System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating and distributing chat content according to an interest of a viewer during an online event using a generative AI model; terminals for displaying the generated chat content; and means for collecting a reaction from the viewer and feeding back the reaction to the AI model in real time.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] With the increase in online events, there tends to be a lack of interactive communication with viewers, making it difficult for viewers to maintain their interest in the event content. There are also issues with limited chat from viewers and difficulty in responding to unexpected questions. These issues are factors that reduce the effectiveness of the event and the viewer's level of understanding. [Means for solving the problem]

[0005] The present invention provides a system that solves the above-mentioned problems by using a generative AI model. Specifically, it provides a means for the generative AI model to generate and distribute chat content tailored to viewer interests during an online event. It also includes a device that displays the generated chat content and a means for collecting viewer responses and providing real-time feedback to the AI ​​model. This increases viewer interest and promotes two-way communication. It also provides a means for setting various personas and for the generative AI model to generate chat content based on those personas, and updates the persona setting information based on viewer responses. This enables communication tailored to each viewer. It also provides a means for saving viewer response data, identifying areas for improvement for the next event, and distributing a feedback form to viewers after the event. This allows for continuous improvement of the quality of events.

[0006] A "generative AI model" is an artificial intelligence algorithm for generating natural language, a system that generates meaningful sentences based on text data.

[0007] "Online events" refer to events, seminars, conferences, etc. that are streamed in real time via the Internet.

[0008] "Viewers" refer to users who are watching an online event and are participants in the event.

[0009] "Chat content" refers to information such as messages, questions, and comments exchanged in text format during an online event.

[0010] "Device" refers to the device used by a viewer to access an online event and to view and enter chat content, including, but not limited to, a computer, smartphone, or tablet.

[0011] "Reaction" refers to a viewer's reaction or response to a chat, including sending a message or providing feedback.

[0012] A "persona" is a virtual character that is designed to represent a user with specific interests and behavioral patterns.

[0013] "Feedback Form" refers to an online survey-style form that viewers can fill out after the event to provide their opinions and impressions. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] This invention is a system that uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[0036] Processing performed by the server

[0037] 1. Loading event data

[0038] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[0039] 2. ChatGPT model initialization and context preparation

[0040] The server initializes the generative AI model (ChatGPT) and passes the loaded event details and persona setting information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[0041] 3. Generating and delivering the first chat

[0042] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[0043] 4. Real-time chat generation

[0044] During the event, the server collects audience reactions and feeds them back to the ChatGPT model in real time, generating new questions and comments that can continuously increase audience engagement.

[0045] 5. Data storage and feedback collection

[0046] After the event is over, the server stores all viewer response data and sends a feedback form to the viewer to collect suggestions for improving the next event.

[0047] Processing performed by the device

[0048] 1. Displaying chat contents

[0049] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[0050] 2. Receiving and sending user input

[0051] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[0052] 3. Providing a feedback form

[0053] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0054] User actions

[0055] 1. View and respond to chat

[0056] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[0057] 2. Submit your own questions or comments

[0058] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[0059] 3. Providing Feedback

[0060] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[0061] Specific examples

[0062] Example 1: Starting an event

[0063] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0064] Terminal: The terminal displays the questions sent from the server in real time.

[0065] User: The user is interested in the question displayed and enters an additional question.

[0066] Example 2: Interaction during an event

[0067] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[0068] Terminal: The terminal will immediately display the new question as it is generated.

[0069] User: The user responds again to the new question and sends a chat.

[0070] Example 3: After the event

[0071] Server: After the event, the server stores all the response data and delivers the feedback form.

[0072] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0073] User: The user fills out a feedback form and submits it.

[0074] This allows the system of the present invention to facilitate two-way communication at online events and increase audience engagement.

[0075] The processing flow will be explained below.

[0076] Step 1:

[0077] The server loads the event details and persona configuration information from the database a few minutes before the start of the online event, capturing basic information about the event theme, speakers, and topics covered.

[0078] Step 2:

[0079] The server initializes the ChatGPT model with the loaded event details and persona configuration information to prepare the context, so that the AI ​​understands the context to generate questions and comments about the event.

[0080] Step 3:

[0081] At the start of an event, the server sends a request to the ChatGPT model to generate initial questions and comments, including those from different perspectives such as personas A, B, and C.

[0082] Step 4:

[0083] The server distributes the generated questions and comments to a real-time chat system, which displays the initial chat content on the viewers' devices.

[0084] Step 5:

[0085] The terminal displays the chat contents delivered from the server to the user, allowing the user to view questions and comments generated in real time.

[0086] Step 6:

[0087] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[0088] Step 7:

[0089] The device sends the user's input responses and questions to the server, and the data sent becomes important information that enhances the interactivity of the event.

[0090] Step 8:

[0091] The server receives the reaction data sent by users and inputs it into the ChatGPT model as feedback, which allows the AI ​​to improve the accuracy of the questions and comments it generates next.

[0092] Step 9:

[0093] The server then sends a request to the ChatGPT model to generate new questions and comments, and distributes the generated content to the real-time chat system. This process is repeated throughout the event.

[0094] Step 10:

[0095] After the event, the server stores all reaction data in a database, which helps identify improvements for future events and audience interests.

[0096] Step 11:

[0097] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[0098] Step 12:

[0099] The terminal causes the feedback form to be displayed and easily accessible for the user to complete.

[0100] Step 13:

[0101] Users can enter their opinions and thoughts in the feedback form and press the submit button. The feedback will help improve the quality of the next event.

[0102] Step 14:

[0103] The device forwards the feedback sent by the user to the server, which stores it in a database and uses it for analysis.

[0104] Example 1

[0105] 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."

[0106] With conventional online events, it was difficult to engage with viewers in two-way communication that matched their interests, making it difficult to encourage active participation. Furthermore, there was a lack of a system for effectively collecting viewers' reactions and feedback and reflecting that information in future events.

[0107] 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.

[0108] In this invention, the server includes means for generating and distributing chat content according to viewer interests during an online event, a display device for displaying the generated chat content, means for collecting viewer reactions and providing real-time feedback to the generative AI model, means for acquiring detailed event information and persona setting information in advance, and means for generating new questions and comments in real-time based on viewer reactions during the online event, thereby attracting viewer interest, effectively promoting two-way communication, and improving the quality of future events.

[0109] A "generative AI model" is an artificial intelligence system designed to perform tasks such as text generation.

[0110] An "online event" is an event such as a lecture, seminar, or conference that is held in real time via the Internet.

[0111] "Viewer" means a person who participates in an online event and watches or listens to the content in real time.

[0112] "Chat content" is a collection of statements, questions, and comments displayed in text format.

[0113] A "display device" is a device used by a user to view the generated chat content, and includes a personal computer, a smartphone, a tablet, and the like.

[0114] "Reactions" are reactions such as comments, questions, and opinions that viewers show during the event.

[0115] "Feedback" refers to information such as opinions, ratings, and impressions collected from viewers.

[0116] "Event details" refers to information about the online event, such as the theme, speakers, and topics.

[0117] "Persona setting information" is information about a virtual character based on the viewer's interests and attributes.

[0118] "New questions and comments" are new textual statements or questions that are generated based on viewer reactions.

[0119] This system uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[0120] Server Roles

[0121] The server loads online event details and persona configuration information from a database such as MySQL. This information includes the event theme, speakers, planned topics, and audience persona information. The server then initializes OpenAI's ChatGPT model and passes it the event details and persona information. Once this is ready, the server generates the initial chat content and distributes it to the chat system in real time.

[0122] For example, the server loads information about a webinar on AI technology and sends a prompt to ChatGPT to ask viewers what topic in AI technology they are most interested in. The generated questions are then sent to the display device in real time.

[0123] Device Role

[0124] The device displays chat content sent from the server in real time, allowing viewers to easily view the questions and comments that have been generated. When viewers type responses to chats or questions, the device sends the information to the server. After the event ends, a feedback form is displayed, allowing viewers to easily enter their opinions and thoughts.

[0125] As a specific example, a viewer types a response into their smartphone, such as "I'm interested in image recognition using AI," and this is sent to the server.

[0126] User Roles

[0127] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, stimulating two-way communication. Users can also enter their own concerns or questions and submit them to the chat system. After the event, users can enter their opinions and impressions in a feedback form and submit it to help improve the quality of the next event.

[0128] As a concrete example, a user types the question "How does AI learn?" and sends it to the chat.

[0129] Prompt Sentence Examples

[0130] Below is an example of a prompt sentence.

[0131] "What topic in AI technology interests you most?"

[0132] "Please tell us about the latest technological trends using AI."

[0133] What was the most memorable aspect of what was said at the event?

[0134] In this system, the server collects viewer reactions and feeds them back into the ChatGPT model to generate new questions and comments in real time. After the event ends, the server saves all viewer reaction data and sends a feedback form. Through this process, the system of the present invention can increase viewer interest and effectively promote two-way communication.

[0135] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0136] Step 1: Loading event data

[0137] Specifically: Before the start of an online event, the server loads the event details (theme, speakers, topics, etc.) and persona configuration information from a database, which is necessary to generate questions and comments related to the event content.

[0138] Input: Database connection information (e.g. MySQL query), event identification information.

[0139] Data processing: Executes SQL queries to retrieve data about events and loads it into memory.

[0140] Output: A data object containing event details and persona configuration information.

[0141] Specific operation: The server executes an SQL query such as "SELECT FROM events WHERE id=1" to retrieve information about the "2023 AI Technology Webinar."

[0142] Step 2: Initializing the ChatGPT model and preparing the context

[0143] Specifically: The server initializes the ChatGPT model with the retrieved event details and persona configuration information, allowing the model to have the right context to generate questions and comments based on the event content.

[0144] Input: Event details and persona configuration information.

[0145] Data processing: Obtaining the API key and configuration parameters required for initialization, and passing context data to the ChatGPT model.

[0146] Output: The initialized ChatGPT model.

[0147] What happens: The server uses the openai.ChatCompletion.create method in Python to pass event information (e.g., the speaker is John Doe, and the topic is AI and medical technology) to the model.

