

[0358] Output: Gist and keyword data

[0359] Step 6: Generate questions

[0360] The server uses a generative AI model (e.g., GPT-4) to generate questions based on the analysis results. The server uses the key points and keywords obtained as prompts to generate appropriate questions.

[0361] Input: Key points and keyword data

[0362] Data processing: Question generation (generative AI model)

[0363] Output: Question text

[0364] Step 7: Posing the Question

[0365] The server presents the generated questions to the viewer's terminal in real time, and transmits the generated questions to the viewer's device and provides them through an interface for display.

[0366] Input: Question text

[0367] Data processing: Submitting and presenting questions

[0368] Output: Question displayed on the viewer's device

[0369] Step 8: Collect and send emotion data

[0370] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. The camera and microphone collect the viewer's emotions and send them to the server as digital data.

[0371] Input: Viewer facial expressions and tone of voice

[0372] Data processing: Sentiment analysis

[0373] Output: Emotion data (sent to server)

[0374] Step 9: Viewer polls

[0375] Users (viewers) vote on questions presented through their terminals. Each viewer votes for the questions they are interested in, and the vote data is sent to the server in real time.

[0376] Input: Viewer votes

[0377] Data processing: Collecting voting data

[0378] Output: Voting data (sent to server)

[0379] Step 10: Prioritize and present questions

[0380] The server dynamically determines the priority of questions based on the collected voting data and sentiment data, and selects the most important question and displays it on the presenter's device.

[0381] Input: Voting data and sentiment data

[0382] Data processing: data integration and prioritization

[0383] Output: Top priority question (displayed to presenter's device)

[0384] The above processing steps enable an effective and engaging Q&A session between the presenter and the audience.

[0385] (Application example 2)

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

[0387] Conventional presentation support systems and robot work support systems have problems in that they are unable to fully realize Q&A during presentations or improve the efficiency of robot work in factories. Specifically, they do not properly manage the priority of audience questions, and do not immediately adjust the robot's behavior in response to the worker's instructions or emotions, hindering efficient work. This results in low satisfaction for viewers and workers, and a lack of two-way communication.

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

[0389] In this invention, the server includes means for generating questions based on the content of the presentation, means for presenting the generated questions to viewer terminals, means for viewers to post questions, means for viewers to vote on the presented questions, means for tallying up vote results and determining priorities, means for acquiring robot work status data, means for acquiring voice instructions from workers and dynamically optimizing work content using a generation AI, and means for acquiring worker emotion data and adjusting work plans based on the emotion data. This improves the efficiency of question and answer sessions between viewers and presenters, enabling more efficient robot work in factories.

[0390] "Presentation audio data" refers to recorded data of the audio uttered by the presenter during the presentation.

[0391] "Slide images" are image data of visual slides used during a presentation.

[0392] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0393] An "image recognition engine" is software or hardware for extracting key text from image data.

[0394] "Integration" is the process of combining multiple different data sets into a single dataset.

[0395] "Question generation" is the process of automatically creating appropriate questions based on the presentation content and analysis results.

[0396] "Viewer device" means the device used by the viewer to view the presentation.

[0397] "Voting results" are the aggregated results of votes cast by viewers on questions.

[0398] "Priority" is the measure by which a particular question or task is judged to be more important than other questions or tasks.

[0399] "Robot work status data" refers to data relating to the work currently being performed by the robot and its status.

[0400] "Generative AI" is artificial intelligence that generates appropriate questions and answers based on given data and conditions.

[0401] "Emotion data" refers to emotional information obtained from the worker's facial expressions, tone of voice, etc.

[0402] "Dynamic optimization" is the process of adjusting to the optimal work plan in real time as conditions change.

[0403] A "work plan" is the design of procedures and schedules for carrying out a specific task.

[0404] This invention relates to a "robot work support system" for improving the efficiency of robot work in factories. This system combines a function to support questions and answers during presentations with a function to dynamically optimize robot work in factories, thereby increasing the satisfaction of viewers and workers.

[0405] System Overview

[0406] Terminal

[0407] The terminal uses the microphone and camera built into the robot body to capture the worker's voice instructions and work status in real time. The smart glasses or tablet used by the worker also function as part of the system.

[0408] server

[0409] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the acquired voice data into text data, an image recognition engine (e.g., Amazon Rekognition) to analyze image data acquired from the robot's camera, and an NLP technology (e.g., spaCy) to integrate the results of speech recognition and image recognition, and a generative AI (e.g., OpenAI GPT) to generate appropriate questions and work plans.

[0410] Emotion Engine

[0411] An emotion engine (e.g., Microsoft Azure Emotion API) is used to analyze workers' facial expressions and voice tones in real time to obtain emotional data, which is used to dynamically adjust the work plan.

[0412] Specific processing examples

[0413] 1. Acquiring and processing audio data

[0414] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits them to the server. The server then converts the voice data into text data using a voice recognition engine. During this process, an instruction such as "Please install the next part" is obtained as text data.

[0415] 2. Image data acquisition and processing

[0416] The terminal uses the robot's built-in camera to capture images of the work status and parts, and sends them to the server. The server then analyzes the image data using an image recognition engine to extract key information. For example, specific parts and work status may be recognized.

[0417] 3. Analysis and adjustment of work content

[0418] The server integrates and analyzes the voice and image recognition results, and uses generative AI to generate questions and work plans, which are then presented to the worker's smart glasses or tablet.

[0419] 4. Acquiring and Processing Emotional Data

[0420] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, capturing emotional data. For example, if the worker is feeling dissatisfied, the data is sent to the server in real time and used to adjust the work plan.

[0421] Prompt Sentence Examples

[0422] "When an AI robot follows instructions in a factory task, suggest ways to optimize the task. Voice command: 'Install the next part.' Worker emotion: Frustration evident."

[0423] This allows for efficient question and answer sessions between the audience and the presenter, while also dynamically optimizing the work of robots in factories, improving worker satisfaction and work efficiency.

[0424] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0425] Step 1:

[0426] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits the voice data to the server. The input is the worker's voice instructions, and the output is the voice data sent to the server. Specifically, the microphone captures the voice, which is then saved and transmitted as digital data.

[0427] Step 2:

[0428] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data, and the output is text data corresponding to the voice instructions. Specifically, the speech recognition engine analyzes the voice data and generates text using a language model.

[0429] Step 3:

[0430] The terminal uses the robot's built-in camera to acquire image data of the work situation and sends that image data to the server. The input is visual information of the work situation, and the output is image data sent to the server. Specifically, the camera captures the image, saves it as digital data, and sends it.

[0431] Step 4:

[0432] The server uses an image recognition engine to analyze the received image data and extract key information. The input is the image data, and the output is extracted key text and recognized objects. Specifically, the image recognition engine analyzes the image data pixel by pixel to identify key text and objects.

[0433] Step 5:

[0434] The server integrates the results of speech recognition and image recognition and analyzes the work content using NLP technology. The input is text data and extracted key information, and the output is the gist of the analyzed work content. Specifically, the text data and image data are integrated into a single dataset and analyzed using an NLP model.

[0435] Step 6:

[0436] The server uses a generation AI to generate appropriate questions and work plans based on the analysis results. The input is the gist of the analyzed work content, and the output is the generated questions and work plans. Specifically, the generation AI generates questions and work plans related to the prompt.

[0437] Step 7:

[0438] The terminal presents the generated questions and work plans to the worker's smart glasses or tablet. The input is the generated questions and work plans, and the output is a visual display. The specific operation is to display the information on the device's display.

[0439] Step 8:

[0440] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, and acquires emotion data. The input is video data of the worker's facial expressions, and the output is the emotion recognition results. Specifically, the camera captures the facial expressions, and the emotion recognition engine analyzes the video data.

[0441] Step 9:

[0442] The server dynamically adjusts the work plan based on the acquired emotion data. The input is emotion data, and the output is an adjusted work plan. Specifically, it analyzes the emotion recognition results and modifies the priority and procedures of the work plan in real time.

[0443] 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.

[0444] 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.

[0445] 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.

[0446] [Second embodiment]

[0447] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0448] 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.

[0449] 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).

[0450] 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.

[0451] 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.

[0452] 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).

[0453] 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.

[0454] 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.

[0455] 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.

[0456] 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.

[0457] 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.

[0458] 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."

[0459] System Overview

[0460] The invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to ensure that the Q&A session proceeds smoothly and satisfies both the presenter and the audience.

[0461] System configuration

[0462] 1. Terminal

[0463] During the presentation, the device uses a microphone to record the presenter's voice.

[0464] At the same time, capture an image of the slide using a camera or a screen capture tool.

[0465] 2. Server

[0466] Voice data is received and converted into text using a voice recognition engine.

[0467] The acquired slide images are received and text is extracted using an image recognition engine.

[0468] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[0469] Questions are generated using generative AI based on the content of the presentation.

[0470] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[0471] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[0472] Specific processing examples

[0473] Acquiring and processing audio data

[0474] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[0475] Slide image acquisition and processing

[0476] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0477] Analysis of presentation content

[0478] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[0479] Question Generation

[0480] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0481] Viewer questions and polls

[0482] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[0483] Prioritizing and presenting questions

[0484] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[0485] System Effects

[0486] This system allows the Q&A session to proceed smoothly and create a lively atmosphere. The audience can actively participate based on their interests, and the presenter can easily understand the audience's interests through questions. The result is high-quality presentations and interactive communication.

[0487] The processing flow will be explained below.

[0488] Step 1:

[0489] As soon as the presentation begins, the device uses a microphone to record the presenter's voice and also captures images of the slides using a camera or screen capture tool.

[0490] Step 2:

[0491] The device transmits the recorded audio data to the server at regular intervals, and simultaneously transmits the acquired slide image data to the server.

[0492] Step 3:

[0493] The server then runs the received voice data through a speech recognition engine and converts it into text data, such as "AI will become mainstream in many industries over the next decade."

[0494] Step 4:

[0495] The server then runs the received slide image data through an image recognition engine to extract key text and figures from the slides. For example, text such as "AI applications: medical, education, and transportation" may be extracted.

[0496] Step 5:

[0497] The server combines the text data generated by the speech recognition engine with the key text extracted by the image recognition engine. Using this combined data, NLP technology is used to analyze the presentation content and extract key points and keywords.

[0498] Step 6:

[0499] The server uses generative AI to generate questions based on the analysis results, such as, "Which field do you think will see the most dramatic changes with the application of AI?"

[0500] Step 7:

[0501] The server presents the generated questions to the viewer's terminal in real time, and the terminal displays the received questions on the viewer's screen.

[0502] Step 8:

[0503] The user (viewer) uses the application's question submission form to enter and submit their question. For example, a question such as "Can this method be applied to other cases?" is submitted. The device then sends this question to the server.

[0504] Step 9:

[0505] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[0506] Step 10:

[0507] The server will tally all the questions and votes it receives and select the question with the most votes, for example, "How will AI revolutionize education?"

[0508] Step 11:

[0509] The server notifies the presenter of the selected high-priority questions, and the presenter can then conduct a Q&A session based on these questions.

[0510] These processing steps ensure that the Q&A session following the presentation proceeds smoothly, providing a satisfying environment for both the presenter and the audience.

[0511] Example 1

[0512] 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."

[0513] In conventional Q&A sessions, there were problems such as the complicated process for viewers to submit questions and the difficulty for presenters to grasp the audience's interests. Furthermore, when a large number of questions are submitted, it becomes difficult to select the most important questions. In such cases, the limited time for Q&A often results in situations where many viewers are unable to receive adequate answers.

[0514] 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.

[0515] In this invention, the server includes a means for generating questions using a generative model based on the content of the presentation, a means for presenting the generated questions to the viewer's terminal, and a means for aggregating voting results in real time and determining priorities. This allows viewers to quickly and easily post questions and select important questions through voting. As a result, the Q&A session proceeds efficiently, and two-way communication between the presenter and the audience is facilitated.

[0516] "Presentation audio data" refers to data that includes audio information of a speaker in a presentation, lecture, or other presentation.

[0517] A "slide image" is image data that includes visual materials and charts used by a presenter during a presentation.

[0518] "Text data" refers to data that includes text information extracted from audio data or image data.

[0519] "Main text" is text data that includes important sentence information and keywords extracted from the slide image.

[0520] "Integration" refers to the process of combining multiple pieces of data or information and organizing them into a single data set or piece of information.

[0521] "Analysis" is the process of examining data or information in detail to make its structure and meaning easier to understand.

[0522] A "generative model" is an algorithm that automatically creates new data and information using machine learning techniques such as neural networks.

[0523] "Means for generating questions" refers to technology for automatically generating questions based on the content of a presentation.

[0524] "Means for presenting on the viewer's device" refers to the technology for displaying the generated questions and information on the screen of the viewer's device.

[0525] "Means for viewers to submit questions" refers to the technology by which viewers submit their questions to the system.

[0526] "Means for viewers to vote on questions presented" refers to technology that allows viewers to rate or select questions displayed on the screen.

[0527] "Means for aggregating voting results in real time" refers to technology for instantly aggregating voting data from viewers and determining rankings and priorities.

[0528] "Means for determining priorities" refers to a technique for determining which questions are most important based on voting results and determining their order.

[0529] "Means of notification" refers to techniques for conveying specific information to a target audience, such as a presenter.

[0530] This invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to smoothly conduct Q&A sessions during presentations and to satisfy both the presenter and the audience.

[0531] System hardware and software configuration

[0532] Terminal

[0533] The terminal is a device for recording the presenter's voice and acquiring slide images. The terminal has the following functions:

[0534] Microphone: Records the presenter's voice in real time.

[0535] Camera or screen capture tool: Take an image of the slide.

[0536] The terminal may be a general-purpose computing device such as a laptop, tablet, or smartphone.

[0537] server

[0538] The server is the core component that analyzes the audio data and slide image data, and generates and presents questions. The specific software and functions used are as follows:

[0539] Speech recognition engine: Converts voice data into text data using Google Cloud Speech-to-Text or similar.

[0540] Image recognition engine: Extract text from slide images using Google Cloud Vision or similar.

[0541] NLP technology: Using natural language processing technology, voice recognition and image recognition results are analyzed to extract key points and keywords.

[0542] Generative AI model: Using OpenAI's GPT-3 and other models, questions are generated based on the analysis results.

[0543] Real-time tallying function: Tally the votes from the audience and notify the presenter of the highest priority questions.

[0544] User

[0545] Users (viewers) can post questions and vote on questions provided from their own devices.

[0546] Specific examples of program processing

[0547] Acquiring and processing audio data

[0548] The device records the speaker's voice and sends the audio data to a server. The server then converts the audio data into text using Google Cloud Speech-to-Text. For example, a statement such as "AI has a wide range of applications" in a presentation can be converted into text.

[0549] Slide image acquisition and processing

[0550] The device uses a screen capture tool to capture an image of the slide and sends it to the server, which then uses Google Cloud Vision to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0551] Analysis of presentation content

[0552] The server combines the received voice and image recognition results and uses NLP technology to analyze the content of the presentation, providing key points such as, "This presentation explains different areas in which AI can be applied."

[0553] Question Generation

[0554] Based on the analysis results, the server uses a generative AI model to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0555] Viewer questions and polls

[0556] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Viewers can also vote on the questions presented, and the results of these votes are tallied in real time by the server.

[0557] Prioritizing and presenting questions

[0558] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[0559] Prompt Sentence Examples

[0560] Generate questions based on the given presentation using the following criteria:

[0561] Announcement: "AI has a wide range of applications. It is being used in a wide variety of fields, from medicine to education and even entertainment."

[0562] Generating conditions: Ask for specific examples and numerical questions.

[0563] This system facilitates smooth Q&A sessions following presentations and allows for two-way communication between presenters and audience members.

[0564] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0565] Step 1:

[0566] The device records the presenter's voice. When the presentation begins, the device's microphone automatically starts up and collects the voice data. The input here is the voice captured in real time through the microphone, and the output is the voice data that is temporarily saved in local storage. Specifically, the voice data is saved periodically every few seconds, and this saved data is put on hold for further processing.