[0148] Step 3: Generate and deliver the first chat

[0149] Specifically: When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate immediately.

[0150] Input: An initialized ChatGPT model, a trigger to start the event.

[0151] Data processing: Enter the prompt text into ChatGPT and format the generated data.

[0152] Output: Data from the initial question and comment.

[0153] How it works: The server generates a question from the ChatGPT model: "What topic are you most interested in in AI technology?" and sends it to the chat system via WebSocket or HTTP API.

[0154] Step 4: Real-time chat generation

[0155] During the event, the server collects audience reactions in real time and feeds them back to the ChatGPT model, allowing it to continuously generate new questions and comments.

[0156] Input: Viewer response data.

[0157] Data processing: Collect and analyze viewer response data and provide it to ChatGPT as feedback.

[0158] Output: Question and comment data updated in real time.

[0159] How it works: A viewer responds by saying, "I'm interested in AI image recognition," and the server then tells ChatGPT, "I'd like to know more about applications in the medical field." The model then generates a new question and distributes it to the chat system.

[0160] Step 5: Store data and collect feedback

[0161] After the event, the server will save all viewer response data and send a feedback form to the viewer, allowing us to collect their opinions for improving the next event.

[0162] Input: Event end trigger, viewer response data.

[0163] Data processing: Response data storage, feedback form generation and distribution.

[0164] Output: Saved data object, feedback form delivered to viewers.

[0165] What it does: The server saves all chat data from viewers in a MySQL database and sends feedback forms such as "How satisfied were you with the event?" via Google Forms or a dedicated application.

[0166] This is the specific processing flow of the program for this system. At each step, the necessary data processing and calculations are carried out based on clearly defined input data, and appropriate output is generated, promoting two-way communication with viewers.

[0167] (Application example 1)

[0168] 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."

[0169] In online events, there is a need for a method to maintain viewer interest, promote interactive communication, and collect event data to improve future events. In particular, there are insufficient methods for generating appropriate topics and questions in real time to encourage active viewer participation, and it is difficult to effectively collect feedback information after the event has ended.

[0170] 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.

[0171] In this invention, the server includes means for using a generative AI model to generate and distribute chat content based on viewer interests during an online event, means for collecting feedback after the event ends and sending it to the server, means for generating product-related questions and comments, a terminal for displaying the generated chat content, and means for collecting reactions from viewers and feeding them back to the AI ​​model in real time. This allows viewers to receive interesting chat content in real time, stimulating two-way communication, and effectively collecting and saving feedback from viewers, making it possible to identify areas for improvement for the next event.

[0172] A "generative AI model" is an artificial intelligence model that generates new text and content based on large amounts of data.

[0173] An "online event" is an event that takes place in real time over the Internet.

[0174] "Viewers" are users who participate in online events, view content, and ask questions.

[0175] "Chat content" refers to text generated by a generative AI model and used for two-way communication with viewers.

[0176] A "terminal" is a device that displays the generated chat content and receives input from viewers.

[0177] "Feedback Form" means an electronic form used to collect opinions and feedback from viewers after an event has concluded.

[0178] "Persona" refers to profile information of a typical audience for a particular event or service.

[0179] The "server" is a central computer system that loads event data, generates chat content, stores viewer reaction data, and so on.

[0180] "Products" means any products or services featured during the Online Event.

[0181] "Real time" means that various processes are carried out at the same time as the event is progressing.

[0182] "Response data" refers to information about interactions, such as comments and questions, that viewers make during an online event.

[0183] This invention is a system that uses generative AI models to generate chat content based on viewer interests at online events, promoting real-time two-way communication, and collecting and saving viewer feedback to help improve future events. The system consists of three main elements: a server, terminals, and users.

[0184] Processing performed by the server

[0185] 1. Loading event data

[0186] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[0187] 2. ChatGPT model initialization and context preparation

[0188] The server initializes a generative AI model (e.g., ChatGPT) and passes the loaded event details and persona configuration information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[0189] 3. Generating and delivering the first chat

[0190] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[0191] 4. Real-time chat generation

[0192] During the event, the server collects viewers' reactions and feeds them back to the ChatGPT model in real time to generate new questions and comments, which can continuously stimulate viewers' interest. It also generates product-related questions and comments, promoting two-way communication.

[0193] 5. Data storage and feedback collection

[0194] After the event is over, the server stores all viewer response data. After the event is over, a feedback form is sent to the viewer to collect suggestions for improvements to the next event. The collected feedback is stored on the server and used for the next event.

[0195] Processing performed by the device

[0196] 1. Displaying chat contents

[0197] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[0198] 2. Receiving and sending user input

[0199] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[0200] 3. Providing a feedback form

[0201] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0202] User actions

[0203] 1. View and respond to chat

[0204] Users can view the chat content generated by the generative AI model on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[0205] 2. Submit your own questions or comments

[0206] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[0207] 3. Providing Feedback

[0208] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[0209] Specific examples

[0210] Example 1: Starting an event

[0211] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0212] Terminal: The terminal displays the questions sent from the server in real time.

[0213] User: The user is interested in the question displayed and enters an additional question.

[0214] Example 2: Interaction during an event

[0215] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[0216] Terminal: The terminal will immediately display the new question as it is generated.

[0217] User: The user responds again to the new question and sends a chat.

[0218] Example 3: After the event

[0219] Server: After the event, the server stores all the response data and delivers the feedback form.

[0220] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0221] User: The user fills out a feedback form and submits it.

[0222] Specific examples of use

[0223] Example prompt sentence:

[0224] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[0225] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0226] Step 1:

[0227] Before the event begins, the server loads the event details and persona setting information from the database. Specifically, it obtains the event theme, speaker information, target topics, etc. This information serves as the basic data for the generative AI model to generate appropriate chat content. The input is the event ID, and the output is the event details and persona setting information.

[0228] Step 2:

[0229] The server initializes the ChatGPT model based on the acquired event details and persona configuration information. This information is input into the model as context data, and the generated chat content is prepared to be event-related. The input is the event details and persona configuration information, and the output is the initialized ChatGPT model.

[0230] Step 3:

[0231] When an event starts, the server sends a request to the ChatGPT model to generate the initial question and comment. The generative AI model generates appropriate questions and comments based on the provided context and returns them to the server. The server then distributes the generated content to the chat system so that viewers can view it. The input is the initialized ChatGPT model, and the output is the initial question and comment.

[0232] Step 4:

[0233] The terminal displays the chat content sent from the server in real time. Viewers can check the generated questions and comments on the terminal and make appropriate responses. The input is the chat content sent from the server, and the output is the displayed chat content.

[0234] Step 5:

[0235] Users input questions and comments in response to the chat content displayed on their devices. The input data is sent from the device to the server. The input is the user's question or comment, and the output is the data sent from the device to the server.

[0236] Step 6:

[0237] The server collects user responses in real time and feeds them back to the generative AI model. The model generates new questions and comments, promoting two-way communication. The input is user response data, and the output is the generated new questions and comments.

[0238] Step 7:

[0239] When the event ends, the server saves all the reaction data. This records what interactions took place during the event and serves as a reference for the next event. The input is all the reaction data after the event ends, and the output is the saved reaction data.

[0240] Step 8:

[0241] After the event, the server distributes a feedback form to viewers. Viewers fill out the form with their opinions and thoughts and send it to the server. The input is the feedback form distributed to the viewer, and the output is the filled-in feedback.

[0242] Step 9:

[0243] The device passes the feedback sent by the viewer to the server, which stores it. This feedback is used to identify improvements for the next event. The input is the viewer feedback, and the output is the stored feedback data.

[0244] For example:

[0245] Regarding prompt generation, by inputting the following prompt into the model, appropriate chat content is generated:

[0246] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[0247] 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.

[0248] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0249] Processing performed by the server

[0250] 1. Loading event data

[0251] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[0252] 2. ChatGPT model initialization and context preparation

[0253] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[0254] 3. Generating and delivering the first chat

[0255] When an event starts, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different perspectives (e.g., persona A, B, C), and are delivered to the chat system in real time.

[0256] 4. Emotion engine integration

[0257] The server collects viewer reaction data in real time and passes it to the emotion engine to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions.

[0258] 5. Real-time chat generation

[0259] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the emotion engine, and distributes the generated content in real time.

[0260] 6. Data storage and feedback collection

[0261] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[0262] Processing performed by the device

[0263] 1. Displaying chat contents

[0264] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[0265] 2. Receiving and sending user input

[0266] When a user enters a chat or a response to a question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[0267] 3. Providing a feedback form

[0268] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0269] User actions

[0270] 1. View and respond to chat

[0271] Users can view the chat content generated by ChatGPT on their device and respond to questions or comments that interest them.

[0272] 2. Submit your own questions or comments

[0273] Users can input their own concerns or questions and send them to the chat system.

[0274] 3. Providing Feedback

[0275] After the event, users can enter their opinions and thoughts in a feedback form and submit it, which will help improve the quality of future events.

[0276] Specific examples

[0277] Example 1: Starting an event

[0278] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0279] Terminal: The terminal displays the questions sent from the server in real time.

[0280] User: A user is interested in a question displayed and enters a new question or comment.

[0281] Example 2: Interaction during an event

[0282] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[0283] Terminal: The terminal will immediately display the new question as it is generated.

[0284] User: A user responds to new questions and sends chats.

[0285] Example 3: After the event

[0286] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[0287] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0288] User: The user fills out a feedback form and submits it.

[0289] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[0290] The processing flow will be explained below.

[0291] Step 1:

[0292] Before the online event starts, the server loads the event details and persona configuration information from the database, including the event theme, speakers, and topics covered.

[0293] Step 2:

[0294] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the questions and comments generated are appropriate for the event.

[0295] Step 3:

[0296] At the start of an event, the server sends a request to the ChatGPT model to generate the initial questions and comments, which include content based on different personas (e.g., personas A, B, C).

[0297] Step 4:

[0298] The server then broadcasts the generated questions and comments to a real-time chat system, allowing viewers to see the original chat transcript.

[0299] Step 5:

[0300] The terminal displays the chat contents sent from the server to the user, allowing the user to view questions and comments generated in real time.

[0301] Step 6:

[0302] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[0303] Step 7:

[0304] The device sends the user's input responses and questions to the server, and simultaneously sends the user's facial expressions and voice data to the emotion engine.

[0305] Step 8:

[0306] The emotion engine analyzes the user's facial and voice data to identify the user's emotional state, and sends the recognized emotion data to the server.

[0307] Step 9:

[0308] The server receives reaction data sent by users and emotion data from the emotion engine, and feeds this back to the ChatGPT model to improve the accuracy of the next questions and comments it generates.

[0309] Step 10:

[0310] The server again sends requests to the ChatGPT model to generate new questions and comments, and delivers them to the chat system in real time. This process is repeated throughout the event.

[0311] Step 11:

[0312] After the event, the server stores all reaction and sentiment data in a database, which allows for analysis and improvement for future events.

[0313] Step 12:

[0314] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[0315] Step 13:

[0316] The terminal displays a feedback form to the user for easy access and completion.

[0317] Step 14:

[0318] Users can enter their opinions and thoughts in the feedback form and press the submit button to send the data to the server, which will help improve the quality of the next event.

[0319] Step 15:

[0320] The server stores the submitted feedback data in a database and uses it for analysis to identify areas for improvement for the next event.

[0321] Example 2

[0322] 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."

[0323] To promote viewer engagement and interactive communication at online events, a system is needed that can generate chat content in real time based on viewer interests and emotions and instantly reflect their feedback. However, conventional systems do not adequately generate content based on viewer emotions or generate chat content from diverse perspectives using personas, resulting in a decline in viewer participation.

[0324] 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 a means for generating and distributing chat content according to viewer interests during an online event using a generative AI model, a terminal for displaying the generated chat content, a means for collecting reactions from viewers and feeding them back to the AI ​​model in real time, an engine for analyzing viewer emotions, and a means for improving the chat content based on the viewer emotions. This enables real-time chat generation and feedback according to viewer emotions and reactions.

[0325] A "generative AI model" is a type of artificial intelligence that automatically generates chat content based on viewers' interests and reactions.

[0326] An "online event" is an event or webinar that takes place live over the internet and allows participants to participate in it in real time.

[0327] "Viewers" are users who participate in an online event, who watch the event content, and participate through chat.

[0328] "Chat content" refers to messages such as questions and comments generated by the generative AI model, and is information used to communicate with viewers during online events.

[0329] "Terminal" refers to a device used by a viewer, such as a computer, smartphone, or tablet, that can display and input chat content.

[0330] "Reactions" refer to responses such as comments, questions, and facial expression analysis data made by viewers in response to chat content.

[0331] "Feedback" refers to the process of collecting and analyzing viewer reaction data, feeding the results back into the generative AI model, and reflecting them in the generation of the next chat content.

[0332] An "emotion engine" is a system that analyzes the viewer's facial expressions and voice data to determine their emotional state.

[0333] A "persona" is a virtual character or user profile created based on the viewer's interests and reactions, and is a factor taken into consideration when generating chat content.

[0334] "Feedback Form" means an input form provided to collect opinions and feedback from viewers after the online event has ended.

[0335] "Analysis" refers to the process of analyzing viewer reaction data and emotional data using statistical or machine learning techniques.

[0336] "Distribution" refers to the process of sending generated chat content, feedback forms, etc. to viewers in real time.

[0337] "Storage" refers to the process of storing data collected during an event in a recording medium such as a database.

[0338] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0339] server

[0340] The server handles the primary data processing for the online event, and specifically uses the following hardware and software:

[0341] Hardware: High-performance server machine (e.g., a server equipped with an Intel Xeon processor)

[0342] Software: Cloud infrastructure using AWS or Google Cloud, database using MySQL or PostgreSQL, ChatGPT as the generative AI model, and Affectiva as the sentiment analysis engine

[0343] 1. Loading event data

[0344] Before the event starts, the server loads the event details (e.g., event name, start time, speaker information) and persona setting information from the database, thereby obtaining information about the event theme and speakers.

[0345] 2. ChatGPT model initialization and context preparation

[0346] Based on the loaded information, we initialize the ChatGPT model and set the context, which allows the model to generate questions and comments related to the event.

[0347] 3. Generating and delivering the first chat

[0348] Once the event begins, the ChatGPT model generates the initial questions and comments, which are then streamed in real time.

[0349] 4. Emotion engine integration

[0350] The server collects viewer reaction data in real time and passes it to the emotion engine for analysis. The analysis results are fed back to the ChatGPT model to refine the chat content based on viewer emotions.

[0351] 5. Real-time chat generation

[0352] Based on viewer reaction data and the analysis results of the emotion engine, the server uses the ChatGPT model to generate new questions and comments and distribute them in real time.

[0353] 6. Data storage and feedback collection

[0354] After the event, all reaction and sentiment data is stored in a database, and a feedback form is sent to viewers to collect suggestions for improvements to be made to the next event.

[0355] Terminal

[0356] A terminal is a device used by a viewer and has the following functions:

[0357] Hardware: PCs, smartphones, tablets

[0358] Software: Web browser or dedicated application

[0359] 1. Displaying chat contents

[0360] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[0361] 2. Receiving and sending user input

[0362] When a user enters a response to a chat or question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[0363] 3. Providing a feedback form

[0364] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0365] User

[0366] Users participate in online events and interact with the system using their devices, performing the following actions in the process:

[0367] 1. View and respond to chat

[0368] Users can view the chat content generated by ChatGPT on their device and respond to any content that interests them.

[0369] 2. Submit your own questions or comments

[0370] Users can input their own concerns or questions and send them to the chat system.

[0371] 3. Providing Feedback

[0372] After the event, users can enter their opinions and thoughts in a feedback form to help improve the quality of the next event.

[0373] Specific examples

[0374] Below is a concrete example of how this system can be actually operated.

[0375] Example 1: Starting an event

[0376] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0377] Terminal: The terminal displays the questions sent from the server in real time.

[0378] User: A user is interested in a question displayed and enters a new question or comment.

[0379] Example 2: Interaction during an event

[0380] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[0381] Terminal: The terminal will immediately display the new question as it is generated.

[0382] User: A user responds to new questions and sends chats.

[0383] Example 3: After the event

[0384] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[0385] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0386] User: The user fills out a feedback form and submits it.

[0387] Prompt Sentence Examples

[0388] "Thank you for joining us for our virtual event today. Please answer the following question: What do you think about the theme of the event?"

[0389] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[0390] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0391] Step 1:

[0392] Loading event data

[0393] Before an online event begins, the server loads event details and persona configuration information from a database.

[0394] Input: Event ID

[0395] Operation:

[0396] The server executes the SQL query "SELECT FROM EventDetails WHERE EventID = X".

[0397] The server obtains detailed information about the event (e.g., event name, start time, speaker information) and persona setting information (e.g., characteristics of Persona A and Persona B).

[0398] Output: Event details, persona setting information

[0399] Step 2:

[0400] ChatGPT model initialization and context preparation

[0401] The server initializes the ChatGPT model and sets the context based on the loaded event details and persona configuration information.

[0402] Input: Event details, persona setting information

[0403] Operation:

[0404] The server calls the "init_model(event_details, persona_info)" method to initialize the ChatGPT model.

[0405] After initialization, the model is provided with an initial prompt appropriate to the event content to set the context.

[0406] Output: Initialized ChatGPT model

[0407] Step 3:

[0408] First chat generation and distribution

[0409] When an event starts, the server will generate the initial questions and comments from the ChatGPT model, and the chat will be streamed in real time.

[0410] Input: Initialized ChatGPT model

[0411] Operation:

[0412] The server calls the "generate_initial_chat(model, context)" method to generate the initial chat content.

[0413] The generated chat content is sent to the chat system using the "send_to_chat_system(chat_content)" method.

[0414] Output: First chat

[0415] Step 4:

[0416] Collecting reactions from viewers

[0417] The device collects chat content and reactions from viewers in real time and sends them to the server.

[0418] Input: Viewer input data (e.g., chat messages, facial expression analysis data)

[0419] Operation:

[0420] The terminal executes the "receive_user_input()" method to collect input from the user.

[0421] The collected data is sent to the server using the "send_to_server(user_input)" method.

[0422] Output: Viewer input data

[0423] Step 5:

[0424] Emotion engine collaboration

[0425] The server passes viewer reaction data to the emotion engine in real time to analyze user emotions.

[0426] Input: Viewer input data

[0427] Operation:

[0428] The server sends reaction data to the emotion engine using the "send_to_emotion_engine(reaction_data)" method.

[0429] The emotion engine analyzes the data and returns emotion results to the server.

[0430] Output: Emotion analysis results

[0431] Step 6:

[0432] Real-time chat generation

[0433] The server uses the ChatGPT model to generate new questions and comments based on viewer reaction data and the analysis results of the emotion engine, and distributes them in real time.

[0434] Input: Sentiment analysis results, viewer input data

[0435] Operation:

[0436] The server executes the "generate_new_chat(model, feedback)" method to generate new chat content.

[0437] The generated new chat content is distributed using "send_to_chat_system(chat_content)".

[0438] Output: New chat transcript

[0439] Step 7:

[0440] Data storage and feedback collection

[0441] After the event, the server stores all reaction and emotion data in a database and delivers a feedback form to viewers.

[0442] Input: All data after the event

[0443] Operation:

[0444] The server saves all data to the database using "save_data_to_database(data)".

[0445] Send the feedback form to viewers using "send_feedback_form()".

[0446] Output: Saved data, delivered feedback form

[0447] (Application example 2)

[0448] 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."

[0449] Conventional online events lacked mechanisms to effectively promote interactive communication with viewers. As a result, it was difficult to respond immediately to viewers' interests and reactions, resulting in a decline in viewer engagement. Furthermore, there was no technology that could analyze viewers' emotions in real time and dynamically improve chat content using generative AI models based on that analysis. This made it difficult to provide optimal content that met viewers' expectations.