[0567] Step 2:

[0568] The device uses a screen capture tool to obtain image data of the slides. Each time a slide changes, the device automatically performs a screen capture and obtains image data. The input here is the slide screen obtained through the screen capture tool, and the output is image data that is temporarily saved in local storage. Specifically, each time the slide changes to a new page, an image is captured and saved.

[0569] Step 3:

[0570] The device sends the recorded audio data to the server. The input here is the audio data stored in the local storage, and the output is the audio data sent to the server. Specifically, the audio files are divided into fixed sizes and sent to the server one by one, and the server waits to receive them.

[0571] Step 4:

[0572] The server converts the received voice data into text data using a speech recognition engine such as Google Cloud Speech-to-Text. The input here is the transmitted voice data, and the output is text data. Specifically, the voice file is input into the engine, and within a few seconds, the corresponding text is generated and stored in a database.

[0573] Step 5:

[0574] The device sends the acquired slide image data to the server. The input here is the image data stored in the local storage, and the output is the image data sent to the server. Specifically, the captured image files are sent to the server one by one, and the server enters a waiting state to receive them.

[0575] Step 6:

[0576] The server analyzes the received slide image data using an image recognition engine such as Google Cloud Vision to extract key text. The input here is the transmitted slide image data, and the output is the extracted text data. Specifically, the image file is input into the engine, and the text is analyzed, extracted, and stored in a database.

[0577] Step 7:

[0578] The server integrates the results of speech recognition and image recognition and uses NLP (natural language processing) technology to perform a detailed analysis of the presentation content. The input here is text data from speech recognition and image recognition, and the output is a summary of the presentation content and key keywords. Specifically, the text data is input into the NLP model, and key points are extracted and summarized.

[0579] Step 8:

[0580] The server uses a generative AI model (e.g., GPT-3) based on the analysis results to generate questions. The input here is the text data of the analyzed presentation, and the output is the generated question text. Specifically, the analysis data is input into the AI ​​model as a prompt, and natural questions are generated.

[0581] Step 9:

[0582] The server presents the generated question to the viewer's device in real time. The input here is the generated question text, and the output is the question displayed on the viewer's device. Specifically, the question text is sent to the viewer and displayed on the viewer's screen.

[0583] Step 10:

[0584] Users (viewers) post their own questions. The input here is text data entered by the viewer, and the output is the question data sent to the server. Specifically, when the viewer enters a question in the input field on their device and presses the send button, the data is sent to the server.

[0585] Step 11:

[0586] Users (viewers) vote on the questions presented. The input here is the viewer's voting operation, and the output is the voting data sent to the server. Specifically, when a viewer presses the voting button on their device, the data is sent to the server in real time.

[0587] Step 12:

[0588] The server tally the votes from viewers in real time and determine the priorities. The input here is the voting data, and the output is a list of questions with determined priorities. Specifically, the voting results are tallyed in real time using a tallying algorithm, and priorities are assigned.

[0589] Step 13:

[0590] The server notifies the presenter of high-priority questions. The input is a prioritized list of questions, and the output is the questions to be displayed on the presenter's device. Specifically, the highest-priority question is sent to the presenter's device and displayed visually.

[0591] (Application example 1)

[0592] 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."

[0593] In conventional presentations and product demonstrations, questions are generated and presented manually, making it difficult to smoothly conduct a Q&A session. Especially in situations with a large number of viewers or customers, important questions can be overlooked, reducing viewer and customer satisfaction. Furthermore, there was no system for collecting and voting on questions from viewers and customers in real time, making it difficult to effectively convey questions of interest to the presenter or presenter. The present invention aims to solve these problems.

[0594] 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.

[0595] In this invention, the server includes means for recording audio data of the presentation, means for acquiring slide images of the presentation, means for converting the audio data into text data, means for analyzing the slide images and extracting key text, means for integrating the text data and the extracted key text to analyze the content of the presentation, means for generating questions based on the content of the presentation, means for presenting the generated questions to user terminals, means for users to submit questions, means for users to vote on the submitted questions, means for tallying up the votes and determining priorities, means for analyzing the audio and images of the product introduction demonstration and automatically generating and presenting questions from customers, and means for presenting the generated questions to user terminals in real time. This enables the Q&A session to proceed smoothly, important questions to be effectively conveyed to the presenter or introducer, and increases the interest and satisfaction of many viewers and customers.

[0596] "Means for recording audio data of a presentation" refers to a device or system for recording the presenter's voice using an audio input device.

[0597] "Means for capturing images of presentation slides" refers to a device or system for capturing images of slides during a presentation using a camera or screen capture tool.

[0598] The "means for converting the voice data into text data" refers to a system or method for converting recorded voice data into text using a voice recognition engine.

[0599] The "means for analyzing the slide images and extracting the main text" refers to a system or method for extracting text information from the acquired slide images using an image recognition engine.

[0600] "Means for integrating the above text data and the extracted main text to analyze the content of the presentation" refers to a system or method for integrating the results of speech recognition and image recognition and analyzing the content of the presentation using natural language processing technology.

[0601] A "means for generating questions based on presentation content" is a system or method for automatically generating questions from presentation content using a generative AI model.

[0602] The "means for presenting the generated question on the user's terminal" is a system or method for displaying the generated question on a device used by a viewer or customer.

[0603] "Means for users to post questions" refers to a device or function that allows viewers or customers to submit their questions to the system.

[0604] "Means for users to vote on questions presented" refers to a system or function that allows viewers or customers to vote on questions that are displayed.

[0605] The "means for aggregating vote results and determining priority" refers to a system or method for collecting voting data from users and determining the question that receives the most votes as the highest priority.

[0606] "Means for analyzing audio and video of a product introduction demonstration and automatically generating and presenting questions from customers" refers to a system or method for analyzing audio and video during a product introduction and generating and presenting questions from customers based on that analysis.

[0607] "Means for presenting generated questions to a user's device in real time" refers to a system or method for displaying generated questions on the spot and instantly on a viewer's or customer's device.

[0608] System Overview

[0609] This invention is a system that analyzes the audio and video of a product introduction demonstration and automatically generates and presents questions to customers, aiming to ensure a smooth Q&A session and satisfy both the customer and the introducer.

[0610] System configuration

[0611] Terminal

[0612] The terminal performs the following functions during the product demonstration:

[0613] Audio recording: Record the introducer's voice using the device's microphone.

[0614] Image capture: Uses the device's camera to capture images during the demonstration.

[0615] server

[0616] The server does the following:

[0617] Speech recognition: Receives voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text).

[0618] Image Analysis: Receive the captured image and extract the key text using Tesseract OCR.

[0619] Content analysis: The results of voice recognition and image recognition are integrated, and the introduction content is analyzed using natural language processing technology.

[0620] Question generation: Based on the analysis results, questions are generated using a generative AI model (e.g., OpenAI's GPT-3).

[0621] Question presentation: The generated questions are presented to the user's device in real time.

[0622] Voting function: Provides a mechanism for users to vote on questions posted by users.

[0623] Prioritization: Tally the votes and present the highest priority questions to the introducer.

[0624] Specific processing examples

[0625] Acquiring and processing audio data

[0626] The device uses a microphone to record the introducer's voice and sends the voice data to the server, which then converts the voice data into text data using Google Speech-to-Text. For example, a statement such as "This new smartphone has an innovative camera function" can be recognized.

[0627] Slide image acquisition and processing

[0628] The device uses its camera to capture an image of the demonstration and sends it to a server, which uses Tesseract OCR to extract key text from the image, such as "Advanced night mode photography with AI."

[0629] Content analysis and question generation

[0630] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. Based on this analysis, a prompt is created to be provided to the generative AI model (GPT-3), which then generates a question. For example, the prompt is as follows:

[0631] Example of question generation

[0632] "Generate customer questions based on the following text: Voiceover: "This new smartphone comes with revolutionary camera features." Slide: "Advanced night mode photography powered by AI.""

[0633] An example of a question that can be generated is:

[0634] "Can you tell me more about the night mode photography feature on this phone? I'm especially interested to know how AI helps with that."

[0635] is.

[0636] Viewer questions and polls

[0637] Users (viewers or customers) can post their own questions from their devices. For example, a question such as, "I'm interested in how AI will affect the filming process." Similarly, viewers can also vote on questions presented. The results of these votes are tallied in real time by the server.

[0638] Prioritizing and presenting questions

[0639] The server tally the votes from the viewers and notify the introducer of the question that received the most votes. For example, if "Please explain the details of the night mode shooting function" received the most votes, this will be displayed to the introducer.

[0640] System Effects

[0641] This system facilitates Q&A sessions during product demonstrations, allowing customers to actively participate based on their interests. It also makes it easier for introducers to understand customer interests through questions, resulting in high-quality demonstrations and interactive communication.

[0642] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0643] Step 1:

[0644] The terminal uses a microphone to record the voice during the product introduction demonstration. This recorded voice data is sent to the server. The input is the recorded voice data, and the output is the voice data sent to the server.

[0645] Step 2:

[0646] The server converts the received voice data into text data using the Google Speech-to-Text API. The input is the voice data, and the output is the converted text data. As part of this conversion process, the server makes an API call and obtains the results.

[0647] Step 3:

[0648] The device uses a camera to capture images of slides during the demonstration. This captured image data is also sent to the server. The input is the captured image data, and the output is the image data sent to the server.

[0649] Step 4:

[0650] The server analyzes the received image data using Tesseract OCR and extracts key character information. The input is image data, and the output is extracted text information. As part of this image analysis process, the server passes the image data to the OCR engine and returns the analysis results as text.

[0651] Step 5:

[0652] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. The input is text data of the speech and text information of the image, and the output is the analyzed introduction. The server combines the text data to extract important keywords and context.

[0653] Step 6:

[0654] The server generates a question using a generative AI model (GPT-3) based on the analysis results. The input is the analyzed introduction content, and the output is the generated question. In this procedure, the server sends a prompt to the generative AI model to generate a question.

[0655] Step 7:

[0656] The server presents the generated questions to the user's terminal in real time. The input is the generated question, and the output is the question displayed on the user's terminal. The server sends data to each terminal and displays it on the screen.

[0657] Step 8:

[0658] Users can post their own questions from their terminals. The input is the question entered by the user, and the output is the question sent to the server. The terminal receives the user's input and sends it to the server.

[0659] Step 9:

[0660] Users can vote on presented questions. The input is the vote selected by the user, and the output is the voting data sent to the server. The user's terminal sends the voting information for the selected question.

[0661] Step 10:

[0662] The server tally the votes and decides the priority questions. The input is the voting data and the output is the priority questions. The server tally each vote and selects the question with the most votes.

[0663] Step 11:

[0664] The server notifies the introducer's terminal of high-priority questions. The input is the high-priority questions, and the output is the questions displayed on the introducer's terminal. The server sends the selected questions to the introducer in real time and displays them on the screen.

[0665] 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.

[0666] System Overview

[0667] This invention provides a presentation support system that combines an emotion engine that recognizes the user's emotions. The system analyzes the presentation's audio data and slide images, automatically generates and presents questions, and combines this with a means to adjust the priority of questions using emotion data, making the Q&A session more effective and engaging.

[0668] System configuration

[0669] 1. Terminal

[0670] During the presentation, the device will record the presenter's voice using a microphone and capture images of the slides using a camera or screen capture tool.

[0671] 2. Server

[0672] Voice data is received and converted into text using a voice recognition engine.

[0673] The acquired slide images are received and text is extracted using an image recognition engine.

[0674] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[0675] Questions are generated using generative AI based on the content of the presentation.

[0676] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[0677] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[0678] 3. Emotion Engine

[0679] It recognizes emotions from viewers' facial expressions and tone of voice, and sends the data to the server in real time.

[0680] Dynamically adjust question priorities based on sentiment data.

[0681] Specific processing examples

[0682] Acquiring and processing audio data

[0683] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[0684] Slide image acquisition and processing

[0685] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0686] Analysis of presentation content

[0687] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[0688] Question Generation

[0689] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0690] Audience emotion recognition

[0691] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. For example, if the viewer has an interested look on their face, that data is used to prioritize questions.

[0692] Viewer questions and polls

[0693] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[0694] Prioritizing and presenting questions

[0695] The server then selects the highest-priority questions based on audience votes and sentiment data and notifies the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, it will be displayed to the presenter.

[0696] System Effects

[0697] This system makes Q&A sessions during presentations more effective and engaging. Audience members can actively participate based on their interests, and presenters can answer questions while understanding the audience's emotions, enabling deeper communication. The result is higher-quality presentations and more interactive engagement.

[0698] The processing flow will be explained below.

[0699] Step 1:

[0700] As soon as the presentation begins, the device will use the microphone to record the presenter's voice and the camera or screen capture tool to capture images of the slides.

[0701] Step 2:

[0702] The terminal transmits the recorded audio data and the acquired slide images to the server at regular intervals.

[0703] Step 3:

[0704] The server passes the received voice data to a speech recognition engine, which converts the voice data into text data. For example, a statement such as "AI will become mainstream in many industries over the next decade" is converted into text data.

[0705] Step 4:

[0706] The server passes the received slide image data to an image recognition engine, which extracts key text and figures from the slide. For example, text such as "AI applications: medical, education, and transportation" is extracted.

[0707] Step 5:

[0708] The server combines the text data generated by speech recognition with the key text extracted by image recognition, and uses NLP technology to analyze the presentation and extract key points and keywords.

[0709] Step 6:

[0710] Based on the analysis results, the server uses generative AI to generate questions such as, "Which field do you think will see the most dramatic changes with the application of AI?" These questions are then displayed in real time on the viewer's device.

[0711] Step 7:

[0712] The emotion engine analyzes viewers' facial expressions and vocal tone in real time, using cameras and microphones to capture viewers' emotional data (e.g., interest, surprise, joy) while they watch.

[0713] Step 8:

[0714] The emotion engine sends the acquired emotion data to the server, and the priority of questions is dynamically adjusted based on that data. For example, questions that show a lot of interest to viewers can be given higher priority.

[0715] Step 9:

[0716] Users (viewers) can post their own questions from their devices. For example, they can post a question such as, "I would like to know more about the risks of AI." The device then sends this question to the server.

[0717] Step 10:

[0718] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[0719] Step 11:

[0720] The server aggregates all questions and votes and uses audience sentiment data to determine priority. For example, if the question "How will AI revolutionize education?" receives a lot of votes and emotional responses from viewers, it will be prioritized.

[0721] Step 12:

[0722] The server then notifies the presenter of the selected high-priority questions, allowing the presenter to conduct a Q&A session based on these questions. For example, by prioritizing answers to questions that evoke strong emotional responses, the presenter can increase audience interest.

[0723] Through these processing steps, the Q&A session during the presentation can be conducted more effectively and attractively, providing a satisfying environment for both the presenter and the audience.

[0724] Example 2

[0725] 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."

[0726] Conventional presentation support systems have the problem that the priority of questions received by the presenter is not optimized based on the audience's interests and emotions, making it difficult to effectively conduct a Q&A session. In addition, there is a lack of a way to present questions without distinguishing between questions from the audience and questions generated by the AI ​​generation system, making it difficult to realize a Q&A session that reflects the audience's interests.

[0727] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for converting audio data into text data, means for analyzing slide images and extracting key text, means for generating questions using a generative AI model, means for using an emotion engine that analyzes viewers' facial expressions and tone of voice, and means for dynamically adjusting the priority of questions based on emotion data. This enables the progress of an effective Q&A session that reflects viewers' interests and emotions.

[0728] "Means for recording audio data of a presentation" refers to a device or method for collecting the presenter's remarks as digital data using audio equipment such as a microphone.

[0729] "Means for acquiring slide images of presentations" refers to a device or method for capturing images of slides used during presentations and collecting them as digital image data.

[0730] "Means for converting voice data into text data" refers to software or a system that uses voice recognition technology to convert recorded voice data into text data.