[0450] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for using a generative AI model to generate and distribute chat content tailored to viewer interests during an online event; a terminal for displaying the generated chat content; means for collecting viewer reactions and providing feedback to the AI ​​model in real time; means for analyzing viewer emotions using an emotion analysis engine and improving the generated chat content based on the analysis results; means for constantly generating new questions and comments based on viewer reactions and distributing them in real time; and means for distributing a feedback form to viewers after the event ends and identifying areas for improvement for the next event. This makes it possible to significantly increase viewer engagement by immediately responding to viewer emotions and reactions and providing optimal chat content.

[0451] A "generative AI model" is an artificial intelligence model that takes data as input and automatically generates content based on specific contexts and conditions.

[0452] "Chat content" refers to text messages such as questions and comments that are generated by a generative AI model during an online event and distributed to viewers.

[0453] A "terminal" is a device that displays the generated chat content, accepts input from viewers, and transmits it to the server.

[0454] "Viewer reactions" refers to feedback such as facial expressions, gestures, voice, and text comments given by viewers during an online event.

[0455] "Real-time feedback" refers to a mechanism that instantly senses viewer reactions and sends them to a generative AI model for use in improving chat content.

[0456] An "emotion analysis engine" is software or hardware that analyzes emotions from viewers' facial expressions and voices and provides the results to a generative AI model.

[0457] A "persona" is a virtual character based on a specific audience or user profile, which the generative AI model uses as a reference when generating chat content.

[0458] A "feedback form" is a form provided to viewers after an online event to allow them to enter their evaluation of the event and suggestions for improvement.

[0459] This invention is a system that uses a generative AI model and an emotion analysis engine to generate chat content in real time during online events based on viewer interests, promoting two-way communication with viewers. This system consists of four main components: a server, terminals, viewers, and an emotion analysis engine.

[0460] Processing performed by the server

[0461] 1. Loading event data

[0462] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[0463] 2. ChatGPT model initialization and context preparation

[0464] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[0465] 3. Generating and delivering the first chat

[0466] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different personas (e.g., persona A, persona B, etc.) and are distributed to the chat system in real time.

[0467] 4. Emotion engine integration

[0468] The server collects viewer reaction data in real time and passes it to a sentiment analysis engine to analyze user emotions. The analysis results are fed back to the ChatGPT model, which refines the generated chat content according to viewer emotions.

[0469] 5. Real-time chat generation

[0470] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the sentiment analysis engine, and distributes the generated content in real time.

[0471] 6. Data storage and feedback collection

[0472] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[0473] Processing performed by the device

[0474] 1. Displaying chat contents

[0475] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[0476] 2. Receiving and sending user input

[0477] When viewers chat or respond to questions, the device sends that information to the server, along with facial expression and voice analysis data from the viewers to the emotion analysis engine.

[0478] 3. Providing a feedback form

[0479] After the event, the device will display a feedback form, allowing viewers to easily enter their opinions and thoughts.

[0480] User actions

[0481] 1. View and respond to chat

[0482] Viewers can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them.

[0483] 2. Submit your own questions or comments

[0484] Viewers can enter their own concerns or questions and send them to the chat system.

[0485] 3. Providing Feedback

[0486] After the event, viewers can fill out a feedback form and submit their thoughts and opinions, which will help improve the quality of future events.

[0487] Hardware / Software used

[0488] Server: AWS EC2 instance, or other cloud-based virtual machine

[0489] Sentiment Analysis Engine: Affectiva SDK, or other sentiment analysis platform

[0490] Chat display devices: smartphones, tablets, PCs

[0491] Real-time databases: Firebase Realtime Database, MongoDB

[0492] Generative AI model: OpenAI GPT-4 API or other advanced generative AI models

[0493] Prompt Sentence Examples

[0494] "User emotion is {emotion}. Event data is:\n{event_data}\nPlease generate an appropriate question."

[0495] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0496] Step 1:

[0497] Before the online event starts, the server loads the event details and persona information from the database, which provides basic information about the event theme, speakers, and topics covered. The input is the event ID, and the output is the event details and persona information.

[0498] Step 2:

[0499] The server initializes the ChatGPT model based on the loaded event details and persona settings, and prepares the context. This provides the context for the generative AI model to generate questions and comments related to the event. The input is the event details and persona settings, and the output is the initialized ChatGPT model.

[0500] Step 3:

[0501] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments. The generated questions and comments reflect different personas (e.g., Persona A, Persona B) and are delivered to the chat system in real time. The input is the initialized ChatGPT model, and the output is the generated initial questions and comments.

[0502] Step 4:

[0503] The terminal displays the chat content sent from the server in real time. Viewers view and respond to generated questions and comments. The input is the chat content sent from the server, and the output is the chat content that viewers see on their terminal.

[0504] Step 5:

[0505] When viewers enter responses to chats or questions, the device sends that information to the server. It also sends the viewer's facial expressions and voice analysis data to the emotion analysis engine. The input is the viewer's responses and emotion data, and the output is the data sent to the server and the emotion analysis engine.

[0506] Step 6:

[0507] The server sends viewer reaction data to the sentiment analysis engine in real time to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions. The input is viewer reaction data and sentiment analysis results, and the output is refined chat content.

[0508] Step 7:

[0509] The server sends a request to the ChatGPT model to generate new questions and comments based on the viewer reaction data and the analysis results from the sentiment analysis engine, and distributes the generated content in real time. The input is the analysis results and the ChatGPT model, and the output is new questions and comments generated in real time.

[0510] Step 8:

[0511] After the event, the server stores all the reaction and emotion data in a database and sends a feedback form to the viewers to collect improvements for the next event. The input is the viewer's reaction and emotion data, and the output is the stored data and the feedback form.

[0512] 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.

[0513] 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.

[0514] 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.

[0515] [Second embodiment]

[0516] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0517] 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.

[0518] 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).

[0519] 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.

[0520] 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.

[0521] 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).

[0522] 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.

[0523] 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.

[0524] 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.

[0525] 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.

[0526] 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.

[0527] 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."

[0528] This invention is a system that uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[0529] Processing performed by the server

[0530] 1. Loading event data

[0531] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[0532] 2. ChatGPT model initialization and context preparation

[0533] The server initializes the generative AI model (ChatGPT) and passes the loaded event details and persona setting information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[0534] 3. Generating and delivering the first chat

[0535] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[0536] 4. Real-time chat generation

[0537] During the event, the server collects audience reactions and feeds them back to the ChatGPT model in real time, generating new questions and comments that can continuously increase audience engagement.

[0538] 5. Data storage and feedback collection

[0539] After the event is over, the server stores all viewer response data and sends a feedback form to the viewer to collect suggestions for improving the next event.

[0540] Processing performed by the device

[0541] 1. Displaying chat contents

[0542] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[0543] 2. Receiving and sending user input

[0544] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[0545] 3. Providing a feedback form

[0546] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0547] User actions

[0548] 1. View and respond to chat

[0549] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[0550] 2. Submit your own questions or comments

[0551] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[0552] 3. Providing Feedback

[0553] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[0554] Specific examples

[0555] Example 1: Starting an event

[0556] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0557] Terminal: The terminal displays the questions sent from the server in real time.

[0558] User: The user is interested in the question displayed and enters an additional question.

[0559] Example 2: Interaction during an event

[0560] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[0561] Terminal: The terminal will immediately display the new question as it is generated.

[0562] User: The user responds again to the new question and sends a chat.

[0563] Example 3: After the event

[0564] Server: After the event, the server stores all the response data and delivers the feedback form.

[0565] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0566] User: The user fills out a feedback form and submits it.

[0567] This allows the system of the present invention to facilitate two-way communication at online events and increase audience engagement.

[0568] The processing flow will be explained below.

[0569] Step 1:

[0570] The server loads the event details and persona configuration information from the database a few minutes before the start of the online event, capturing basic information about the event theme, speakers, and topics covered.

[0571] Step 2:

[0572] The server initializes the ChatGPT model with the loaded event details and persona configuration information to prepare the context, so that the AI ​​understands the context to generate questions and comments about the event.

[0573] Step 3:

[0574] At the start of an event, the server sends a request to the ChatGPT model to generate initial questions and comments, including those from different perspectives such as personas A, B, and C.

[0575] Step 4:

[0576] The server distributes the generated questions and comments to a real-time chat system, which displays the initial chat content on the viewers' devices.

[0577] Step 5:

[0578] The terminal displays the chat contents delivered from the server to the user, allowing the user to view questions and comments generated in real time.

[0579] Step 6:

[0580] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[0581] Step 7:

[0582] The device sends the user's input responses and questions to the server, and the data sent becomes important information that enhances the interactivity of the event.

[0583] Step 8:

[0584] The server receives the reaction data sent by users and inputs it into the ChatGPT model as feedback, which allows the AI ​​to improve the accuracy of the questions and comments it generates next.

[0585] Step 9:

[0586] The server then sends a request to the ChatGPT model to generate new questions and comments, and distributes the generated content to the real-time chat system. This process is repeated throughout the event.

[0587] Step 10:

[0588] After the event, the server stores all reaction data in a database, which helps identify improvements for future events and audience interests.

[0589] Step 11:

[0590] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[0591] Step 12:

[0592] The terminal causes the feedback form to be displayed and easily accessible for the user to complete.

[0593] Step 13:

[0594] Users can enter their opinions and thoughts in the feedback form and press the submit button. The feedback will help improve the quality of the next event.

[0595] Step 14:

[0596] The device forwards the feedback sent by the user to the server, which stores it in a database and uses it for analysis.

[0597] Example 1

[0598] 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."

[0599] With conventional online events, it was difficult to engage with viewers in two-way communication that matched their interests, making it difficult to encourage active participation. Furthermore, there was a lack of a system for effectively collecting viewers' reactions and feedback and reflecting that information in future events.

[0600] 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.