[0731] "Means for analyzing slide images and extracting key text" refers to software or a system that extracts text information from slide images acquired using image recognition technology.

[0732] "Means for integrating text data and extracted key texts to analyze presentation content" refers to software or systems that integrate collected text data and use natural language processing technology to understand and analyze presentation content.

[0733] "Means using a generative AI model that generates questions based on the content of a presentation" refers to an artificial intelligence model and method that automatically generates questions based on the content of a presentation analyzed using natural language generation technology.

[0734] The "means for presenting the generated questions to the viewer terminal" refers to software or communication technology for displaying the questions generated by the server on the device used by the viewer.

[0735] "Means for viewers to post questions and vote on submitted questions" means an interface or system through which viewers can submit their own questions and rate other submitted questions.

[0736] The "means for aggregating vote results and determining question priority" refers to an algorithm or system that aggregates vote data from viewers and determines the importance of each question based on the results.

[0737] "Means using an emotion engine that analyzes the viewer's facial expressions and tone of voice" refers to software or equipment for analyzing the viewer's facial expressions and tone of voice and extracting emotion data.

[0738] "Means for dynamically adjusting question priority based on emotional data" refers to an algorithm or system that changes the display order and importance of questions in real time, taking into account the acquired emotional data.

[0739] The present invention relates to a presentation support system that optimizes a question and answer session using user emotion data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following describes in detail the preferred embodiments of the present invention.

[0740] First, we will describe the main embodiment of the system, which uses hardware and software to collect and analyze audio data, image data, and emotional data to effectively facilitate a question-and-answer session between the presenter and the audience.

[0741] Hardware and software used

[0742] The device includes a microphone, camera, and screen capture tool. These devices are used to collect the presenter's voice and slide images in real time. For example, when a presenter explains "Medical Applications of AI Technology," the device records the voice with the microphone and captures slide images with the screen capture tool.

[0743] The server is a central system for analyzing and integrating the collected data. It receives the audio data and slide image data sent from the devices via an internet connection and processes them as follows:

[0744] 1. Speech recognition: Using a speech recognition engine such as Google Cloud Speech-to-Text, the audio data is converted into text data. For example, a speaker's words such as "AI technology is being used in the medical field" are obtained as text data.

[0745] 2. Image Recognition: Using an image recognition engine such as Google Cloud Vision, key text is extracted from the slide image, specifically the slide title "Applications of AI in Medicine."

[0746] 3. Analysis of presentation content: The acquired text data is integrated and the presentation content is analyzed using natural language processing (NLP) technology. Here, the key points and key keywords of the presentation are identified. For example, key points such as "This presentation explains different application areas of AI" are extracted.

[0747] 4. Question generation: Based on the analysis results, questions are generated using generative AI (such as OpenAI's GPT-4). The generated questions are displayed in real time on the viewer's device. For example, a question such as, "What specific changes can we expect to see in the medical field through the use of AI?"

[0748] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotion data to the server. For example, if the viewer has an interested expression, the data will be collected by the emotion engine and sent to the server.

[0749] Prompt Sentence Examples

[0750] "Generate a question based on the following presentation: AI has achieved remarkable results in the field of medical imaging diagnosis. In particular, AI algorithms are now able to detect cellular abnormalities with high accuracy. This has significantly improved the speed and accuracy of diagnoses and reduced the burden on doctors."

[0751] This allows viewers to post questions based on their own interests and also vote on questions generated by other viewers. The server dynamically prioritizes questions based on the collected voting and sentiment data, and provides the presenter with the most relevant questions.

[0752] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0753] Processing flow and step details

[0754] Step 1: Acquiring and sending audio data

[0755] The device records the presenter's voice with a microphone and sends the audio data to the server. Specifically, the audio during the presentation is recorded in real time and converted into a digital audio file. This audio file is then sent to the server via the Internet.

[0756] Input: Speaker's voice

[0757] Data processing: audio recording and digital conversion

[0758] Output: Digital audio file (sent to server)

[0759] Step 2: Capture and send slide images

[0760] The device uses a screen capture tool to capture image data of the slides and sends it to the server. When the presenter switches slides, the device automatically performs a screen capture and sends the captured image to the server.

[0761] Input: Slide screen

[0762] Data processing: Screen capture

[0763] Output: Image data file (sent to server)

[0764] Step 3: Convert audio data to text

[0765] The server receives the voice data and converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The digital voice file sent to the server is analyzed and converted by the speech recognition engine and saved as a string of characters.

[0766] Input: Digital audio data

[0767] Data processing: Speech recognition (voice to text conversion)

[0768] Output: Text data

[0769] Step 4: Extract text from slide images

[0770] The server extracts key text from the slide images using an image recognition engine (e.g., Google Cloud Vision). It analyzes the received image data files to identify and extract text within the images.

[0771] Input: Slide image data

[0772] Data processing: Image recognition (text extraction)

[0773] Output: Text data

[0774] Step 5: Analyze the presentation

[0775] The server integrates the text data obtained from the speech and image recognition and analyzes the presentation using natural language processing (NLP) technology. Based on the integrated text data, the main points and key keywords of the presentation are identified.

[0776] Input: Text data extracted from audio and slides

[0777] Data Processing: Integration and NLP Analysis

[0778] Output: Gist and keyword data

[0779] Step 6: Generate questions

[0780] The server uses a generative AI model (e.g., GPT-4) to generate questions based on the analysis results. The server uses the key points and keywords obtained as prompts to generate appropriate questions.

[0781] Input: Key points and keyword data

[0782] Data processing: Question generation (generative AI model)

[0783] Output: Question text

[0784] Step 7: Posing the Question

[0785] The server presents the generated questions to the viewer's terminal in real time, and transmits the generated questions to the viewer's device and provides them through an interface for display.

[0786] Input: Question text

[0787] Data processing: Submitting and presenting questions

[0788] Output: Question displayed on the viewer's device

[0789] Step 8: Collect and send emotion data

[0790] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. The camera and microphone collect the viewer's emotions and send them to the server as digital data.

[0791] Input: Viewer facial expressions and tone of voice

[0792] Data processing: Sentiment analysis

[0793] Output: Emotion data (sent to server)

[0794] Step 9: Viewer polls

[0795] Users (viewers) vote on questions presented through their terminals. Each viewer votes for the questions they are interested in, and the vote data is sent to the server in real time.

[0796] Input: Viewer votes

[0797] Data processing: Collecting voting data

[0798] Output: Voting data (sent to server)

[0799] Step 10: Prioritize and present questions

[0800] The server dynamically determines the priority of questions based on the collected voting data and sentiment data, and selects the most important question and displays it on the presenter's device.

[0801] Input: Voting data and sentiment data

[0802] Data processing: data integration and prioritization

[0803] Output: Top priority question (displayed to presenter's device)

[0804] The above processing steps enable an effective and engaging Q&A session between the presenter and the audience.

[0805] (Application example 2)

[0806] 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."

[0807] Conventional presentation support systems and robot work support systems have problems in that they are unable to fully realize Q&A during presentations or improve the efficiency of robot work in factories. Specifically, they do not properly manage the priority of audience questions, and do not immediately adjust the robot's behavior in response to the worker's instructions or emotions, hindering efficient work. This results in low satisfaction for viewers and workers, and a lack of two-way communication.

[0808] 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.

[0809] In this invention, the server includes means for generating questions based on the content of the presentation, means for presenting the generated questions to viewer terminals, means for viewers to post questions, means for viewers to vote on the presented questions, means for tallying up vote results and determining priorities, means for acquiring robot work status data, means for acquiring voice instructions from workers and dynamically optimizing work content using a generation AI, and means for acquiring worker emotion data and adjusting work plans based on the emotion data. This improves the efficiency of question and answer sessions between viewers and presenters, enabling more efficient robot work in factories.

[0810] "Presentation audio data" refers to recorded data of the audio uttered by the presenter during the presentation.

[0811] "Slide images" are image data of visual slides used during a presentation.

[0812] A "voice recognition engine" is software or hardware for converting voice data into text data.

[0813] An "image recognition engine" is software or hardware for extracting key text from image data.

[0814] "Integration" is the process of combining multiple different data sets into a single dataset.

[0815] "Question generation" is the process of automatically creating appropriate questions based on the presentation content and analysis results.

[0816] "Viewer device" means the device used by the viewer to view the presentation.

[0817] "Voting results" are the aggregated results of votes cast by viewers on questions.

[0818] "Priority" is the measure by which a particular question or task is judged to be more important than other questions or tasks.

[0819] "Robot work status data" refers to data relating to the work currently being performed by the robot and its status.

[0820] "Generative AI" is artificial intelligence that generates appropriate questions and answers based on given data and conditions.

[0821] "Emotion data" refers to emotional information obtained from the worker's facial expressions, tone of voice, etc.

[0822] "Dynamic optimization" is the process of adjusting to the optimal work plan in real time as conditions change.

[0823] A "work plan" is the design of procedures and schedules for carrying out a specific task.

[0824] This invention relates to a "robot work support system" for improving the efficiency of robot work in factories. This system combines a function to support questions and answers during presentations with a function to dynamically optimize robot work in factories, thereby increasing the satisfaction of viewers and workers.

[0825] System Overview

[0826] Terminal

[0827] The terminal uses the microphone and camera built into the robot body to capture the worker's voice instructions and work status in real time. The smart glasses or tablet used by the worker also function as part of the system.

[0828] server

[0829] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the acquired voice data into text data, an image recognition engine (e.g., Amazon Rekognition) to analyze image data acquired from the robot's camera, and an NLP technology (e.g., spaCy) to integrate the results of speech recognition and image recognition, and a generative AI (e.g., OpenAI GPT) to generate appropriate questions and work plans.

[0830] Emotion Engine

[0831] An emotion engine (e.g., Microsoft Azure Emotion API) is used to analyze workers' facial expressions and voice tones in real time to obtain emotional data, which is used to dynamically adjust the work plan.

[0832] Specific processing examples

[0833] 1. Acquiring and processing audio data

[0834] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits them to the server. The server then converts the voice data into text data using a voice recognition engine. During this process, an instruction such as "Please install the next part" is obtained as text data.

[0835] 2. Image data acquisition and processing

[0836] The terminal uses the robot's built-in camera to capture images of the work status and parts, and sends them to the server. The server then analyzes the image data using an image recognition engine to extract key information. For example, specific parts and work status may be recognized.

[0837] 3. Analysis and adjustment of work content

[0838] The server integrates and analyzes the voice and image recognition results, and uses generative AI to generate questions and work plans, which are then presented to the worker's smart glasses or tablet.

[0839] 4. Acquiring and Processing Emotional Data

[0840] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, capturing emotional data. For example, if the worker is feeling dissatisfied, the data is sent to the server in real time and used to adjust the work plan.

[0841] Prompt Sentence Examples

[0842] "When an AI robot follows instructions in a factory task, suggest ways to optimize the task. Voice command: 'Install the next part.' Worker emotion: Frustration evident."

[0843] This allows for efficient question and answer sessions between the audience and the presenter, while also dynamically optimizing the work of robots in factories, improving worker satisfaction and work efficiency.

[0844] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0845] Step 1:

[0846] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits the voice data to the server. The input is the worker's voice instructions, and the output is the voice data sent to the server. Specifically, the microphone captures the voice, which is then saved and transmitted as digital data.

[0847] Step 2:

[0848] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data, and the output is text data corresponding to the voice instructions. Specifically, the speech recognition engine analyzes the voice data and generates text using a language model.

[0849] Step 3:

[0850] The terminal uses the robot's built-in camera to acquire image data of the work situation and sends that image data to the server. The input is visual information of the work situation, and the output is image data sent to the server. Specifically, the camera captures the image, saves it as digital data, and sends it.

[0851] Step 4:

[0852] The server uses an image recognition engine to analyze the received image data and extract key information. The input is the image data, and the output is extracted key text and recognized objects. Specifically, the image recognition engine analyzes the image data pixel by pixel to identify key text and objects.

[0853] Step 5:

[0854] The server integrates the results of speech recognition and image recognition and analyzes the work content using NLP technology. The input is text data and extracted key information, and the output is the gist of the analyzed work content. Specifically, the text data and image data are integrated into a single dataset and analyzed using an NLP model.

[0855] Step 6:

[0856] The server uses a generation AI to generate appropriate questions and work plans based on the analysis results. The input is the gist of the analyzed work content, and the output is the generated questions and work plans. Specifically, the generation AI generates questions and work plans related to the prompt.

[0857] Step 7:

[0858] The terminal presents the generated questions and work plans to the worker's smart glasses or tablet. The input is the generated questions and work plans, and the output is a visual display. The specific operation is to display the information on the device's display.

[0859] Step 8:

[0860] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, and acquires emotion data. The input is video data of the worker's facial expressions, and the output is the emotion recognition results. Specifically, the camera captures the facial expressions, and the emotion recognition engine analyzes the video data.

[0861] Step 9:

[0862] The server dynamically adjusts the work plan based on the acquired emotion data. The input is emotion data, and the output is an adjusted work plan. Specifically, it analyzes the emotion recognition results and modifies the priority and procedures of the work plan in real time.

[0863] 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.

[0864] 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.

[0865] 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.

[0866] [Third embodiment]

[0867] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0868] 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.

[0869] 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).

[0870] 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.

[0871] 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.

[0872] 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).

[0873] 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.

[0874] 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.

[0875] 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.

[0876] 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.

[0877] 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.

[0878] 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."

[0879] System Overview

[0880] The invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to ensure that the Q&A session proceeds smoothly and satisfies both the presenter and the audience.

[0881] System configuration

[0882] 1. Terminal

[0883] During the presentation, the device uses a microphone to record the presenter's voice.

[0884] At the same time, capture an image of the slide using a camera or a screen capture tool.

[0885] 2. Server

[0886] Voice data is received and converted into text using a voice recognition engine.

[0887] The acquired slide images are received and text is extracted using an image recognition engine.

[0888] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[0889] Questions are generated using generative AI based on the content of the presentation.

[0890] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[0891] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[0892] Specific processing examples

[0893] Acquiring and processing audio data

[0894] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[0895] Slide image acquisition and processing

[0896] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0897] Analysis of presentation content

[0898] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[0899] Question Generation

[0900] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0901] Viewer questions and polls

[0902] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[0903] Prioritizing and presenting questions

[0904] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[0905] System Effects

[0906] This system allows the Q&A session to proceed smoothly and create a lively atmosphere. The audience can actively participate based on their interests, and the presenter can easily understand the audience's interests through questions. The result is high-quality presentations and interactive communication.

[0907] The processing flow will be explained below.

[0908] Step 1:

[0909] As soon as the presentation begins, the device uses a microphone to record the presenter's voice and also captures images of the slides using a camera or screen capture tool.

[0910] Step 2:

[0911] The device transmits the recorded audio data to the server at regular intervals, and simultaneously transmits the acquired slide image data to the server.

[0912] Step 3:

[0913] The server then runs the received voice data through a speech recognition engine and converts it into text data, such as "AI will become mainstream in many industries over the next decade."

[0914] Step 4:

[0915] The server then runs the received slide image data through an image recognition engine to extract key text and figures from the slides. For example, text such as "AI applications: medical, education, and transportation" may be extracted.

[0916] Step 5:

[0917] The server combines the text data generated by the speech recognition engine with the key text extracted by the image recognition engine. Using this combined data, NLP technology is used to analyze the presentation content and extract key points and keywords.

[0918] Step 6:

[0919] The server uses generative AI to generate questions based on the analysis results, such as, "Which field do you think will see the most dramatic changes with the application of AI?"

[0920] Step 7:

[0921] The server presents the generated questions to the viewer's terminal in real time, and the terminal displays the received questions on the viewer's screen.

[0922] Step 8:

[0923] The user (viewer) uses the application's question submission form to enter and submit their question. For example, a question such as "Can this method be applied to other cases?" is submitted. The device then sends this question to the server.