[0601] In this invention, the server includes means for generating and distributing chat content according to viewer interests during an online event, a display device for displaying the generated chat content, means for collecting viewer reactions and providing real-time feedback to the generative AI model, means for acquiring detailed event information and persona setting information in advance, and means for generating new questions and comments in real-time based on viewer reactions during the online event, thereby attracting viewer interest, effectively promoting two-way communication, and improving the quality of future events.

[0602] A "generative AI model" is an artificial intelligence system designed to perform tasks such as text generation.

[0603] An "online event" is an event such as a lecture, seminar, or conference that is held in real time via the Internet.

[0604] "Viewer" means a person who participates in an online event and watches or listens to the content in real time.

[0605] "Chat content" is a collection of statements, questions, and comments displayed in text format.

[0606] A "display device" is a device used by a user to view the generated chat content, and includes a personal computer, a smartphone, a tablet, and the like.

[0607] "Reactions" are reactions such as comments, questions, and opinions that viewers show during the event.

[0608] "Feedback" refers to information such as opinions, ratings, and impressions collected from viewers.

[0609] "Event details" refers to information about the online event, such as the theme, speakers, and topics.

[0610] "Persona setting information" is information about a virtual character based on the viewer's interests and attributes.

[0611] "New questions and comments" are new textual statements or questions that are generated based on viewer reactions.

[0612] This system uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[0613] Server Roles

[0614] The server loads online event details and persona configuration information from a database such as MySQL. This information includes the event theme, speakers, planned topics, and audience persona information. The server then initializes OpenAI's ChatGPT model and passes it the event details and persona information. Once this is ready, the server generates the initial chat content and distributes it to the chat system in real time.

[0615] For example, the server loads information about a webinar on AI technology and sends a prompt to ChatGPT to ask viewers what topic in AI technology they are most interested in. The generated questions are then sent to the display device in real time.

[0616] Device Role

[0617] The device displays chat content sent from the server in real time, allowing viewers to easily view the questions and comments that have been generated. When viewers type responses to chats or questions, the device sends the information to the server. After the event ends, a feedback form is displayed, allowing viewers to easily enter their opinions and thoughts.

[0618] As a specific example, a viewer types a response into their smartphone, such as "I'm interested in image recognition using AI," and this is sent to the server.

[0619] User Roles

[0620] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, stimulating two-way communication. Users can also enter their own concerns or questions and submit them to the chat system. After the event, users can enter their opinions and impressions in a feedback form and submit it to help improve the quality of the next event.

[0621] As a concrete example, a user types the question "How does AI learn?" and sends it to the chat.

[0622] Prompt Sentence Examples

[0623] Below is an example of a prompt sentence.

[0624] "What topic in AI technology interests you most?"

[0625] "Please tell us about the latest technological trends using AI."

[0626] What was the most memorable aspect of what was said at the event?

[0627] In this system, the server collects viewer reactions and feeds them back into the ChatGPT model to generate new questions and comments in real time. After the event ends, the server saves all viewer reaction data and sends a feedback form. Through this process, the system of the present invention can increase viewer interest and effectively promote two-way communication.

[0628] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0629] Step 1: Loading event data

[0630] Specifically: Before the start of an online event, the server loads the event details (theme, speakers, topics, etc.) and persona configuration information from a database, which is necessary to generate questions and comments related to the event content.

[0631] Input: Database connection information (e.g. MySQL query), event identification information.

[0632] Data processing: Executes SQL queries to retrieve data about events and loads it into memory.

[0633] Output: A data object containing event details and persona configuration information.

[0634] Specific operation: The server executes an SQL query such as "SELECT FROM events WHERE id=1" to retrieve information about the "2023 AI Technology Webinar."

[0635] Step 2: Initializing the ChatGPT model and preparing the context

[0636] Specifically: The server initializes the ChatGPT model with the retrieved event details and persona configuration information, allowing the model to have the right context to generate questions and comments based on the event content.

[0637] Input: Event details and persona configuration information.

[0638] Data processing: Obtaining the API key and configuration parameters required for initialization, and passing context data to the ChatGPT model.

[0639] Output: The initialized ChatGPT model.

[0640] What happens: The server uses the openai.ChatCompletion.create method in Python to pass event information (e.g., the speaker is John Doe, and the topic is AI and medical technology) to the model.

[0641] Step 3: Generate and deliver the first chat

[0642] Specifically: When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate immediately.

[0643] Input: An initialized ChatGPT model, a trigger to start the event.

[0644] Data processing: Enter the prompt text into ChatGPT and format the generated data.

[0645] Output: Data from the initial question and comment.

[0646] How it works: The server generates a question from the ChatGPT model: "What topic are you most interested in in AI technology?" and sends it to the chat system via WebSocket or HTTP API.

[0647] Step 4: Real-time chat generation

[0648] During the event, the server collects audience reactions in real time and feeds them back to the ChatGPT model, allowing it to continuously generate new questions and comments.

[0649] Input: Viewer response data.

[0650] Data processing: Collect and analyze viewer response data and provide it to ChatGPT as feedback.

[0651] Output: Question and comment data updated in real time.

[0652] How it works: A viewer responds by saying, "I'm interested in AI image recognition," and the server then tells ChatGPT, "I'd like to know more about applications in the medical field." The model then generates a new question and distributes it to the chat system.

[0653] Step 5: Store data and collect feedback

[0654] After the event, the server will save all viewer response data and send a feedback form to the viewer, allowing us to collect their opinions for improving the next event.

[0655] Input: Event end trigger, viewer response data.

[0656] Data processing: Response data storage, feedback form generation and distribution.

[0657] Output: Saved data object, feedback form delivered to viewers.

[0658] What it does: The server saves all chat data from viewers in a MySQL database and sends feedback forms such as "How satisfied were you with the event?" via Google Forms or a dedicated application.

[0659] This is the specific processing flow of the program for this system. At each step, the necessary data processing and calculations are carried out based on clearly defined input data, and appropriate output is generated, promoting two-way communication with viewers.

[0660] (Application example 1)

[0661] 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."

[0662] In online events, there is a need for a method to maintain viewer interest, promote interactive communication, and collect event data to improve future events. In particular, there are insufficient methods for generating appropriate topics and questions in real time to encourage active viewer participation, and it is difficult to effectively collect feedback information after the event has ended.

[0663] 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.

[0664] In this invention, the server includes means for using a generative AI model to generate and distribute chat content based on viewer interests during an online event, means for collecting feedback after the event ends and sending it to the server, means for generating product-related questions and comments, a terminal for displaying the generated chat content, and means for collecting reactions from viewers and feeding them back to the AI ​​model in real time. This allows viewers to receive interesting chat content in real time, stimulating two-way communication, and effectively collecting and saving feedback from viewers, making it possible to identify areas for improvement for the next event.

[0665] A "generative AI model" is an artificial intelligence model that generates new text and content based on large amounts of data.

[0666] An "online event" is an event that takes place in real time over the Internet.

[0667] "Viewers" are users who participate in online events, view content, and ask questions.

[0668] "Chat content" refers to text generated by a generative AI model and used for two-way communication with viewers.

[0669] A "terminal" is a device that displays the generated chat content and receives input from viewers.

[0670] "Feedback Form" means an electronic form used to collect opinions and feedback from viewers after an event has concluded.

[0671] "Persona" refers to profile information of a typical audience for a particular event or service.

[0672] The "server" is a central computer system that loads event data, generates chat content, stores viewer reaction data, and so on.

[0673] "Products" means any products or services featured during the Online Event.

[0674] "Real time" means that various processes are carried out at the same time as the event is progressing.

[0675] "Response data" refers to information about interactions, such as comments and questions, that viewers make during an online event.

[0676] This invention is a system that uses generative AI models to generate chat content based on viewer interests at online events, promoting real-time two-way communication, and collecting and saving viewer feedback to help improve future events. The system consists of three main elements: a server, terminals, and users.

[0677] Processing performed by the server

[0678] 1. Loading event data

[0679] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[0680] 2. ChatGPT model initialization and context preparation

[0681] The server initializes a generative AI model (e.g., ChatGPT) and passes the loaded event details and persona configuration information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[0682] 3. Generating and delivering the first chat

[0683] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[0684] 4. Real-time chat generation

[0685] During the event, the server collects viewers' reactions and feeds them back to the ChatGPT model in real time to generate new questions and comments, which can continuously stimulate viewers' interest. It also generates product-related questions and comments, promoting two-way communication.

[0686] 5. Data storage and feedback collection

[0687] After the event is over, the server stores all viewer response data. After the event is over, a feedback form is sent to the viewer to collect suggestions for improvements to the next event. The collected feedback is stored on the server and used for the next event.

[0688] Processing performed by the device

[0689] 1. Displaying chat contents

[0690] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[0691] 2. Receiving and sending user input

[0692] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[0693] 3. Providing a feedback form

[0694] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0695] User actions

[0696] 1. View and respond to chat

[0697] Users can view the chat content generated by the generative AI model on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[0698] 2. Submit your own questions or comments

[0699] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[0700] 3. Providing Feedback

[0701] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[0702] Specific examples

[0703] Example 1: Starting an event

[0704] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0705] Terminal: The terminal displays the questions sent from the server in real time.

[0706] User: The user is interested in the question displayed and enters an additional question.

[0707] Example 2: Interaction during an event

[0708] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[0709] Terminal: The terminal will immediately display the new question as it is generated.

[0710] User: The user responds again to the new question and sends a chat.

[0711] Example 3: After the event

[0712] Server: After the event, the server stores all the response data and delivers the feedback form.

[0713] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0714] User: The user fills out a feedback form and submits it.

[0715] Specific examples of use

[0716] Example prompt sentence:

[0717] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[0718] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0719] Step 1:

[0720] Before the event begins, the server loads the event details and persona setting information from the database. Specifically, it obtains the event theme, speaker information, target topics, etc. This information serves as the basic data for the generative AI model to generate appropriate chat content. The input is the event ID, and the output is the event details and persona setting information.

[0721] Step 2:

[0722] The server initializes the ChatGPT model based on the acquired event details and persona configuration information. This information is input into the model as context data, and the generated chat content is prepared to be event-related. The input is the event details and persona configuration information, and the output is the initialized ChatGPT model.

[0723] Step 3:

[0724] When an event starts, the server sends a request to the ChatGPT model to generate the initial question and comment. The generative AI model generates appropriate questions and comments based on the provided context and returns them to the server. The server then distributes the generated content to the chat system so that viewers can view it. The input is the initialized ChatGPT model, and the output is the initial question and comment.

[0725] Step 4:

[0726] The terminal displays the chat content sent from the server in real time. Viewers can check the generated questions and comments on the terminal and make appropriate responses. The input is the chat content sent from the server, and the output is the displayed chat content.

[0727] Step 5:

[0728] Users input questions and comments in response to the chat content displayed on their devices. The input data is sent from the device to the server. The input is the user's question or comment, and the output is the data sent from the device to the server.

[0729] Step 6:

[0730] The server collects user responses in real time and feeds them back to the generative AI model. The model generates new questions and comments, promoting two-way communication. The input is user response data, and the output is the generated new questions and comments.

[0731] Step 7:

[0732] When the event ends, the server saves all the reaction data. This records what interactions took place during the event and serves as a reference for the next event. The input is all the reaction data after the event ends, and the output is the saved reaction data.

[0733] Step 8:

[0734] After the event, the server distributes a feedback form to viewers. Viewers fill out the form with their opinions and thoughts and send it to the server. The input is the feedback form distributed to the viewer, and the output is the filled-in feedback.

[0735] Step 9:

[0736] The device passes the feedback sent by the viewer to the server, which stores it. This feedback is used to identify improvements for the next event. The input is the viewer feedback, and the output is the stored feedback data.

[0737] For example:

[0738] Regarding prompt generation, by inputting the following prompt into the model, appropriate chat content is generated:

[0739] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[0740] 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.

[0741] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0742] Processing performed by the server

[0743] 1. Loading event data

[0744] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[0745] 2. ChatGPT model initialization and context preparation

[0746] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[0747] 3. Generating and delivering the first chat

[0748] When an event starts, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different perspectives (e.g., persona A, B, C), and are delivered to the chat system in real time.

[0749] 4. Emotion engine integration

[0750] The server collects viewer reaction data in real time and passes it to the emotion engine to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions.

[0751] 5. Real-time chat generation

[0752] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the emotion engine, and distributes the generated content in real time.

[0753] 6. Data storage and feedback collection

[0754] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[0755] Processing performed by the device

[0756] 1. Displaying chat contents

[0757] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[0758] 2. Receiving and sending user input

[0759] When a user enters a chat or a response to a question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[0760] 3. Providing a feedback form

[0761] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0762] User actions

[0763] 1. View and respond to chat

[0764] Users can view the chat content generated by ChatGPT on their device and respond to questions or comments that interest them.

[0765] 2. Submit your own questions or comments

[0766] Users can input their own concerns or questions and send them to the chat system.

[0767] 3. Providing Feedback

[0768] After the event, users can enter their opinions and thoughts in a feedback form and submit it, which will help improve the quality of future events.

[0769] Specific examples

[0770] Example 1: Starting an event

[0771] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0772] Terminal: The terminal displays the questions sent from the server in real time.

[0773] User: A user is interested in a question displayed and enters a new question or comment.

[0774] Example 2: Interaction during an event

[0775] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[0776] Terminal: The terminal will immediately display the new question as it is generated.

[0777] User: A user responds to new questions and sends chats.

[0778] Example 3: After the event

[0779] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[0780] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0781] User: The user fills out a feedback form and submits it.

[0782] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[0783] The processing flow will be explained below.

[0784] Step 1:

[0785] Before the online event starts, the server loads the event details and persona configuration information from the database, including the event theme, speakers, and topics covered.

[0786] Step 2:

[0787] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the questions and comments generated are appropriate for the event.

[0788] Step 3:

[0789] At the start of an event, the server sends a request to the ChatGPT model to generate the initial questions and comments, which include content based on different personas (e.g., personas A, B, C).

[0790] Step 4:

[0791] The server then broadcasts the generated questions and comments to a real-time chat system, allowing viewers to see the original chat transcript.

[0792] Step 5:

[0793] The terminal displays the chat contents sent from the server to the user, allowing the user to view questions and comments generated in real time.

[0794] Step 6:

[0795] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[0796] Step 7:

[0797] The device sends the user's input responses and questions to the server, and simultaneously sends the user's facial expressions and voice data to the emotion engine.

[0798] Step 8:

[0799] The emotion engine analyzes the user's facial and voice data to identify the user's emotional state, and sends the recognized emotion data to the server.

[0800] Step 9:

[0801] The server receives reaction data sent by users and emotion data from the emotion engine, and feeds this back to the ChatGPT model to improve the accuracy of the next questions and comments it generates.

[0802] Step 10:

[0803] The server again sends requests to the ChatGPT model to generate new questions and comments, and delivers them to the chat system in real time. This process is repeated throughout the event.

[0804] Step 11:

[0805] After the event, the server stores all reaction and sentiment data in a database, which allows for analysis and improvement for future events.

[0806] Step 12:

[0807] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[0808] Step 13:

[0809] The terminal displays a feedback form to the user for easy access and completion.

[0810] Step 14:

[0811] Users can enter their opinions and thoughts in the feedback form and press the submit button to send the data to the server, which will help improve the quality of the next event.

[0812] Step 15:

[0813] The server stores the submitted feedback data in a database and uses it for analysis to identify areas for improvement for the next event.

[0814] Example 2

[0815] 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."

[0816] To promote viewer engagement and interactive communication at online events, a system is needed that can generate chat content in real time based on viewer interests and emotions and instantly reflect their feedback. However, conventional systems do not adequately generate content based on viewer emotions or generate chat content from diverse perspectives using personas, resulting in a decline in viewer participation.

[0817] 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 a means for generating and distributing chat content according to viewer interests during an online event using a generative AI model, a terminal for displaying the generated chat content, a means for collecting reactions from viewers and feeding them back to the AI ​​model in real time, an engine for analyzing viewer emotions, and a means for improving the chat content based on the viewer emotions. This enables real-time chat generation and feedback according to viewer emotions and reactions.

[0818] A "generative AI model" is a type of artificial intelligence that automatically generates chat content based on viewers' interests and reactions.

[0819] An "online event" is an event or webinar that takes place live over the internet and allows participants to participate in it in real time.

[0820] "Viewers" are users who participate in an online event, who watch the event content, and participate through chat.

[0821] "Chat content" refers to messages such as questions and comments generated by the generative AI model, and is information used to communicate with viewers during online events.

[0822] "Terminal" refers to a device used by a viewer, such as a computer, smartphone, or tablet, that can display and input chat content.

[0823] "Reactions" refer to responses such as comments, questions, and facial expression analysis data made by viewers in response to chat content.

[0824] "Feedback" refers to the process of collecting and analyzing viewer reaction data, feeding the results back into the generative AI model, and reflecting them in the generation of the next chat content.

[0825] An "emotion engine" is a system that analyzes the viewer's facial expressions and voice data to determine their emotional state.

[0826] A "persona" is a virtual character or user profile created based on the viewer's interests and reactions, and is a factor taken into consideration when generating chat content.

[0827] "Feedback Form" means an input form provided to collect opinions and feedback from viewers after the online event has ended.

[0828] "Analysis" refers to the process of analyzing viewer reaction data and emotional data using statistical or machine learning techniques.

[0829] "Distribution" refers to the process of sending generated chat content, feedback forms, etc. to viewers in real time.

[0830] "Storage" refers to the process of storing data collected during an event in a recording medium such as a database.

[0831] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0832] server

[0833] The server handles the primary data processing for the online event, and specifically uses the following hardware and software:

[0834] Hardware: High-performance server machine (e.g., a server equipped with an Intel Xeon processor)

[0835] Software: Cloud infrastructure using AWS or Google Cloud, database using MySQL or PostgreSQL, ChatGPT as the generative AI model, and Affectiva as the sentiment analysis engine

[0836] 1. Loading event data

[0837] Before the event starts, the server loads the event details (e.g., event name, start time, speaker information) and persona setting information from the database, thereby obtaining information about the event theme and speakers.

[0838] 2. ChatGPT model initialization and context preparation

[0839] Based on the loaded information, we initialize the ChatGPT model and set the context, which allows the model to generate questions and comments related to the event.

[0840] 3. Generating and delivering the first chat

[0841] Once the event begins, the ChatGPT model generates the initial questions and comments, which are then streamed in real time.

[0842] 4. Emotion engine integration

[0843] The server collects viewer reaction data in real time and passes it to the emotion engine for analysis. The analysis results are fed back to the ChatGPT model to refine the chat content based on viewer emotions.

[0844] 5. Real-time chat generation

[0845] Based on viewer reaction data and the analysis results of the emotion engine, the server uses the ChatGPT model to generate new questions and comments and distribute them in real time.

[0846] 6. Data storage and feedback collection

[0847] After the event, all reaction and sentiment data is stored in a database, and a feedback form is sent to viewers to collect suggestions for improvements to be made to the next event.

[0848] Terminal

[0849] A terminal is a device used by a viewer and has the following functions:

[0850] Hardware: PCs, smartphones, tablets

[0851] Software: Web browser or dedicated application

[0852] 1. Displaying chat contents

[0853] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[0854] 2. Receiving and sending user input

[0855] When a user enters a response to a chat or question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[0856] 3. Providing a feedback form

[0857] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[0858] User

[0859] Users participate in online events and interact with the system using their devices, performing the following actions in the process:

[0860] 1. View and respond to chat

[0861] Users can view the chat content generated by ChatGPT on their device and respond to any content that interests them.

[0862] 2. Submit your own questions or comments

[0863] Users can input their own concerns or questions and send them to the chat system.

[0864] 3. Providing Feedback

[0865] After the event, users can enter their opinions and thoughts in a feedback form to help improve the quality of the next event.