[0924] Step 9:

[0925] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[0926] Step 10:

[0927] The server will tally all the questions and votes it receives and select the question with the most votes, for example, "How will AI revolutionize education?"

[0928] Step 11:

[0929] The server notifies the presenter of the selected high-priority questions, and the presenter can then conduct a Q&A session based on these questions.

[0930] These processing steps ensure that the Q&A session following the presentation proceeds smoothly, providing a satisfying environment for both the presenter and the audience.

[0931] Example 1

[0932] 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."

[0933] In conventional Q&A sessions, there were problems such as the complicated process for viewers to submit questions and the difficulty for presenters to grasp the audience's interests. Furthermore, when a large number of questions are submitted, it becomes difficult to select the most important questions. In such cases, the limited time for Q&A often results in situations where many viewers are unable to receive adequate answers.

[0934] 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.

[0935] In this invention, the server includes a means for generating questions using a generative model based on the content of the presentation, a means for presenting the generated questions to the viewer's terminal, and a means for aggregating voting results in real time and determining priorities. This allows viewers to quickly and easily post questions and select important questions through voting. As a result, the Q&A session proceeds efficiently, and two-way communication between the presenter and the audience is facilitated.

[0936] "Presentation audio data" refers to data that includes audio information of a speaker in a presentation, lecture, or other presentation.

[0937] A "slide image" is image data that includes visual materials and charts used by a presenter during a presentation.

[0938] "Text data" refers to data that includes text information extracted from audio data or image data.

[0939] "Main text" is text data that includes important sentence information and keywords extracted from the slide image.

[0940] "Integration" refers to the process of combining multiple pieces of data or information and organizing them into a single data set or piece of information.

[0941] "Analysis" is the process of examining data or information in detail to make its structure and meaning easier to understand.

[0942] A "generative model" is an algorithm that automatically creates new data and information using machine learning techniques such as neural networks.

[0943] "Means for generating questions" refers to technology for automatically generating questions based on the content of a presentation.

[0944] "Means for presenting on the viewer's device" refers to the technology for displaying the generated questions and information on the screen of the viewer's device.

[0945] "Means for viewers to submit questions" refers to the technology by which viewers submit their questions to the system.

[0946] "Means for viewers to vote on questions presented" refers to technology that allows viewers to rate or select questions displayed on the screen.

[0947] "Means for aggregating voting results in real time" refers to technology for instantly aggregating voting data from viewers and determining rankings and priorities.

[0948] "Means for determining priorities" refers to a technique for determining which questions are most important based on voting results and determining their order.

[0949] "Means of notification" refers to techniques for conveying specific information to a target audience, such as a presenter.

[0950] This invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to smoothly conduct Q&A sessions during presentations and to satisfy both the presenter and the audience.

[0951] System hardware and software configuration

[0952] Terminal

[0953] The terminal is a device for recording the presenter's voice and acquiring slide images. The terminal has the following functions:

[0954] Microphone: Records the presenter's voice in real time.

[0955] Camera or screen capture tool: Take an image of the slide.

[0956] The terminal may be a general-purpose computing device such as a laptop, tablet, or smartphone.

[0957] server

[0958] The server is the core component that analyzes the audio data and slide image data, and generates and presents questions. The specific software and functions used are as follows:

[0959] Speech recognition engine: Converts voice data into text data using Google Cloud Speech-to-Text or similar.

[0960] Image recognition engine: Extract text from slide images using Google Cloud Vision or similar.

[0961] NLP technology: Using natural language processing technology, voice recognition and image recognition results are analyzed to extract key points and keywords.

[0962] Generative AI model: Using OpenAI's GPT-3 and other models, questions are generated based on the analysis results.

[0963] Real-time tallying function: Tally the votes from the audience and notify the presenter of the highest priority questions.

[0964] User

[0965] Users (viewers) can post questions and vote on questions provided from their own devices.

[0966] Specific examples of program processing

[0967] Acquiring and processing audio data

[0968] The device records the speaker's voice and sends the audio data to a server. The server then converts the audio data into text using Google Cloud Speech-to-Text. For example, a statement such as "AI has a wide range of applications" in a presentation can be converted into text.

[0969] Slide image acquisition and processing

[0970] The device uses a screen capture tool to capture an image of the slide and sends it to the server, which then uses Google Cloud Vision to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[0971] Analysis of presentation content

[0972] The server combines the received voice and image recognition results and uses NLP technology to analyze the content of the presentation, providing key points such as, "This presentation explains different areas in which AI can be applied."

[0973] Question Generation

[0974] Based on the analysis results, the server uses a generative AI model to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[0975] Viewer questions and polls

[0976] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Viewers can also vote on the questions presented, and the results of these votes are tallied in real time by the server.

[0977] Prioritizing and presenting questions

[0978] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[0979] Prompt Sentence Examples

[0980] Generate questions based on the given presentation using the following criteria:

[0981] Announcement: "AI has a wide range of applications. It is being used in a wide variety of fields, from medicine to education and even entertainment."

[0982] Generating conditions: Ask for specific examples and numerical questions.

[0983] This system facilitates smooth Q&A sessions following presentations and allows for two-way communication between presenters and audience members.

[0984] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0985] Step 1:

[0986] The device records the presenter's voice. When the presentation begins, the device's microphone automatically starts up and collects the voice data. The input here is the voice captured in real time through the microphone, and the output is the voice data that is temporarily saved in local storage. Specifically, the voice data is saved periodically every few seconds, and this saved data is put on hold for further processing.

[0987] Step 2:

[0988] The device uses a screen capture tool to obtain image data of the slides. Each time a slide changes, the device automatically performs a screen capture and obtains image data. The input here is the slide screen obtained through the screen capture tool, and the output is image data that is temporarily saved in local storage. Specifically, each time the slide changes to a new page, an image is captured and saved.

[0989] Step 3:

[0990] The device sends the recorded audio data to the server. The input here is the audio data stored in the local storage, and the output is the audio data sent to the server. Specifically, the audio files are divided into fixed sizes and sent to the server one by one, and the server waits to receive them.

[0991] Step 4:

[0992] The server converts the received voice data into text data using a speech recognition engine such as Google Cloud Speech-to-Text. The input here is the transmitted voice data, and the output is text data. Specifically, the voice file is input into the engine, and within a few seconds, the corresponding text is generated and stored in a database.

[0993] Step 5:

[0994] The device sends the acquired slide image data to the server. The input here is the image data stored in the local storage, and the output is the image data sent to the server. Specifically, the captured image files are sent to the server one by one, and the server enters a waiting state to receive them.

[0995] Step 6:

[0996] The server analyzes the received slide image data using an image recognition engine such as Google Cloud Vision to extract key text. The input here is the transmitted slide image data, and the output is the extracted text data. Specifically, the image file is input into the engine, and the text is analyzed, extracted, and stored in a database.

[0997] Step 7:

[0998] The server integrates the results of speech recognition and image recognition and uses NLP (natural language processing) technology to perform a detailed analysis of the presentation content. The input here is text data from speech recognition and image recognition, and the output is a summary of the presentation content and key keywords. Specifically, the text data is input into the NLP model, and key points are extracted and summarized.

[0999] Step 8:

[1000] The server uses a generative AI model (e.g., GPT-3) based on the analysis results to generate questions. The input here is the text data of the analyzed presentation, and the output is the generated question text. Specifically, the analysis data is input into the AI ​​model as a prompt, and natural questions are generated.

[1001] Step 9:

[1002] The server presents the generated question to the viewer's device in real time. The input here is the generated question text, and the output is the question displayed on the viewer's device. Specifically, the question text is sent to the viewer and displayed on the viewer's screen.

[1003] Step 10:

[1004] Users (viewers) post their own questions. The input here is text data entered by the viewer, and the output is the question data sent to the server. Specifically, when the viewer enters a question in the input field on their device and presses the send button, the data is sent to the server.

[1005] Step 11:

[1006] Users (viewers) vote on the questions presented. The input here is the viewer's voting operation, and the output is the voting data sent to the server. Specifically, when a viewer presses the voting button on their device, the data is sent to the server in real time.

[1007] Step 12:

[1008] The server tally the votes from viewers in real time and determine the priorities. The input here is the voting data, and the output is a list of questions with determined priorities. Specifically, the voting results are tallyed in real time using a tallying algorithm, and priorities are assigned.

[1009] Step 13:

[1010] The server notifies the presenter of high-priority questions. The input is a prioritized list of questions, and the output is the questions to be displayed on the presenter's device. Specifically, the highest-priority question is sent to the presenter's device and displayed visually.

[1011] (Application example 1)

[1012] 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."

[1013] In conventional presentations and product demonstrations, questions are generated and presented manually, making it difficult to smoothly conduct a Q&A session. Especially in situations with a large number of viewers or customers, important questions can be overlooked, reducing viewer and customer satisfaction. Furthermore, there was no system for collecting and voting on questions from viewers and customers in real time, making it difficult to effectively convey questions of interest to the presenter or presenter. The present invention aims to solve these problems.

[1014] 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.

[1015] In this invention, the server includes means for recording audio data of the presentation, means for acquiring slide images of the presentation, means for converting the audio data into text data, means for analyzing the slide images and extracting key text, means for integrating the text data and the extracted key text to analyze the content of the presentation, means for generating questions based on the content of the presentation, means for presenting the generated questions to user terminals, means for users to submit questions, means for users to vote on the submitted questions, means for tallying up the votes and determining priorities, means for analyzing the audio and images of the product introduction demonstration and automatically generating and presenting questions from customers, and means for presenting the generated questions to user terminals in real time. This enables the Q&A session to proceed smoothly, important questions to be effectively conveyed to the presenter or introducer, and increases the interest and satisfaction of many viewers and customers.

[1016] "Means for recording audio data of a presentation" refers to a device or system for recording the presenter's voice using an audio input device.

[1017] "Means for capturing images of presentation slides" refers to a device or system for capturing images of slides during a presentation using a camera or screen capture tool.

[1018] The "means for converting the voice data into text data" refers to a system or method for converting recorded voice data into text using a voice recognition engine.

[1019] The "means for analyzing the slide images and extracting the main text" refers to a system or method for extracting text information from the acquired slide images using an image recognition engine.

[1020] "Means for integrating the above text data and the extracted main text to analyze the content of the presentation" refers to a system or method for integrating the results of speech recognition and image recognition and analyzing the content of the presentation using natural language processing technology.

[1021] A "means for generating questions based on presentation content" is a system or method for automatically generating questions from presentation content using a generative AI model.

[1022] The "means for presenting the generated question on the user's terminal" is a system or method for displaying the generated question on a device used by a viewer or customer.

[1023] "Means for users to post questions" refers to a device or function that allows viewers or customers to submit their questions to the system.

[1024] "Means for users to vote on questions presented" refers to a system or function that allows viewers or customers to vote on questions that are displayed.

[1025] The "means for aggregating vote results and determining priority" refers to a system or method for collecting voting data from users and determining the question that receives the most votes as the highest priority.

[1026] "Means for analyzing audio and video of a product introduction demonstration and automatically generating and presenting questions from customers" refers to a system or method for analyzing audio and video during a product introduction and generating and presenting questions from customers based on that analysis.

[1027] "Means for presenting generated questions to a user's device in real time" refers to a system or method for displaying generated questions on the spot and instantly on a viewer's or customer's device.

[1028] System Overview

[1029] This invention is a system that analyzes the audio and video of a product introduction demonstration and automatically generates and presents questions to customers, aiming to ensure a smooth Q&A session and satisfy both the customer and the introducer.

[1030] System configuration

[1031] Terminal

[1032] The terminal performs the following functions during the product demonstration:

[1033] Audio recording: Record the introducer's voice using the device's microphone.

[1034] Image capture: Uses the device's camera to capture images during the demonstration.

[1035] server

[1036] The server does the following:

[1037] Speech recognition: Receives voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text).

[1038] Image Analysis: Receive the captured image and extract the key text using Tesseract OCR.

[1039] Content analysis: The results of voice recognition and image recognition are integrated, and the introduction content is analyzed using natural language processing technology.

[1040] Question generation: Based on the analysis results, questions are generated using a generative AI model (e.g., OpenAI's GPT-3).

[1041] Question presentation: The generated questions are presented to the user's device in real time.

[1042] Voting function: Provides a mechanism for users to vote on questions posted by users.

[1043] Prioritization: Tally the votes and present the highest priority questions to the introducer.

[1044] Specific processing examples

[1045] Acquiring and processing audio data

[1046] The device uses a microphone to record the introducer's voice and sends the voice data to the server, which then converts the voice data into text data using Google Speech-to-Text. For example, a statement such as "This new smartphone has an innovative camera function" can be recognized.

[1047] Slide image acquisition and processing

[1048] The device uses its camera to capture an image of the demonstration and sends it to a server, which uses Tesseract OCR to extract key text from the image, such as "Advanced night mode photography with AI."

[1049] Content analysis and question generation

[1050] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. Based on this analysis, a prompt is created to be provided to the generative AI model (GPT-3), which then generates a question. For example, the prompt is as follows:

[1051] Example of question generation

[1052] "Generate customer questions based on the following text: Voiceover: "This new smartphone comes with revolutionary camera features." Slide: "Advanced night mode photography powered by AI.""

[1053] An example of a question that can be generated is:

[1054] "Can you tell me more about the night mode photography feature on this phone? I'm especially interested to know how AI helps with that."

[1055] is.

[1056] Viewer questions and polls

[1057] Users (viewers or customers) can post their own questions from their devices. For example, a question such as, "I'm interested in how AI will affect the filming process." Similarly, viewers can also vote on questions presented. The results of these votes are tallied in real time by the server.

[1058] Prioritizing and presenting questions

[1059] The server tally the votes from the viewers and notify the introducer of the question that received the most votes. For example, if "Please explain the details of the night mode shooting function" received the most votes, this will be displayed to the introducer.

[1060] System Effects

[1061] This system facilitates Q&A sessions during product demonstrations, allowing customers to actively participate based on their interests. It also makes it easier for introducers to understand customer interests through questions, resulting in high-quality demonstrations and interactive communication.

[1062] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1063] Step 1:

[1064] The terminal uses a microphone to record the voice during the product introduction demonstration. This recorded voice data is sent to the server. The input is the recorded voice data, and the output is the voice data sent to the server.

[1065] Step 2:

[1066] The server converts the received voice data into text data using the Google Speech-to-Text API. The input is the voice data, and the output is the converted text data. As part of this conversion process, the server makes an API call and obtains the results.

[1067] Step 3:

[1068] The device uses a camera to capture images of slides during the demonstration. This captured image data is also sent to the server. The input is the captured image data, and the output is the image data sent to the server.

[1069] Step 4:

[1070] The server analyzes the received image data using Tesseract OCR and extracts key character information. The input is image data, and the output is extracted text information. As part of this image analysis process, the server passes the image data to the OCR engine and returns the analysis results as text.

[1071] Step 5:

[1072] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. The input is text data of the speech and text information of the image, and the output is the analyzed introduction. The server combines the text data to extract important keywords and context.

[1073] Step 6:

[1074] The server generates a question using a generative AI model (GPT-3) based on the analysis results. The input is the analyzed introduction content, and the output is the generated question. In this procedure, the server sends a prompt to the generative AI model to generate a question.

[1075] Step 7:

[1076] The server presents the generated questions to the user's terminal in real time. The input is the generated question, and the output is the question displayed on the user's terminal. The server sends data to each terminal and displays it on the screen.

[1077] Step 8:

[1078] Users can post their own questions from their terminals. The input is the question entered by the user, and the output is the question sent to the server. The terminal receives the user's input and sends it to the server.

[1079] Step 9:

[1080] Users can vote on presented questions. The input is the vote selected by the user, and the output is the voting data sent to the server. The user's terminal sends the voting information for the selected question.

[1081] Step 10:

[1082] The server tally the votes and decides the priority questions. The input is the voting data and the output is the priority questions. The server tally each vote and selects the question with the most votes.