[0866] Specific examples

[0867] Below is a concrete example of how this system can be actually operated.

[0868] Example 1: Starting an event

[0869] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[0870] Terminal: The terminal displays the questions sent from the server in real time.

[0871] User: A user is interested in a question displayed and enters a new question or comment.

[0872] Example 2: Interaction during an event

[0873] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[0874] Terminal: The terminal will immediately display the new question as it is generated.

[0875] User: A user responds to new questions and sends chats.

[0876] Example 3: After the event

[0877] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[0878] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[0879] User: The user fills out a feedback form and submits it.

[0880] Prompt Sentence Examples

[0881] "Thank you for joining us for our virtual event today. Please answer the following question: What do you think about the theme of the event?"

[0882] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[0883] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0884] Step 1:

[0885] Loading event data

[0886] Before an online event begins, the server loads event details and persona configuration information from a database.

[0887] Input: Event ID

[0888] Operation:

[0889] The server executes the SQL query "SELECT FROM EventDetails WHERE EventID = X".

[0890] The server obtains detailed information about the event (e.g., event name, start time, speaker information) and persona setting information (e.g., characteristics of Persona A and Persona B).

[0891] Output: Event details, persona setting information

[0892] Step 2:

[0893] ChatGPT model initialization and context preparation

[0894] The server initializes the ChatGPT model and sets the context based on the loaded event details and persona configuration information.

[0895] Input: Event details, persona setting information

[0896] Operation:

[0897] The server calls the "init_model(event_details, persona_info)" method to initialize the ChatGPT model.

[0898] After initialization, the model is provided with an initial prompt appropriate to the event content to set the context.

[0899] Output: Initialized ChatGPT model

[0900] Step 3:

[0901] First chat generation and distribution

[0902] When an event starts, the server will generate the initial questions and comments from the ChatGPT model, and the chat will be streamed in real time.

[0903] Input: Initialized ChatGPT model

[0904] Operation:

[0905] The server calls the "generate_initial_chat(model, context)" method to generate the initial chat content.

[0906] The generated chat content is sent to the chat system using the "send_to_chat_system(chat_content)" method.

[0907] Output: First chat

[0908] Step 4:

[0909] Collecting reactions from viewers

[0910] The device collects chat content and reactions from viewers in real time and sends them to the server.

[0911] Input: Viewer input data (e.g., chat messages, facial expression analysis data)

[0912] Operation:

[0913] The terminal executes the "receive_user_input()" method to collect input from the user.

[0914] The collected data is sent to the server using the "send_to_server(user_input)" method.

[0915] Output: Viewer input data

[0916] Step 5:

[0917] Emotion engine collaboration

[0918] The server passes viewer reaction data to the emotion engine in real time to analyze user emotions.

[0919] Input: Viewer input data

[0920] Operation:

[0921] The server sends reaction data to the emotion engine using the "send_to_emotion_engine(reaction_data)" method.

[0922] The emotion engine analyzes the data and returns emotion results to the server.

[0923] Output: Emotion analysis results

[0924] Step 6:

[0925] Real-time chat generation

[0926] The server uses the ChatGPT model to generate new questions and comments based on viewer reaction data and the analysis results of the emotion engine, and distributes them in real time.

[0927] Input: Sentiment analysis results, viewer input data

[0928] Operation:

[0929] The server executes the "generate_new_chat(model, feedback)" method to generate new chat content.

[0930] The generated new chat content is distributed using "send_to_chat_system(chat_content)".

[0931] Output: New chat transcript

[0932] Step 7:

[0933] Data storage and feedback collection

[0934] After the event, the server stores all reaction and emotion data in a database and delivers a feedback form to viewers.

[0935] Input: All data after the event

[0936] Operation:

[0937] The server saves all data to the database using "save_data_to_database(data)".

[0938] Send the feedback form to viewers using "send_feedback_form()".

[0939] Output: Saved data, delivered feedback form

[0940] (Application example 2)

[0941] 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."

[0942] Conventional online events lacked mechanisms to effectively promote interactive communication with viewers. As a result, it was difficult to respond immediately to viewers' interests and reactions, resulting in a decline in viewer engagement. Furthermore, there was no technology that could analyze viewers' emotions in real time and dynamically improve chat content using generative AI models based on that analysis. This made it difficult to provide optimal content that met viewers' expectations.

[0943] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for using a generative AI model to generate and distribute chat content tailored to viewer interests during an online event; a terminal for displaying the generated chat content; means for collecting viewer reactions and providing feedback to the AI ​​model in real time; means for analyzing viewer emotions using an emotion analysis engine and improving the generated chat content based on the analysis results; means for constantly generating new questions and comments based on viewer reactions and distributing them in real time; and means for distributing a feedback form to viewers after the event ends and identifying areas for improvement for the next event. This makes it possible to significantly increase viewer engagement by immediately responding to viewer emotions and reactions and providing optimal chat content.

[0944] A "generative AI model" is an artificial intelligence model that takes data as input and automatically generates content based on specific contexts and conditions.

[0945] "Chat content" refers to text messages such as questions and comments that are generated by a generative AI model during an online event and distributed to viewers.

[0946] A "terminal" is a device that displays the generated chat content, accepts input from viewers, and transmits it to the server.

[0947] "Viewer reactions" refers to feedback such as facial expressions, gestures, voice, and text comments given by viewers during an online event.

[0948] "Real-time feedback" refers to a mechanism that instantly senses viewer reactions and sends them to a generative AI model for use in improving chat content.

[0949] An "emotion analysis engine" is software or hardware that analyzes emotions from viewers' facial expressions and voices and provides the results to a generative AI model.

[0950] A "persona" is a virtual character based on a specific audience or user profile, which the generative AI model uses as a reference when generating chat content.

[0951] A "feedback form" is a form provided to viewers after an online event to allow them to enter their evaluation of the event and suggestions for improvement.

[0952] This invention is a system that uses a generative AI model and an emotion analysis engine to generate chat content in real time during online events based on viewer interests, promoting two-way communication with viewers. This system consists of four main components: a server, terminals, viewers, and an emotion analysis engine.

[0953] Processing performed by the server

[0954] 1. Loading event data

[0955] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[0956] 2. ChatGPT model initialization and context preparation

[0957] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[0958] 3. Generating and delivering the first chat

[0959] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different personas (e.g., persona A, persona B, etc.) and are distributed to the chat system in real time.

[0960] 4. Emotion engine integration

[0961] The server collects viewer reaction data in real time and passes it to a sentiment analysis engine to analyze user emotions. The analysis results are fed back to the ChatGPT model, which refines the generated chat content according to viewer emotions.

[0962] 5. Real-time chat generation

[0963] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the sentiment analysis engine, and distributes the generated content in real time.

[0964] 6. Data storage and feedback collection

[0965] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[0966] Processing performed by the device

[0967] 1. Displaying chat contents

[0968] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[0969] 2. Receiving and sending user input

[0970] When viewers chat or respond to questions, the device sends that information to the server, along with facial expression and voice analysis data from the viewers to the emotion analysis engine.

[0971] 3. Providing a feedback form

[0972] After the event, the device will display a feedback form, allowing viewers to easily enter their opinions and thoughts.

[0973] User actions

[0974] 1. View and respond to chat

[0975] Viewers can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them.

[0976] 2. Submit your own questions or comments

[0977] Viewers can enter their own concerns or questions and send them to the chat system.

[0978] 3. Providing Feedback

[0979] After the event, viewers can fill out a feedback form and submit their thoughts and opinions, which will help improve the quality of future events.

[0980] Hardware / Software used

[0981] Server: AWS EC2 instance, or other cloud-based virtual machine

[0982] Sentiment Analysis Engine: Affectiva SDK, or other sentiment analysis platform

[0983] Chat display devices: smartphones, tablets, PCs

[0984] Real-time databases: Firebase Realtime Database, MongoDB

[0985] Generative AI model: OpenAI GPT-4 API or other advanced generative AI models

[0986] Prompt Sentence Examples

[0987] "User emotion is {emotion}. Event data is:\n{event_data}\nPlease generate an appropriate question."

[0988] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0989] Step 1:

[0990] Before the online event starts, the server loads the event details and persona information from the database, which provides basic information about the event theme, speakers, and topics covered. The input is the event ID, and the output is the event details and persona information.

[0991] Step 2:

[0992] The server initializes the ChatGPT model based on the loaded event details and persona settings, and prepares the context. This provides the context for the generative AI model to generate questions and comments related to the event. The input is the event details and persona settings, and the output is the initialized ChatGPT model.

[0993] Step 3:

[0994] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments. The generated questions and comments reflect different personas (e.g., Persona A, Persona B) and are delivered to the chat system in real time. The input is the initialized ChatGPT model, and the output is the generated initial questions and comments.

[0995] Step 4:

[0996] The terminal displays the chat content sent from the server in real time. Viewers view and respond to generated questions and comments. The input is the chat content sent from the server, and the output is the chat content that viewers see on their terminal.

[0997] Step 5:

[0998] When viewers enter responses to chats or questions, the device sends that information to the server. It also sends the viewer's facial expressions and voice analysis data to the emotion analysis engine. The input is the viewer's responses and emotion data, and the output is the data sent to the server and the emotion analysis engine.

[0999] Step 6:

[1000] The server sends viewer reaction data to the sentiment analysis engine in real time to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions. The input is viewer reaction data and sentiment analysis results, and the output is refined chat content.

[1001] Step 7:

[1002] The server sends a request to the ChatGPT model to generate new questions and comments based on the viewer reaction data and the analysis results from the sentiment analysis engine, and distributes the generated content in real time. The input is the analysis results and the ChatGPT model, and the output is new questions and comments generated in real time.

[1003] Step 8:

[1004] After the event, the server stores all the reaction and emotion data in a database and sends a feedback form to the viewers to collect improvements for the next event. The input is the viewer's reaction and emotion data, and the output is the stored data and the feedback form.

[1005] 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.

[1006] 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.

[1007] 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.

[1008] [Third embodiment]

[1009] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1010] 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.

[1011] 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).