[1083] Step 11:

[1084] The server notifies the introducer's terminal of high-priority questions. The input is the high-priority questions, and the output is the questions displayed on the introducer's terminal. The server sends the selected questions to the introducer in real time and displays them on the screen.

[1085] 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.

[1086] System Overview

[1087] This invention provides a presentation support system that combines an emotion engine that recognizes the user's emotions. The system analyzes the presentation's audio data and slide images, automatically generates and presents questions, and combines this with a means to adjust the priority of questions using emotion data, making the Q&A session more effective and engaging.

[1088] System configuration

[1089] 1. Terminal

[1090] During the presentation, the device will record the presenter's voice using a microphone and capture images of the slides using a camera or screen capture tool.

[1091] 2. Server

[1092] Voice data is received and converted into text using a voice recognition engine.

[1093] The acquired slide images are received and text is extracted using an image recognition engine.

[1094] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[1095] Questions are generated using generative AI based on the content of the presentation.

[1096] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[1097] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[1098] 3. Emotion Engine

[1099] It recognizes emotions from viewers' facial expressions and tone of voice, and sends the data to the server in real time.

[1100] Dynamically adjust question priorities based on sentiment data.

[1101] Specific processing examples

[1102] Acquiring and processing audio data

[1103] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[1104] Slide image acquisition and processing

[1105] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[1106] Analysis of presentation content

[1107] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[1108] Question Generation

[1109] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[1110] Audience emotion recognition

[1111] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. For example, if the viewer has an interested look on their face, that data is used to prioritize questions.

[1112] Viewer questions and polls

[1113] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[1114] Prioritizing and presenting questions

[1115] The server then selects the highest-priority questions based on audience votes and sentiment data and notifies the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, it will be displayed to the presenter.

[1116] System Effects

[1117] This system makes Q&A sessions during presentations more effective and engaging. Audience members can actively participate based on their interests, and presenters can answer questions while understanding the audience's emotions, enabling deeper communication. The result is higher-quality presentations and more interactive engagement.

[1118] The processing flow will be explained below.

[1119] Step 1:

[1120] As soon as the presentation begins, the device will use the microphone to record the presenter's voice and the camera or screen capture tool to capture images of the slides.

[1121] Step 2:

[1122] The terminal transmits the recorded audio data and the acquired slide images to the server at regular intervals.

[1123] Step 3:

[1124] The server passes the received voice data to a speech recognition engine, which converts the voice data into text data. For example, a statement such as "AI will become mainstream in many industries over the next decade" is converted into text data.

[1125] Step 4:

[1126] The server passes the received slide image data to an image recognition engine, which extracts key text and figures from the slide. For example, text such as "AI applications: medical, education, and transportation" is extracted.

[1127] Step 5:

[1128] The server combines the text data generated by speech recognition with the key text extracted by image recognition, and uses NLP technology to analyze the presentation and extract key points and keywords.

[1129] Step 6:

[1130] Based on the analysis results, the server uses generative AI to generate questions such as, "Which field do you think will see the most dramatic changes with the application of AI?" These questions are then displayed in real time on the viewer's device.

[1131] Step 7:

[1132] The emotion engine analyzes viewers' facial expressions and vocal tone in real time, using cameras and microphones to capture viewers' emotional data (e.g., interest, surprise, joy) while they watch.

[1133] Step 8:

[1134] The emotion engine sends the acquired emotion data to the server, and the priority of questions is dynamically adjusted based on that data. For example, questions that show a lot of interest to viewers can be given higher priority.

[1135] Step 9:

[1136] Users (viewers) can post their own questions from their devices. For example, they can post a question such as, "I would like to know more about the risks of AI." The device then sends this question to the server.

[1137] Step 10:

[1138] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[1139] Step 11:

[1140] The server aggregates all questions and votes and uses audience sentiment data to determine priority. For example, if the question "How will AI revolutionize education?" receives a lot of votes and emotional responses from viewers, it will be prioritized.

[1141] Step 12:

[1142] The server then notifies the presenter of the selected high-priority questions, allowing the presenter to conduct a Q&A session based on these questions. For example, by prioritizing answers to questions that evoke strong emotional responses, the presenter can increase audience interest.

[1143] Through these processing steps, the Q&A session during the presentation can be conducted more effectively and attractively, providing a satisfying environment for both the presenter and the audience.

[1144] Example 2

[1145] 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."

[1146] Conventional presentation support systems have the problem that the priority of questions received by the presenter is not optimized based on the audience's interests and emotions, making it difficult to effectively conduct a Q&A session. In addition, there is a lack of a way to present questions without distinguishing between questions from the audience and questions generated by the AI ​​generation system, making it difficult to realize a Q&A session that reflects the audience's interests.

[1147] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for converting audio data into text data, means for analyzing slide images and extracting key text, means for generating questions using a generative AI model, means for using an emotion engine that analyzes viewers' facial expressions and tone of voice, and means for dynamically adjusting the priority of questions based on emotion data. This enables the progress of an effective Q&A session that reflects viewers' interests and emotions.

[1148] "Means for recording audio data of a presentation" refers to a device or method for collecting the presenter's remarks as digital data using audio equipment such as a microphone.

[1149] "Means for acquiring slide images of presentations" refers to a device or method for capturing images of slides used during presentations and collecting them as digital image data.

[1150] "Means for converting voice data into text data" refers to software or a system that uses voice recognition technology to convert recorded voice data into text data.

[1151] "Means for analyzing slide images and extracting key text" refers to software or a system that extracts text information from slide images acquired using image recognition technology.

[1152] "Means for integrating text data and extracted key texts to analyze presentation content" refers to software or systems that integrate collected text data and use natural language processing technology to understand and analyze presentation content.

[1153] "Means using a generative AI model that generates questions based on the content of a presentation" refers to an artificial intelligence model and method that automatically generates questions based on the content of a presentation analyzed using natural language generation technology.

[1154] The "means for presenting the generated questions to the viewer terminal" refers to software or communication technology for displaying the questions generated by the server on the device used by the viewer.

[1155] "Means for viewers to post questions and vote on submitted questions" means an interface or system through which viewers can submit their own questions and rate other submitted questions.

[1156] The "means for aggregating vote results and determining question priority" refers to an algorithm or system that aggregates vote data from viewers and determines the importance of each question based on the results.

[1157] "Means using an emotion engine that analyzes the viewer's facial expressions and tone of voice" refers to software or equipment for analyzing the viewer's facial expressions and tone of voice and extracting emotion data.

[1158] "Means for dynamically adjusting question priority based on emotional data" refers to an algorithm or system that changes the display order and importance of questions in real time, taking into account the acquired emotional data.

[1159] The present invention relates to a presentation support system that optimizes a question and answer session using user emotion data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following describes in detail the preferred embodiments of the present invention.

[1160] First, we will describe the main embodiment of the system, which uses hardware and software to collect and analyze audio data, image data, and emotional data to effectively facilitate a question-and-answer session between the presenter and the audience.

[1161] Hardware and software used

[1162] The device includes a microphone, camera, and screen capture tool. These devices are used to collect the presenter's voice and slide images in real time. For example, when a presenter explains "Medical Applications of AI Technology," the device records the voice with the microphone and captures slide images with the screen capture tool.

[1163] The server is a central system for analyzing and integrating the collected data. It receives the audio data and slide image data sent from the devices via an internet connection and processes them as follows:

[1164] 1. Speech recognition: Using a speech recognition engine such as Google Cloud Speech-to-Text, the audio data is converted into text data. For example, a speaker's words such as "AI technology is being used in the medical field" are obtained as text data.

[1165] 2. Image Recognition: Using an image recognition engine such as Google Cloud Vision, key text is extracted from the slide image, specifically the slide title "Applications of AI in Medicine."

[1166] 3. Analysis of presentation content: The acquired text data is integrated and the presentation content is analyzed using natural language processing (NLP) technology. Here, the key points and key keywords of the presentation are identified. For example, key points such as "This presentation explains different application areas of AI" are extracted.

[1167] 4. Question generation: Based on the analysis results, questions are generated using generative AI (such as OpenAI's GPT-4). The generated questions are displayed in real time on the viewer's device. For example, a question such as, "What specific changes can we expect to see in the medical field through the use of AI?"

[1168] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotion data to the server. For example, if the viewer has an interested expression, the data will be collected by the emotion engine and sent to the server.

[1169] Prompt Sentence Examples

[1170] "Generate a question based on the following presentation: AI has achieved remarkable results in the field of medical imaging diagnosis. In particular, AI algorithms are now able to detect cellular abnormalities with high accuracy. This has significantly improved the speed and accuracy of diagnoses and reduced the burden on doctors."

[1171] This allows viewers to post questions based on their own interests and also vote on questions generated by other viewers. The server dynamically prioritizes questions based on the collected voting and sentiment data, and provides the presenter with the most relevant questions.

[1172] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1173] Processing flow and step details

[1174] Step 1: Acquiring and sending audio data

[1175] The device records the presenter's voice with a microphone and sends the audio data to the server. Specifically, the audio during the presentation is recorded in real time and converted into a digital audio file. This audio file is then sent to the server via the Internet.

[1176] Input: Speaker's voice

[1177] Data processing: audio recording and digital conversion

[1178] Output: Digital audio file (sent to server)

[1179] Step 2: Capture and send slide images

[1180] The device uses a screen capture tool to capture image data of the slides and sends it to the server. When the presenter switches slides, the device automatically performs a screen capture and sends the captured image to the server.

[1181] Input: Slide screen

[1182] Data processing: Screen capture

[1183] Output: Image data file (sent to server)

[1184] Step 3: Convert audio data to text

[1185] The server receives the voice data and converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The digital voice file sent to the server is analyzed and converted by the speech recognition engine and saved as a string of characters.

[1186] Input: Digital audio data

[1187] Data processing: Speech recognition (voice to text conversion)

[1188] Output: Text data

[1189] Step 4: Extract text from slide images

[1190] The server extracts key text from the slide images using an image recognition engine (e.g., Google Cloud Vision). It analyzes the received image data files to identify and extract text within the images.

[1191] Input: Slide image data

[1192] Data processing: Image recognition (text extraction)

[1193] Output: Text data

[1194] Step 5: Analyze the presentation

[1195] The server integrates the text data obtained from the speech and image recognition and analyzes the presentation using natural language processing (NLP) technology. Based on the integrated text data, the main points and key keywords of the presentation are identified.

[1196] Input: Text data extracted from audio and slides

[1197] Data Processing: Integration and NLP Analysis

[1198] Output: Gist and keyword data

[1199] Step 6: Generate questions

[1200] The server uses a generative AI model (e.g., GPT-4) to generate questions based on the analysis results. The server uses the key points and keywords obtained as prompts to generate appropriate questions.

[1201] Input: Key points and keyword data

[1202] Data processing: Question generation (generative AI model)

[1203] Output: Question text

[1204] Step 7: Posing the Question

[1205] The server presents the generated questions to the viewer's terminal in real time, and transmits the generated questions to the viewer's device and provides them through an interface for display.

[1206] Input: Question text

[1207] Data processing: Submitting and presenting questions

[1208] Output: Question displayed on the viewer's device

[1209] Step 8: Collect and send emotion data

[1210] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. The camera and microphone collect the viewer's emotions and send them to the server as digital data.

[1211] Input: Viewer facial expressions and tone of voice

[1212] Data processing: Sentiment analysis

[1213] Output: Emotion data (sent to server)

[1214] Step 9: Viewer polls

[1215] Users (viewers) vote on questions presented through their terminals. Each viewer votes for the questions they are interested in, and the vote data is sent to the server in real time.

[1216] Input: Viewer votes

[1217] Data processing: Collecting voting data

[1218] Output: Voting data (sent to server)

[1219] Step 10: Prioritize and present questions

[1220] The server dynamically determines the priority of questions based on the collected voting data and sentiment data, and selects the most important question and displays it on the presenter's device.

[1221] Input: Voting data and sentiment data

[1222] Data processing: data integration and prioritization

[1223] Output: Top priority question (displayed to presenter's device)

[1224] The above processing steps enable an effective and engaging Q&A session between the presenter and the audience.

[1225] (Application example 2)

[1226] 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."

[1227] Conventional presentation support systems and robot work support systems have problems in that they are unable to fully realize Q&A during presentations or improve the efficiency of robot work in factories. Specifically, they do not properly manage the priority of audience questions, and do not immediately adjust the robot's behavior in response to the worker's instructions or emotions, hindering efficient work. This results in low satisfaction for viewers and workers, and a lack of two-way communication.

[1228] 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.

[1229] In this invention, the server includes means for generating questions based on the content of the presentation, means for presenting the generated questions to viewer terminals, means for viewers to post questions, means for viewers to vote on the presented questions, means for tallying up vote results and determining priorities, means for acquiring robot work status data, means for acquiring voice instructions from workers and dynamically optimizing work content using a generation AI, and means for acquiring worker emotion data and adjusting work plans based on the emotion data. This improves the efficiency of question and answer sessions between viewers and presenters, enabling more efficient robot work in factories.

[1230] "Presentation audio data" refers to recorded data of the audio uttered by the presenter during the presentation.

[1231] "Slide images" are image data of visual slides used during a presentation.

[1232] A "voice recognition engine" is software or hardware for converting voice data into text data.

[1233] An "image recognition engine" is software or hardware for extracting key text from image data.

[1234] "Integration" is the process of combining multiple different data sets into a single dataset.

[1235] "Question generation" is the process of automatically creating appropriate questions based on the presentation content and analysis results.

[1236] "Viewer device" means the device used by the viewer to view the presentation.

[1237] "Voting results" are the aggregated results of votes cast by viewers on questions.

[1238] "Priority" is the measure by which a particular question or task is judged to be more important than other questions or tasks.

[1239] "Robot work status data" refers to data relating to the work currently being performed by the robot and its status.

[1240] "Generative AI" is artificial intelligence that generates appropriate questions and answers based on given data and conditions.

[1241] "Emotion data" refers to emotional information obtained from the worker's facial expressions, tone of voice, etc.

[1242] "Dynamic optimization" is the process of adjusting to the optimal work plan in real time as conditions change.

[1243] A "work plan" is the design of procedures and schedules for carrying out a specific task.

[1244] This invention relates to a "robot work support system" for improving the efficiency of robot work in factories. This system combines a function to support questions and answers during presentations with a function to dynamically optimize robot work in factories, thereby increasing the satisfaction of viewers and workers.

[1245] System Overview

[1246] Terminal

[1247] The terminal uses the microphone and camera built into the robot body to capture the worker's voice instructions and work status in real time. The smart glasses or tablet used by the worker also function as part of the system.

[1248] server

[1249] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the acquired voice data into text data, an image recognition engine (e.g., Amazon Rekognition) to analyze image data acquired from the robot's camera, and an NLP technology (e.g., spaCy) to integrate the results of speech recognition and image recognition, and a generative AI (e.g., OpenAI GPT) to generate appropriate questions and work plans.

[1250] Emotion Engine

[1251] An emotion engine (e.g., Microsoft Azure Emotion API) is used to analyze workers' facial expressions and voice tones in real time to obtain emotional data, which is used to dynamically adjust the work plan.

[1252] Specific processing examples

[1253] 1. Acquiring and processing audio data

[1254] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits them to the server. The server then converts the voice data into text data using a voice recognition engine. During this process, an instruction such as "Please install the next part" is obtained as text data.

[1255] 2. Image data acquisition and processing

[1256] The terminal uses the robot's built-in camera to capture images of the work status and parts, and sends them to the server. The server then analyzes the image data using an image recognition engine to extract key information. For example, specific parts and work status may be recognized.

[1257] 3. Analysis and adjustment of work content

[1258] The server integrates and analyzes the voice and image recognition results, and uses generative AI to generate questions and work plans, which are then presented to the worker's smart glasses or tablet.

[1259] 4. Acquiring and Processing Emotional Data

[1260] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, capturing emotional data. For example, if the worker is feeling dissatisfied, the data is sent to the server in real time and used to adjust the work plan.

[1261] Prompt Sentence Examples

[1262] "When an AI robot follows instructions in a factory task, suggest ways to optimize the task. Voice command: 'Install the next part.' Worker emotion: Frustration evident."

[1263] This allows for efficient question and answer sessions between the audience and the presenter, while also dynamically optimizing the work of robots in factories, improving worker satisfaction and work efficiency.

[1264] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1265] Step 1:

[1266] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits the voice data to the server. The input is the worker's voice instructions, and the output is the voice data sent to the server. Specifically, the microphone captures the voice, which is then saved and transmitted as digital data.

[1267] Step 2:

[1268] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data, and the output is text data corresponding to the voice instructions. Specifically, the speech recognition engine analyzes the voice data and generates text using a language model.

[1269] Step 3:

[1270] The terminal uses the robot's built-in camera to acquire image data of the work situation and sends that image data to the server. The input is visual information of the work situation, and the output is image data sent to the server. Specifically, the camera captures the image, saves it as digital data, and sends it.

[1271] Step 4:

[1272] The server uses an image recognition engine to analyze the received image data and extract key information. The input is the image data, and the output is extracted key text and recognized objects. Specifically, the image recognition engine analyzes the image data pixel by pixel to identify key text and objects.

[1273] Step 5:

[1274] The server integrates the results of speech recognition and image recognition and analyzes the work content using NLP technology. The input is text data and extracted key information, and the output is the gist of the analyzed work content. Specifically, the text data and image data are integrated into a single dataset and analyzed using an NLP model.

[1275] Step 6:

[1276] The server uses a generation AI to generate appropriate questions and work plans based on the analysis results. The input is the gist of the analyzed work content, and the output is the generated questions and work plans. Specifically, the generation AI generates questions and work plans related to the prompt.

[1277] Step 7:

[1278] The terminal presents the generated questions and work plans to the worker's smart glasses or tablet. The input is the generated questions and work plans, and the output is a visual display. The specific operation is to display the information on the device's display.

[1279] Step 8:

[1280] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, and acquires emotion data. The input is video data of the worker's facial expressions, and the output is the emotion recognition results. Specifically, the camera captures the facial expressions, and the emotion recognition engine analyzes the video data.

[1281] Step 9:

[1282] The server dynamically adjusts the work plan based on the acquired emotion data. The input is emotion data, and the output is an adjusted work plan. Specifically, it analyzes the emotion recognition results and modifies the priority and procedures of the work plan in real time.

[1283] 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.

[1284] 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.

[1285] 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.

[1286] [Fourth embodiment]

[1287] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1288] 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.

[1289] 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).

[1290] 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.

[1291] 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.

[1292] 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).

[1293] 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.

[1294] 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.

[1295] 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.

[1296] 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.

[1297] 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.

[1298] 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.

[1299] 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."

[1300] System Overview

[1301] The invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to ensure that the Q&A session proceeds smoothly and satisfies both the presenter and the audience.

[1302] System configuration

[1303] 1. Terminal

[1304] During the presentation, the device uses a microphone to record the presenter's voice.

[1305] At the same time, capture an image of the slide using a camera or a screen capture tool.

[1306] 2. Server

[1307] Voice data is received and converted into text using a voice recognition engine.

[1308] The acquired slide images are received and text is extracted using an image recognition engine.

[1309] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[1310] Questions are generated using generative AI based on the content of the presentation.

[1311] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[1312] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[1313] Specific processing examples

[1314] Acquiring and processing audio data

[1315] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[1316] Slide image acquisition and processing

[1317] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[1318] Analysis of presentation content

[1319] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[1320] Question Generation

[1321] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[1322] Viewer questions and polls

[1323] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[1324] Prioritizing and presenting questions

[1325] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[1326] System Effects

[1327] This system allows the Q&A session to proceed smoothly and create a lively atmosphere. The audience can actively participate based on their interests, and the presenter can easily understand the audience's interests through questions. The result is high-quality presentations and interactive communication.

[1328] The processing flow will be explained below.

[1329] Step 1:

[1330] As soon as the presentation begins, the device uses a microphone to record the presenter's voice and also captures images of the slides using a camera or screen capture tool.

[1331] Step 2:

[1332] The device transmits the recorded audio data to the server at regular intervals, and simultaneously transmits the acquired slide image data to the server.

[1333] Step 3:

[1334] The server then runs the received voice data through a speech recognition engine and converts it into text data, such as "AI will become mainstream in many industries over the next decade."

[1335] Step 4:

[1336] The server then runs the received slide image data through an image recognition engine to extract key text and figures from the slides. For example, text such as "AI applications: medical, education, and transportation" may be extracted.

[1337] Step 5:

[1338] The server combines the text data generated by the speech recognition engine with the key text extracted by the image recognition engine. Using this combined data, NLP technology is used to analyze the presentation content and extract key points and keywords.

[1339] Step 6:

[1340] The server uses generative AI to generate questions based on the analysis results, such as, "Which field do you think will see the most dramatic changes with the application of AI?"

[1341] Step 7:

[1342] The server presents the generated questions to the viewer's terminal in real time, and the terminal displays the received questions on the viewer's screen.

[1343] Step 8:

[1344] The user (viewer) uses the application's question submission form to enter and submit their question. For example, a question such as "Can this method be applied to other cases?" is submitted. The device then sends this question to the server.

[1345] Step 9:

[1346] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[1347] Step 10:

[1348] The server will tally all the questions and votes it receives and select the question with the most votes, for example, "How will AI revolutionize education?"

[1349] Step 11:

[1350] The server notifies the presenter of the selected high-priority questions, and the presenter can then conduct a Q&A session based on these questions.

[1351] These processing steps ensure that the Q&A session following the presentation proceeds smoothly, providing a satisfying environment for both the presenter and the audience.

[1352] Example 1

[1353] 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."

[1354] In conventional Q&A sessions, there were problems such as the complicated process for viewers to submit questions and the difficulty for presenters to grasp the audience's interests. Furthermore, when a large number of questions are submitted, it becomes difficult to select the most important questions. In such cases, the limited time for Q&A often results in situations where many viewers are unable to receive adequate answers.

[1355] 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.

[1356] In this invention, the server includes a means for generating questions using a generative model based on the content of the presentation, a means for presenting the generated questions to the viewer's terminal, and a means for aggregating voting results in real time and determining priorities. This allows viewers to quickly and easily post questions and select important questions through voting. As a result, the Q&A session proceeds efficiently, and two-way communication between the presenter and the audience is facilitated.

[1357] "Presentation audio data" refers to data that includes audio information of a speaker in a presentation, lecture, or other presentation.

[1358] A "slide image" is image data that includes visual materials and charts used by a presenter during a presentation.

[1359] "Text data" refers to data that includes text information extracted from audio data or image data.

[1360] "Main text" is text data that includes important sentence information and keywords extracted from the slide image.

[1361] "Integration" refers to the process of combining multiple pieces of data or information and organizing them into a single data set or piece of information.

[1362] "Analysis" is the process of examining data or information in detail to make its structure and meaning easier to understand.

[1363] A "generative model" is an algorithm that automatically creates new data and information using machine learning techniques such as neural networks.

[1364] "Means for generating questions" refers to technology for automatically generating questions based on the content of a presentation.

[1365] "Means for presenting on the viewer's device" refers to the technology for displaying the generated questions and information on the screen of the viewer's device.

[1366] "Means for viewers to submit questions" refers to the technology by which viewers submit their questions to the system.

[1367] "Means for viewers to vote on questions presented" refers to technology that allows viewers to rate or select questions displayed on the screen.

[1368] "Means for aggregating voting results in real time" refers to technology for instantly aggregating voting data from viewers and determining rankings and priorities.

[1369] "Means for determining priorities" refers to a technique for determining which questions are most important based on voting results and determining their order.

[1370] "Means of notification" refers to techniques for conveying specific information to a target audience, such as a presenter.

[1371] This invention is a system that analyzes the audio data and slide images of a presentation and automatically generates and presents questions. This system aims to smoothly conduct Q&A sessions during presentations and to satisfy both the presenter and the audience.

[1372] System hardware and software configuration

[1373] Terminal

[1374] The terminal is a device for recording the presenter's voice and acquiring slide images. The terminal has the following functions:

[1375] Microphone: Records the presenter's voice in real time.

[1376] Camera or screen capture tool: Take an image of the slide.

[1377] The terminal may be a general-purpose computing device such as a laptop, tablet, or smartphone.

[1378] server

[1379] The server is the core component that analyzes the audio data and slide image data, and generates and presents questions. The specific software and functions used are as follows:

[1380] Speech recognition engine: Converts voice data into text data using Google Cloud Speech-to-Text or similar.

[1381] Image recognition engine: Extract text from slide images using Google Cloud Vision or similar.

[1382] NLP technology: Using natural language processing technology, voice recognition and image recognition results are analyzed to extract key points and keywords.

[1383] Generative AI model: Using OpenAI's GPT-3 and other models, questions are generated based on the analysis results.

[1384] Real-time tallying function: Tally the votes from the audience and notify the presenter of the highest priority questions.

[1385] User

[1386] Users (viewers) can post questions and vote on questions provided from their own devices.

[1387] Specific examples of program processing

[1388] Acquiring and processing audio data

[1389] The device records the speaker's voice and sends the audio data to a server. The server then converts the audio data into text using Google Cloud Speech-to-Text. For example, a statement such as "AI has a wide range of applications" in a presentation can be converted into text.

[1390] Slide image acquisition and processing

[1391] The device uses a screen capture tool to capture an image of the slide and sends it to the server, which then uses Google Cloud Vision to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[1392] Analysis of presentation content

[1393] The server combines the received voice and image recognition results and uses NLP technology to analyze the content of the presentation, providing key points such as, "This presentation explains different areas in which AI can be applied."

[1394] Question Generation

[1395] Based on the analysis results, the server uses a generative AI model to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[1396] Viewer questions and polls

[1397] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Viewers can also vote on the questions presented, and the results of these votes are tallied in real time by the server.

[1398] Prioritizing and presenting questions

[1399] Based on the audience votes, the server will select the highest priority questions and notify the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, this will be displayed to the presenter.

[1400] Prompt Sentence Examples

[1401] Generate questions based on the given presentation using the following criteria:

[1402] Announcement: "AI has a wide range of applications. It is being used in a wide variety of fields, from medicine to education and even entertainment."

[1403] Generating conditions: Ask for specific examples and numerical questions.

[1404] This system facilitates smooth Q&A sessions following presentations and allows for two-way communication between presenters and audience members.

[1405] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1406] Step 1:

[1407] The device records the presenter's voice. When the presentation begins, the device's microphone automatically starts up and collects the voice data. The input here is the voice captured in real time through the microphone, and the output is the voice data that is temporarily saved in local storage. Specifically, the voice data is saved periodically every few seconds, and this saved data is put on hold for further processing.

[1408] Step 2:

[1409] The device uses a screen capture tool to obtain image data of the slides. Each time a slide changes, the device automatically performs a screen capture and obtains image data. The input here is the slide screen obtained through the screen capture tool, and the output is image data that is temporarily saved in local storage. Specifically, each time the slide changes to a new page, an image is captured and saved.

[1410] Step 3:

[1411] The device sends the recorded audio data to the server. The input here is the audio data stored in the local storage, and the output is the audio data sent to the server. Specifically, the audio files are divided into fixed sizes and sent to the server one by one, and the server waits to receive them.

[1412] Step 4:

[1413] The server converts the received voice data into text data using a speech recognition engine such as Google Cloud Speech-to-Text. The input here is the transmitted voice data, and the output is text data. Specifically, the voice file is input into the engine, and within a few seconds, the corresponding text is generated and stored in a database.

[1414] Step 5:

[1415] The device sends the acquired slide image data to the server. The input here is the image data stored in the local storage, and the output is the image data sent to the server. Specifically, the captured image files are sent to the server one by one, and the server enters a waiting state to receive them.

[1416] Step 6:

[1417] The server analyzes the received slide image data using an image recognition engine such as Google Cloud Vision to extract key text. The input here is the transmitted slide image data, and the output is the extracted text data. Specifically, the image file is input into the engine, and the text is analyzed, extracted, and stored in a database.

[1418] Step 7:

[1419] The server integrates the results of speech recognition and image recognition and uses NLP (natural language processing) technology to perform a detailed analysis of the presentation content. The input here is text data from speech recognition and image recognition, and the output is a summary of the presentation content and key keywords. Specifically, the text data is input into the NLP model, and key points are extracted and summarized.

[1420] Step 8:

[1421] The server uses a generative AI model (e.g., GPT-3) based on the analysis results to generate questions. The input here is the text data of the analyzed presentation, and the output is the generated question text. Specifically, the analysis data is input into the AI ​​model as a prompt, and natural questions are generated.

[1422] Step 9:

[1423] The server presents the generated question to the viewer's device in real time. The input here is the generated question text, and the output is the question displayed on the viewer's device. Specifically, the question text is sent to the viewer and displayed on the viewer's screen.

[1424] Step 10:

[1425] Users (viewers) post their own questions. The input here is text data entered by the viewer, and the output is the question data sent to the server. Specifically, when the viewer enters a question in the input field on their device and presses the send button, the data is sent to the server.

[1426] Step 11:

[1427] Users (viewers) vote on the questions presented. The input here is the viewer's voting operation, and the output is the voting data sent to the server. Specifically, when a viewer presses the voting button on their device, the data is sent to the server in real time.

[1428] Step 12:

[1429] The server tally the votes from viewers in real time and determine the priorities. The input here is the voting data, and the output is a list of questions with determined priorities. Specifically, the voting results are tallyed in real time using a tallying algorithm, and priorities are assigned.

[1430] Step 13:

[1431] The server notifies the presenter of high-priority questions. The input is a prioritized list of questions, and the output is the questions to be displayed on the presenter's device. Specifically, the highest-priority question is sent to the presenter's device and displayed visually.

[1432] (Application example 1)

[1433] 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."

[1434] In conventional presentations and product demonstrations, questions are generated and presented manually, making it difficult to smoothly conduct a Q&A session. Especially in situations with a large number of viewers or customers, important questions can be overlooked, reducing viewer and customer satisfaction. Furthermore, there was no system for collecting and voting on questions from viewers and customers in real time, making it difficult to effectively convey questions of interest to the presenter or presenter. The present invention aims to solve these problems.

[1435] 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.

[1436] In this invention, the server includes means for recording audio data of the presentation, means for acquiring slide images of the presentation, means for converting the audio data into text data, means for analyzing the slide images and extracting key text, means for integrating the text data and the extracted key text to analyze the content of the presentation, means for generating questions based on the content of the presentation, means for presenting the generated questions to user terminals, means for users to submit questions, means for users to vote on the submitted questions, means for tallying up the votes and determining priorities, means for analyzing the audio and images of the product introduction demonstration and automatically generating and presenting questions from customers, and means for presenting the generated questions to user terminals in real time. This enables the Q&A session to proceed smoothly, important questions to be effectively conveyed to the presenter or introducer, and increases the interest and satisfaction of many viewers and customers.

[1437] "Means for recording audio data of a presentation" refers to a device or system for recording the presenter's voice using an audio input device.

[1438] "Means for capturing images of presentation slides" refers to a device or system for capturing images of slides during a presentation using a camera or screen capture tool.

[1439] The "means for converting the voice data into text data" refers to a system or method for converting recorded voice data into text using a voice recognition engine.

[1440] The "means for analyzing the slide images and extracting the main text" refers to a system or method for extracting text information from the acquired slide images using an image recognition engine.

[1441] "Means for integrating the above text data and the extracted main text to analyze the content of the presentation" refers to a system or method for integrating the results of speech recognition and image recognition and analyzing the content of the presentation using natural language processing technology.

[1442] A "means for generating questions based on presentation content" is a system or method for automatically generating questions from presentation content using a generative AI model.

[1443] The "means for presenting the generated question on the user's terminal" is a system or method for displaying the generated question on a device used by a viewer or customer.

[1444] "Means for users to post questions" refers to a device or function that allows viewers or customers to submit their questions to the system.

[1445] "Means for users to vote on questions presented" refers to a system or function that allows viewers or customers to vote on questions that are displayed.

[1446] The "means for aggregating vote results and determining priority" refers to a system or method for collecting voting data from users and determining the question that receives the most votes as the highest priority.

[1447] "Means for analyzing audio and video of a product introduction demonstration and automatically generating and presenting questions from customers" refers to a system or method for analyzing audio and video during a product introduction and generating and presenting questions from customers based on that analysis.

[1448] "Means for presenting generated questions to a user's device in real time" refers to a system or method for displaying generated questions on the spot and instantly on a viewer's or customer's device.

[1449] System Overview

[1450] This invention is a system that analyzes the audio and video of a product introduction demonstration and automatically generates and presents questions to customers, aiming to ensure a smooth Q&A session and satisfy both the customer and the introducer.

[1451] System configuration

[1452] Terminal

[1453] The terminal performs the following functions during the product demonstration:

[1454] Audio recording: Record the introducer's voice using the device's microphone.

[1455] Image capture: Uses the device's camera to capture images during the demonstration.

[1456] server

[1457] The server does the following:

[1458] Speech recognition: Receives voice data and converts it into text using a speech recognition engine (e.g., Google Speech-to-Text).

[1459] Image Analysis: Receive the captured image and extract the key text using Tesseract OCR.

[1460] Content analysis: The results of voice recognition and image recognition are integrated, and the introduction content is analyzed using natural language processing technology.

[1461] Question generation: Based on the analysis results, questions are generated using a generative AI model (e.g., OpenAI's GPT-3).

[1462] Question presentation: The generated questions are presented to the user's device in real time.

[1463] Voting function: Provides a mechanism for users to vote on questions posted by users.

[1464] Prioritization: Tally the votes and present the highest priority questions to the introducer.

[1465] Specific processing examples

[1466] Acquiring and processing audio data

[1467] The device uses a microphone to record the introducer's voice and sends the voice data to the server, which then converts the voice data into text data using Google Speech-to-Text. For example, a statement such as "This new smartphone has an innovative camera function" can be recognized.

[1468] Slide image acquisition and processing

[1469] The device uses its camera to capture an image of the demonstration and sends it to a server, which uses Tesseract OCR to extract key text from the image, such as "Advanced night mode photography with AI."

[1470] Content analysis and question generation

[1471] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. Based on this analysis, a prompt is created to be provided to the generative AI model (GPT-3), which then generates a question. For example, the prompt is as follows:

[1472] Example of question generation

[1473] "Generate customer questions based on the following text: Voiceover: "This new smartphone comes with revolutionary camera features." Slide: "Advanced night mode photography powered by AI.""

[1474] An example of a question that can be generated is:

[1475] "Can you tell me more about the night mode photography feature on this phone? I'm especially interested to know how AI helps with that."

[1476] is.

[1477] Viewer questions and polls

[1478] Users (viewers or customers) can post their own questions from their devices. For example, a question such as, "I'm interested in how AI will affect the filming process." Similarly, viewers can also vote on questions presented. The results of these votes are tallied in real time by the server.

[1479] Prioritizing and presenting questions

[1480] The server tally the votes from the viewers and notify the introducer of the question that received the most votes. For example, if "Please explain the details of the night mode shooting function" received the most votes, this will be displayed to the introducer.

[1481] System Effects

[1482] This system facilitates Q&A sessions during product demonstrations, allowing customers to actively participate based on their interests. It also makes it easier for introducers to understand customer interests through questions, resulting in high-quality demonstrations and interactive communication.

[1483] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1484] Step 1:

[1485] The terminal uses a microphone to record the voice during the product introduction demonstration. This recorded voice data is sent to the server. The input is the recorded voice data, and the output is the voice data sent to the server.

[1486] Step 2:

[1487] The server converts the received voice data into text data using the Google Speech-to-Text API. The input is the voice data, and the output is the converted text data. As part of this conversion process, the server makes an API call and obtains the results.

[1488] Step 3:

[1489] The device uses a camera to capture images of slides during the demonstration. This captured image data is also sent to the server. The input is the captured image data, and the output is the image data sent to the server.

[1490] Step 4:

[1491] The server analyzes the received image data using Tesseract OCR and extracts key character information. The input is image data, and the output is extracted text information. As part of this image analysis process, the server passes the image data to the OCR engine and returns the analysis results as text.

[1492] Step 5:

[1493] The server integrates the results of speech recognition and image recognition and analyzes the introduction using natural language processing technology. The input is text data of the speech and text information of the image, and the output is the analyzed introduction. The server combines the text data to extract important keywords and context.

[1494] Step 6:

[1495] The server generates a question using a generative AI model (GPT-3) based on the analysis results. The input is the analyzed introduction content, and the output is the generated question. In this procedure, the server sends a prompt to the generative AI model to generate a question.

[1496] Step 7:

[1497] The server presents the generated questions to the user's terminal in real time. The input is the generated question, and the output is the question displayed on the user's terminal. The server sends data to each terminal and displays it on the screen.

[1498] Step 8:

[1499] Users can post their own questions from their terminals. The input is the question entered by the user, and the output is the question sent to the server. The terminal receives the user's input and sends it to the server.

[1500] Step 9:

[1501] Users can vote on presented questions. The input is the vote selected by the user, and the output is the voting data sent to the server. The user's terminal sends the voting information for the selected question.

[1502] Step 10:

[1503] The server tally the votes and decides the priority questions. The input is the voting data and the output is the priority questions. The server tally each vote and selects the question with the most votes.

[1504] Step 11:

[1505] The server notifies the introducer's terminal of high-priority questions. The input is the high-priority questions, and the output is the questions displayed on the introducer's terminal. The server sends the selected questions to the introducer in real time and displays them on the screen.

[1506] 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.

[1507] System Overview

[1508] This invention provides a presentation support system that combines an emotion engine that recognizes the user's emotions. The system analyzes the presentation's audio data and slide images, automatically generates and presents questions, and combines this with a means to adjust the priority of questions using emotion data, making the Q&A session more effective and engaging.

[1509] System configuration

[1510] 1. Terminal

[1511] During the presentation, the device will record the presenter's voice using a microphone and capture images of the slides using a camera or screen capture tool.

[1512] 2. Server

[1513] Voice data is received and converted into text using a voice recognition engine.

[1514] The acquired slide images are received and text is extracted using an image recognition engine.

[1515] The results of speech recognition and image recognition are integrated to analyze the content of the presentation.

[1516] Questions are generated using generative AI based on the content of the presentation.

[1517] The system provides a mechanism for presenting generated questions and questions submitted by viewers to viewers' devices, allowing viewers to vote on questions.

[1518] Voting results will be tallied and the highest priority questions will be presented to the presenters.

[1519] 3. Emotion Engine

[1520] It recognizes emotions from viewers' facial expressions and tone of voice, and sends the data to the server in real time.

[1521] Dynamically adjust question priorities based on sentiment data.

[1522] Specific processing examples

[1523] Acquiring and processing audio data

[1524] The device records the speaker's voice and sends the audio data to the server. The server then uses a speech recognition engine to convert the voice data into text. During this speech recognition process, an example of a speech being made is a statement such as, "AI has a wide range of applications."

[1525] Slide image acquisition and processing

[1526] The device uses a screen capture tool to capture the image data of the slide and sends it to the server, which then uses an image recognition engine to extract key text from the image, such as the slide title, "Applications of AI in Medicine."

[1527] Analysis of presentation content

[1528] The server combines the received voice and image recognition results and analyzes the presentation using NLP technology. This analysis extracts key points and key keywords from the presentation. For example, key points such as "This presentation explains different application areas of AI" can be obtained.

[1529] Question Generation

[1530] Based on the analysis results, the server uses generative AI to generate questions such as, "What specific changes can we expect in the medical field through the use of AI?" These generated questions are then displayed in real time on the viewer's device.

[1531] Audience emotion recognition

[1532] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. For example, if the viewer has an interested look on their face, that data is used to prioritize questions.

[1533] Viewer questions and polls

[1534] Users (viewers) can post their own questions from their devices. For example, a question such as "Please tell me more about the impact of AI on education" can be posted. Similarly, viewers can vote on the questions presented. The results of these votes are tallied in real time by the server.

[1535] Prioritizing and presenting questions

[1536] The server then selects the highest-priority questions based on audience votes and sentiment data and notifies the presenter. For example, if the question "Please tell us more about the specific impact of AI on the labor market" receives the most votes, it will be displayed to the presenter.

[1537] System Effects

[1538] This system makes Q&A sessions during presentations more effective and engaging. Audience members can actively participate based on their interests, and presenters can answer questions while understanding the audience's emotions, enabling deeper communication. The result is higher-quality presentations and more interactive engagement.

[1539] The processing flow will be explained below.

[1540] Step 1:

[1541] As soon as the presentation begins, the device will use the microphone to record the presenter's voice and the camera or screen capture tool to capture images of the slides.

[1542] Step 2:

[1543] The terminal transmits the recorded audio data and the acquired slide images to the server at regular intervals.

[1544] Step 3:

[1545] The server passes the received voice data to a speech recognition engine, which converts the voice data into text data. For example, a statement such as "AI will become mainstream in many industries over the next decade" is converted into text data.

[1546] Step 4:

[1547] The server passes the received slide image data to an image recognition engine, which extracts key text and figures from the slide. For example, text such as "AI applications: medical, education, and transportation" is extracted.

[1548] Step 5:

[1549] The server combines the text data generated by speech recognition with the key text extracted by image recognition, and uses NLP technology to analyze the presentation and extract key points and keywords.

[1550] Step 6:

[1551] Based on the analysis results, the server uses generative AI to generate questions such as, "Which field do you think will see the most dramatic changes with the application of AI?" These questions are then displayed in real time on the viewer's device.

[1552] Step 7:

[1553] The emotion engine analyzes viewers' facial expressions and vocal tone in real time, using cameras and microphones to capture viewers' emotional data (e.g., interest, surprise, joy) while they watch.

[1554] Step 8:

[1555] The emotion engine sends the acquired emotion data to the server, and the priority of questions is dynamically adjusted based on that data. For example, questions that show a lot of interest to viewers can be given higher priority.

[1556] Step 9:

[1557] Users (viewers) can post their own questions from their devices. For example, they can post a question such as, "I would like to know more about the risks of AI." The device then sends this question to the server.

[1558] Step 10:

[1559] Users (viewers) vote for questions posed by other viewers by clicking "Like." The terminal sends the voting results to the server in real time.

[1560] Step 11:

[1561] The server aggregates all questions and votes and uses audience sentiment data to determine priority. For example, if the question "How will AI revolutionize education?" receives a lot of votes and emotional responses from viewers, it will be prioritized.

[1562] Step 12:

[1563] The server then notifies the presenter of the selected high-priority questions, allowing the presenter to conduct a Q&A session based on these questions. For example, by prioritizing answers to questions that evoke strong emotional responses, the presenter can increase audience interest.

[1564] Through these processing steps, the Q&A session during the presentation can be conducted more effectively and attractively, providing a satisfying environment for both the presenter and the audience.

[1565] Example 2

[1566] 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."

[1567] Conventional presentation support systems have the problem that the priority of questions received by the presenter is not optimized based on the audience's interests and emotions, making it difficult to effectively conduct a Q&A session. In addition, there is a lack of a way to present questions without distinguishing between questions from the audience and questions generated by the AI ​​generation system, making it difficult to realize a Q&A session that reflects the audience's interests.

[1568] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for converting audio data into text data, means for analyzing slide images and extracting key text, means for generating questions using a generative AI model, means for using an emotion engine that analyzes viewers' facial expressions and tone of voice, and means for dynamically adjusting the priority of questions based on emotion data. This enables the progress of an effective Q&A session that reflects viewers' interests and emotions.

[1569] "Means for recording audio data of a presentation" refers to a device or method for collecting the presenter's remarks as digital data using audio equipment such as a microphone.

[1570] "Means for acquiring slide images of presentations" refers to a device or method for capturing images of slides used during presentations and collecting them as digital image data.

[1571] "Means for converting voice data into text data" refers to software or a system that uses voice recognition technology to convert recorded voice data into text data.

[1572] "Means for analyzing slide images and extracting key text" refers to software or a system that extracts text information from slide images acquired using image recognition technology.

[1573] "Means for integrating text data and extracted key texts to analyze presentation content" refers to software or systems that integrate collected text data and use natural language processing technology to understand and analyze presentation content.

[1574] "Means using a generative AI model that generates questions based on the content of a presentation" refers to an artificial intelligence model and method that automatically generates questions based on the content of a presentation analyzed using natural language generation technology.

[1575] The "means for presenting the generated questions to the viewer terminal" refers to software or communication technology for displaying the questions generated by the server on the device used by the viewer.

[1576] "Means for viewers to post questions and vote on submitted questions" means an interface or system through which viewers can submit their own questions and rate other submitted questions.

[1577] The "means for aggregating vote results and determining question priority" refers to an algorithm or system that aggregates vote data from viewers and determines the importance of each question based on the results.

[1578] "Means using an emotion engine that analyzes the viewer's facial expressions and tone of voice" refers to software or equipment for analyzing the viewer's facial expressions and tone of voice and extracting emotion data.

[1579] "Means for dynamically adjusting question priority based on emotional data" refers to an algorithm or system that changes the display order and importance of questions in real time, taking into account the acquired emotional data.

[1580] The present invention relates to a presentation support system that optimizes a question and answer session using user emotion data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following describes in detail the preferred embodiments of the present invention.

[1581] First, we will describe the main embodiment of the system, which uses hardware and software to collect and analyze audio data, image data, and emotional data to effectively facilitate a question-and-answer session between the presenter and the audience.

[1582] Hardware and software used

[1583] The device includes a microphone, camera, and screen capture tool. These devices are used to collect the presenter's voice and slide images in real time. For example, when a presenter explains "Medical Applications of AI Technology," the device records the voice with the microphone and captures slide images with the screen capture tool.

[1584] The server is a central system for analyzing and integrating the collected data. It receives the audio data and slide image data sent from the devices via an internet connection and processes them as follows:

[1585] 1. Speech recognition: Using a speech recognition engine such as Google Cloud Speech-to-Text, the audio data is converted into text data. For example, a speaker's words such as "AI technology is being used in the medical field" are obtained as text data.

[1586] 2. Image Recognition: Using an image recognition engine such as Google Cloud Vision, key text is extracted from the slide image, specifically the slide title "Applications of AI in Medicine."

[1587] 3. Analysis of presentation content: The acquired text data is integrated and the presentation content is analyzed using natural language processing (NLP) technology. Here, the key points and key keywords of the presentation are identified. For example, key points such as "This presentation explains different application areas of AI" are extracted.

[1588] 4. Question generation: Based on the analysis results, questions are generated using generative AI (such as OpenAI's GPT-4). The generated questions are displayed in real time on the viewer's device. For example, a question such as, "What specific changes can we expect to see in the medical field through the use of AI?"

[1589] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotion data to the server. For example, if the viewer has an interested expression, the data will be collected by the emotion engine and sent to the server.

[1590] Prompt Sentence Examples

[1591] "Generate a question based on the following presentation: AI has achieved remarkable results in the field of medical imaging diagnosis. In particular, AI algorithms are now able to detect cellular abnormalities with high accuracy. This has significantly improved the speed and accuracy of diagnoses and reduced the burden on doctors."

[1592] This allows viewers to post questions based on their own interests and also vote on questions generated by other viewers. The server dynamically prioritizes questions based on the collected voting and sentiment data, and provides the presenter with the most relevant questions.

[1593] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1594] Processing flow and step details

[1595] Step 1: Acquiring and sending audio data

[1596] The device records the presenter's voice with a microphone and sends the audio data to the server. Specifically, the audio during the presentation is recorded in real time and converted into a digital audio file. This audio file is then sent to the server via the Internet.

[1597] Input: Speaker's voice

[1598] Data processing: audio recording and digital conversion

[1599] Output: Digital audio file (sent to server)

[1600] Step 2: Capture and send slide images

[1601] The device uses a screen capture tool to capture image data of the slides and sends it to the server. When the presenter switches slides, the device automatically performs a screen capture and sends the captured image to the server.

[1602] Input: Slide screen

[1603] Data processing: Screen capture

[1604] Output: Image data file (sent to server)

[1605] Step 3: Convert audio data to text

[1606] The server receives the voice data and converts it into text using a speech recognition engine (e.g., Google Cloud Speech-to-Text). The digital voice file sent to the server is analyzed and converted by the speech recognition engine and saved as a string of characters.

[1607] Input: Digital audio data

[1608] Data processing: Speech recognition (voice to text conversion)

[1609] Output: Text data

[1610] Step 4: Extract text from slide images

[1611] The server extracts key text from the slide images using an image recognition engine (e.g., Google Cloud Vision). It analyzes the received image data files to identify and extract text within the images.

[1612] Input: Slide image data

[1613] Data processing: Image recognition (text extraction)

[1614] Output: Text data

[1615] Step 5: Analyze the presentation

[1616] The server integrates the text data obtained from the speech and image recognition and analyzes the presentation using natural language processing (NLP) technology. Based on the integrated text data, the main points and key keywords of the presentation are identified.

[1617] Input: Text data extracted from audio and slides

[1618] Data Processing: Integration and NLP Analysis

[1619] Output: Gist and keyword data

[1620] Step 6: Generate questions

[1621] The server uses a generative AI model (e.g., GPT-4) to generate questions based on the analysis results. The server uses the key points and keywords obtained as prompts to generate appropriate questions.

[1622] Input: Key points and keyword data

[1623] Data processing: Question generation (generative AI model)

[1624] Output: Question text

[1625] Step 7: Posing the Question

[1626] The server presents the generated questions to the viewer's terminal in real time, and transmits the generated questions to the viewer's device and provides them through an interface for display.

[1627] Input: Question text

[1628] Data processing: Submitting and presenting questions

[1629] Output: Question displayed on the viewer's device

[1630] Step 8: Collect and send emotion data

[1631] The emotion engine analyzes the viewer's facial expressions and tone of voice in real time and sends the emotional data to the server. The camera and microphone collect the viewer's emotions and send them to the server as digital data.

[1632] Input: Viewer facial expressions and tone of voice

[1633] Data processing: Sentiment analysis

[1634] Output: Emotion data (sent to server)

[1635] Step 9: Viewer polls

[1636] Users (viewers) vote on questions presented through their terminals. Each viewer votes for the questions they are interested in, and the vote data is sent to the server in real time.

[1637] Input: Viewer votes

[1638] Data processing: Collecting voting data

[1639] Output: Voting data (sent to server)

[1640] Step 10: Prioritize and present questions

[1641] The server dynamically determines the priority of questions based on the collected voting data and sentiment data, and selects the most important question and displays it on the presenter's device.

[1642] Input: Voting data and sentiment data

[1643] Data processing: data integration and prioritization

[1644] Output: Top priority question (displayed to presenter's device)

[1645] The above processing steps enable an effective and engaging Q&A session between the presenter and the audience.

[1646] (Application example 2)

[1647] 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."

[1648] Conventional presentation support systems and robot work support systems have problems in that they are unable to fully realize Q&A during presentations or improve the efficiency of robot work in factories. Specifically, they do not properly manage the priority of audience questions, and do not immediately adjust the robot's behavior in response to the worker's instructions or emotions, hindering efficient work. This results in low satisfaction for viewers and workers, and a lack of two-way communication.

[1649] 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.

[1650] In this invention, the server includes means for generating questions based on the content of the presentation, means for presenting the generated questions to viewer terminals, means for viewers to post questions, means for viewers to vote on the presented questions, means for tallying up vote results and determining priorities, means for acquiring robot work status data, means for acquiring voice instructions from workers and dynamically optimizing work content using a generation AI, and means for acquiring worker emotion data and adjusting work plans based on the emotion data. This improves the efficiency of question and answer sessions between viewers and presenters, enabling more efficient robot work in factories.

[1651] "Presentation audio data" refers to recorded data of the audio uttered by the presenter during the presentation.

[1652] "Slide images" are image data of visual slides used during a presentation.

[1653] A "voice recognition engine" is software or hardware for converting voice data into text data.

[1654] An "image recognition engine" is software or hardware for extracting key text from image data.

[1655] "Integration" is the process of combining multiple different data sets into a single dataset.

[1656] "Question generation" is the process of automatically creating appropriate questions based on the presentation content and analysis results.

[1657] "Viewer device" means the device used by the viewer to view the presentation.

[1658] "Voting results" are the aggregated results of votes cast by viewers on questions.

[1659] "Priority" is the measure by which a particular question or task is judged to be more important than other questions or tasks.

[1660] "Robot work status data" refers to data relating to the work currently being performed by the robot and its status.

[1661] "Generative AI" is artificial intelligence that generates appropriate questions and answers based on given data and conditions.

[1662] "Emotion data" refers to emotional information obtained from the worker's facial expressions, tone of voice, etc.

[1663] "Dynamic optimization" is the process of adjusting to the optimal work plan in real time as conditions change.

[1664] A "work plan" is the design of procedures and schedules for carrying out a specific task.

[1665] This invention relates to a "robot work support system" for improving the efficiency of robot work in factories. This system combines a function to support questions and answers during presentations with a function to dynamically optimize robot work in factories, thereby increasing the satisfaction of viewers and workers.

[1666] System Overview

[1667] Terminal

[1668] The terminal uses the microphone and camera built into the robot body to capture the worker's voice instructions and work status in real time. The smart glasses or tablet used by the worker also function as part of the system.

[1669] server

[1670] The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text) to convert the acquired voice data into text data, an image recognition engine (e.g., Amazon Rekognition) to analyze image data acquired from the robot's camera, and an NLP technology (e.g., spaCy) to integrate the results of speech recognition and image recognition, and a generative AI (e.g., OpenAI GPT) to generate appropriate questions and work plans.

[1671] Emotion Engine

[1672] An emotion engine (e.g., Microsoft Azure Emotion API) is used to analyze workers' facial expressions and voice tones in real time to obtain emotional data, which is used to dynamically adjust the work plan.

[1673] Specific processing examples

[1674] 1. Acquiring and processing audio data

[1675] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits them to the server. The server then converts the voice data into text data using a voice recognition engine. During this process, an instruction such as "Please install the next part" is obtained as text data.

[1676] 2. Image data acquisition and processing

[1677] The terminal uses the robot's built-in camera to capture images of the work status and parts, and sends them to the server. The server then analyzes the image data using an image recognition engine to extract key information. For example, specific parts and work status may be recognized.

[1678] 3. Analysis and adjustment of work content

[1679] The server integrates and analyzes the voice and image recognition results, and uses generative AI to generate questions and work plans, which are then presented to the worker's smart glasses or tablet.

[1680] 4. Acquiring and Processing Emotional Data

[1681] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, capturing emotional data. For example, if the worker is feeling dissatisfied, the data is sent to the server in real time and used to adjust the work plan.

[1682] Prompt Sentence Examples

[1683] "When an AI robot follows instructions in a factory task, suggest ways to optimize the task. Voice command: 'Install the next part.' Worker emotion: Frustration evident."

[1684] This allows for efficient question and answer sessions between the audience and the presenter, while also dynamically optimizing the work of robots in factories, improving worker satisfaction and work efficiency.

[1685] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1686] Step 1:

[1687] The terminal uses the robot's built-in microphone to record the worker's voice instructions and transmits the voice data to the server. The input is the worker's voice instructions, and the output is the voice data sent to the server. Specifically, the microphone captures the voice, which is then saved and transmitted as digital data.

[1688] Step 2:

[1689] The server uses a speech recognition engine to convert the received voice data into text data. The input is voice data, and the output is text data corresponding to the voice instructions. Specifically, the speech recognition engine analyzes the voice data and generates text using a language model.

[1690] Step 3:

[1691] The terminal uses the robot's built-in camera to acquire image data of the work situation and sends that image data to the server. The input is visual information of the work situation, and the output is image data sent to the server. Specifically, the camera captures the image, saves it as digital data, and sends it.

[1692] Step 4:

[1693] The server uses an image recognition engine to analyze the received image data and extract key information. The input is the image data, and the output is extracted key text and recognized objects. Specifically, the image recognition engine analyzes the image data pixel by pixel to identify key text and objects.

[1694] Step 5:

[1695] The server integrates the results of speech recognition and image recognition and analyzes the work content using NLP technology. The input is text data and extracted key information, and the output is the gist of the analyzed work content. Specifically, the text data and image data are integrated into a single dataset and analyzed using an NLP model.

[1696] Step 6:

[1697] The server uses a generation AI to generate appropriate questions and work plans based on the analysis results. The input is the gist of the analyzed work content, and the output is the generated questions and work plans. Specifically, the generation AI generates questions and work plans related to the prompt.

[1698] Step 7:

[1699] The terminal presents the generated questions and work plans to the worker's smart glasses or tablet. The input is the generated questions and work plans, and the output is a visual display. The specific operation is to display the information on the device's display.

[1700] Step 8:

[1701] The emotion engine captures the worker's facial expressions using a camera built into the worker's smart glasses or tablet, and acquires emotion data. The input is video data of the worker's facial expressions, and the output is the emotion recognition results. Specifically, the camera captures the facial expressions, and the emotion recognition engine analyzes the video data.

[1702] Step 9:

[1703] The server dynamically adjusts the work plan based on the acquired emotion data. The input is emotion data, and the output is an adjusted work plan. Specifically, it analyzes the emotion recognition results and modifies the priority and procedures of the work plan in real time.

[1704] 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.

[1705] 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.

[1706] 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.

[1707] 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.

[1708] 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.

[1709] 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.

[1710] 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).

[1711] 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.

[1712] 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."

[1713] 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.

[1714] 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).

[1715] 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.

[1716] 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.

[1717] 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.

[1718] 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.

[1719] 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.

[1720] 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.

[1721] 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.

[1722] 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.

[1723] 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.

[1724] 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.

[1725] The following is further disclosed regarding the above embodiment.

[1726] (Claim 1)

[1727] A means of recording the audio data of the presentation;

[1728] a means for obtaining slide images of the presentation;

[1729] means for converting the voice data into text data;

[1730] means for analyzing the slide images and extracting key text;

[1731] A means for integrating the text data and the extracted key texts to analyze the presentation content;

[1732] a means for generating questions based on the presentation;

[1733] A means for presenting the generated question to a viewer's terminal;

[1734] A way for viewers to submit questions;

[1735] a means for viewers to vote on questions posed;

[1736] A system including a means for tabulating vote results and determining priorities.

[1737] (Claim 2)

[1738] The system according to claim 1, further comprising means for receiving questions posted by viewers and presenting them in a manner indistinguishable from questions generated by AI.

[1739] (Claim 3)

[1740] 10. The system of claim 1, further comprising means for presenting the question with the most votes to the presenter.

[1741] "Example 1"

[1742] (Claim 1)

[1743] A means of recording the audio data of the presentation;

[1744] a means for obtaining slide images of the presentation;

[1745] means for converting the voice data into text data;

[1746] means for analyzing the slide images and extracting key text;

[1747] A means for integrating the text data and the extracted key texts to analyze the presentation content;

[1748] A means for generating questions using a generative model based on the presentation content;

[1749] A means for presenting the generated question to a viewer's terminal;

[1750] A way for viewers to submit questions;

[1751] a means for viewers to vote on questions posed;

[1752] A means of tallying vote results in real time and determining priorities;

[1753] A system that includes a means for notifying presenters of high priority questions.

[1754] (Claim 2)

[1755] 2. The system of claim 1, further comprising means for receiving questions posted by viewers and presenting the questions indistinguishable from questions generated by the generative model.

[1756] (Claim 3)

[1757] 10. The system of claim 1, further comprising means for presenting the question with the most votes to the presenter.

[1758] "Application Example 1"

[1759] (Claim 1)

[1760] A means of recording the audio data of the presentation;

[1761] a means for obtaining slide images of the presentation;

[1762] means for converting the voice data into text data;

[1763] means for analyzing the slide images and extracting key text;

[1764] A means for integrating the text data and the extracted key texts to analyze the presentation content;

[1765] a means for generating questions based on the presentation;

[1766] means for presenting the generated question to a user's terminal;

[1767] a means for users to post questions;

[1768] a means for users to vote on questions posed;

[1769] a means for tabulating the votes and determining priorities; and

[1770] A method for automatically generating and presenting questions from customers by analyzing the audio and video of product demonstrations,

[1771] A means of presenting generated questions to the user's device in real time

[1772] A system including:

[1773] (Claim 2)

[1774] The system of claim 1, further comprising means for receiving questions posted by viewers and presenting them in a manner indistinguishable from questions generated by AI.

[1775] (Claim 3)

[1776] 10. The system of claim 1, further comprising means for presenting the question with the most votes to the presenter.

[1777] "Example 2: Combining Emotion Engines"

[1778] (Claim 1)

[1779] A means of recording the audio data of the presentation;

[1780] a means for obtaining slide images of the presentation;

[1781] means for converting voice data into text data;

[1782] means for analyzing the slide images and extracting key text;

[1783] A means for integrating the text data and extracted key texts to analyze the presentation content;

[1784] using a generative AI model to generate questions based on the presentation;

[1785] a means for presenting the generated question to a viewer terminal;

[1786] a means for viewers to submit questions and vote on the questions posed;

[1787] a means for tabulating the votes and determining the priority of questions; and

[1788] Using an emotion engine that analyzes the viewer's facial expressions and tone of voice;

[1789] The system includes a means for dynamically adjusting the priority of questions based on sentiment data.

[1790] (Claim 2)

[1791] The system of claim 1, further comprising a means for receiving questions posted by viewers and presenting them in a manner indistinguishable from questions generated by the generation AI.

[1792] (Claim 3)

[1793] 10. The system of claim 1, further comprising means for presenting the question with the most votes to the presenter.

[1794] "Application example 2 when combining emotion engines"

[1795] (Claim 1)

[1796] A means of recording the audio data of the presentation;

[1797] a means for obtaining slide images of the presentation;

[1798] means for converting the voice data into text data;

[1799] means for analyzing the slide images and extracting key text;

[1800] A means for integrating the text data and the extracted key texts to analyze the presentation content;

[1801] a means for generating questions based on the presentation;

[1802] A means for presenting the generated question to a viewer's terminal;

[1803] A way for viewers to submit questions;

[1804] a means for viewers to vote on questions posed;

[1805] a means for tabulating the votes and determining priorities; and

[1806] A means for acquiring working state data of the robot;

[1807] A means for acquiring voice instructions from workers and dynamically optimizing the work content using generation AI;

[1808] A system including means for acquiring emotion data of a worker and adjusting a work plan based on the emotion data.

[1809] (Claim 2)

[1810] The system of claim 1, further comprising a means for receiving questions posted by viewers and presenting them in a manner indistinguishable from questions generated by the generation AI.

[1811] (Claim 3)

[1812] 10. The system of claim 1, further comprising means for presenting the question with the most votes to the presenter. [Explanation of symbols]

[1813] 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. A means of recording the audio data of the presentation; a means for obtaining slide images of the presentation; means for converting the voice data into text data; means for analyzing the slide images and extracting key text; A means for integrating the text data and the extracted key texts to analyze the presentation content; a means for generating questions based on the presentation; A means for presenting the generated question to a viewer's terminal; A way for viewers to submit questions; a means for viewers to vote on questions posed; A system including a means for tabulating vote results and determining priorities.

2. The system according to claim 1, further comprising means for receiving questions posted by viewers and presenting the questions in a manner indistinguishable from questions generated by AI.

3. 2. The system of claim 1, further comprising means for presenting the question with the most votes to the presenter.

Images