[1012] 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.

[1013] 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.

[1014] 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).

[1015] 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.

[1016] 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.

[1017] 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.

[1018] 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.

[1019] 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.

[1020] 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."

[1021] This invention is a system that uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[1022] Processing performed by the server

[1023] 1. Loading event data

[1024] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[1025] 2. ChatGPT model initialization and context preparation

[1026] The server initializes the generative AI model (ChatGPT) and passes the loaded event details and persona setting information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[1027] 3. Generating and delivering the first chat

[1028] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[1029] 4. Real-time chat generation

[1030] During the event, the server collects audience reactions and feeds them back to the ChatGPT model in real time, generating new questions and comments that can continuously increase audience engagement.

[1031] 5. Data storage and feedback collection

[1032] After the event is over, the server stores all viewer response data and sends a feedback form to the viewer to collect suggestions for improving the next event.

[1033] Processing performed by the device

[1034] 1. Displaying chat contents

[1035] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[1036] 2. Receiving and sending user input

[1037] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[1038] 3. Providing a feedback form

[1039] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1040] User actions

[1041] 1. View and respond to chat

[1042] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[1043] 2. Submit your own questions or comments

[1044] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[1045] 3. Providing Feedback

[1046] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[1047] Specific examples

[1048] Example 1: Starting an event

[1049] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1050] Terminal: The terminal displays the questions sent from the server in real time.

[1051] User: The user is interested in the question displayed and enters an additional question.

[1052] Example 2: Interaction during an event

[1053] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[1054] Terminal: The terminal will immediately display the new question as it is generated.

[1055] User: The user responds again to the new question and sends a chat.

[1056] Example 3: After the event

[1057] Server: After the event, the server stores all the response data and delivers the feedback form.

[1058] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1059] User: The user fills out a feedback form and submits it.

[1060] This allows the system of the present invention to facilitate two-way communication at online events and increase audience engagement.

[1061] The processing flow will be explained below.

[1062] Step 1:

[1063] The server loads the event details and persona configuration information from the database a few minutes before the start of the online event, capturing basic information about the event theme, speakers, and topics covered.

[1064] Step 2:

[1065] The server initializes the ChatGPT model with the loaded event details and persona configuration information to prepare the context, so that the AI ​​understands the context to generate questions and comments about the event.

[1066] Step 3:

[1067] At the start of an event, the server sends a request to the ChatGPT model to generate initial questions and comments, including those from different perspectives such as personas A, B, and C.

[1068] Step 4:

[1069] The server distributes the generated questions and comments to a real-time chat system, which displays the initial chat content on the viewers' devices.

[1070] Step 5:

[1071] The terminal displays the chat contents delivered from the server to the user, allowing the user to view questions and comments generated in real time.

[1072] Step 6:

[1073] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[1074] Step 7:

[1075] The device sends the user's input responses and questions to the server, and the data sent becomes important information that enhances the interactivity of the event.

[1076] Step 8:

[1077] The server receives the reaction data sent by users and inputs it into the ChatGPT model as feedback, which allows the AI ​​to improve the accuracy of the questions and comments it generates next.

[1078] Step 9:

[1079] The server then sends a request to the ChatGPT model to generate new questions and comments, and distributes the generated content to the real-time chat system. This process is repeated throughout the event.

[1080] Step 10:

[1081] After the event, the server stores all reaction data in a database, which helps identify improvements for future events and audience interests.

[1082] Step 11:

[1083] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[1084] Step 12:

[1085] The terminal causes the feedback form to be displayed and easily accessible for the user to complete.

[1086] Step 13:

[1087] Users can enter their opinions and thoughts in the feedback form and press the submit button. The feedback will help improve the quality of the next event.

[1088] Step 14:

[1089] The device forwards the feedback sent by the user to the server, which stores it in a database and uses it for analysis.

[1090] Example 1

[1091] 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."

[1092] With conventional online events, it was difficult to engage with viewers in two-way communication that matched their interests, making it difficult to encourage active participation. Furthermore, there was a lack of a system for effectively collecting viewers' reactions and feedback and reflecting that information in future events.

[1093] 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.

[1094] In this invention, the server includes means for generating and distributing chat content according to viewer interests during an online event, a display device for displaying the generated chat content, means for collecting viewer reactions and providing real-time feedback to the generative AI model, means for acquiring detailed event information and persona setting information in advance, and means for generating new questions and comments in real-time based on viewer reactions during the online event, thereby attracting viewer interest, effectively promoting two-way communication, and improving the quality of future events.

[1095] A "generative AI model" is an artificial intelligence system designed to perform tasks such as text generation.

[1096] An "online event" is an event such as a lecture, seminar, or conference that is held in real time via the Internet.

[1097] "Viewer" means a person who participates in an online event and watches or listens to the content in real time.

[1098] "Chat content" is a collection of statements, questions, and comments displayed in text format.

[1099] A "display device" is a device used by a user to view the generated chat content, and includes a personal computer, a smartphone, a tablet, and the like.

[1100] "Reactions" are reactions such as comments, questions, and opinions that viewers show during the event.

[1101] "Feedback" refers to information such as opinions, ratings, and impressions collected from viewers.

[1102] "Event details" refers to information about the online event, such as the theme, speakers, and topics.

[1103] "Persona setting information" is information about a virtual character based on the viewer's interests and attributes.

[1104] "New questions and comments" are new textual statements or questions that are generated based on viewer reactions.

[1105] This system uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[1106] Server Roles

[1107] The server loads online event details and persona configuration information from a database such as MySQL. This information includes the event theme, speakers, planned topics, and audience persona information. The server then initializes OpenAI's ChatGPT model and passes it the event details and persona information. Once this is ready, the server generates the initial chat content and distributes it to the chat system in real time.

[1108] For example, the server loads information about a webinar on AI technology and sends a prompt to ChatGPT to ask viewers what topic in AI technology they are most interested in. The generated questions are then sent to the display device in real time.

[1109] Device Role

[1110] The device displays chat content sent from the server in real time, allowing viewers to easily view the questions and comments that have been generated. When viewers type responses to chats or questions, the device sends the information to the server. After the event ends, a feedback form is displayed, allowing viewers to easily enter their opinions and thoughts.

[1111] As a specific example, a viewer types a response into their smartphone, such as "I'm interested in image recognition using AI," and this is sent to the server.

[1112] User Roles

[1113] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, stimulating two-way communication. Users can also enter their own concerns or questions and submit them to the chat system. After the event, users can enter their opinions and impressions in a feedback form and submit it to help improve the quality of the next event.

[1114] As a concrete example, a user types the question "How does AI learn?" and sends it to the chat.

[1115] Prompt Sentence Examples

[1116] Below is an example of a prompt sentence.

[1117] "What topic in AI technology interests you most?"

[1118] "Please tell us about the latest technological trends using AI."

[1119] What was the most memorable aspect of what was said at the event?

[1120] In this system, the server collects viewer reactions and feeds them back into the ChatGPT model to generate new questions and comments in real time. After the event ends, the server saves all viewer reaction data and sends a feedback form. Through this process, the system of the present invention can increase viewer interest and effectively promote two-way communication.

[1121] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1122] Step 1: Loading event data

[1123] Specifically: Before the start of an online event, the server loads the event details (theme, speakers, topics, etc.) and persona configuration information from a database, which is necessary to generate questions and comments related to the event content.

[1124] Input: Database connection information (e.g. MySQL query), event identification information.

[1125] Data processing: Executes SQL queries to retrieve data about events and loads it into memory.

[1126] Output: A data object containing event details and persona configuration information.

[1127] Specific operation: The server executes an SQL query such as "SELECT FROM events WHERE id=1" to retrieve information about the "2023 AI Technology Webinar."

[1128] Step 2: Initializing the ChatGPT model and preparing the context

[1129] Specifically: The server initializes the ChatGPT model with the retrieved event details and persona configuration information, allowing the model to have the right context to generate questions and comments based on the event content.

[1130] Input: Event details and persona configuration information.

[1131] Data processing: Obtaining the API key and configuration parameters required for initialization, and passing context data to the ChatGPT model.

[1132] Output: The initialized ChatGPT model.

[1133] What happens: The server uses the openai.ChatCompletion.create method in Python to pass event information (e.g., the speaker is John Doe, and the topic is AI and medical technology) to the model.

[1134] Step 3: Generate and deliver the first chat

[1135] Specifically: When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate immediately.

[1136] Input: An initialized ChatGPT model, a trigger to start the event.

[1137] Data processing: Enter the prompt text into ChatGPT and format the generated data.

[1138] Output: Data from the initial question and comment.

[1139] How it works: The server generates a question from the ChatGPT model: "What topic are you most interested in in AI technology?" and sends it to the chat system via WebSocket or HTTP API.

[1140] Step 4: Real-time chat generation

[1141] During the event, the server collects audience reactions in real time and feeds them back to the ChatGPT model, allowing it to continuously generate new questions and comments.

[1142] Input: Viewer response data.

[1143] Data processing: Collect and analyze viewer response data and provide it to ChatGPT as feedback.

[1144] Output: Question and comment data updated in real time.

[1145] How it works: A viewer responds by saying, "I'm interested in AI image recognition," and the server then tells ChatGPT, "I'd like to know more about applications in the medical field." The model then generates a new question and distributes it to the chat system.

[1146] Step 5: Store data and collect feedback

[1147] After the event, the server will save all viewer response data and send a feedback form to the viewer, allowing us to collect their opinions for improving the next event.

[1148] Input: Event end trigger, viewer response data.

[1149] Data processing: Response data storage, feedback form generation and distribution.

[1150] Output: Saved data object, feedback form delivered to viewers.

[1151] What it does: The server saves all chat data from viewers in a MySQL database and sends feedback forms such as "How satisfied were you with the event?" via Google Forms or a dedicated application.

[1152] This is the specific processing flow of the program for this system. At each step, the necessary data processing and calculations are carried out based on clearly defined input data, and appropriate output is generated, promoting two-way communication with viewers.

[1153] (Application example 1)

[1154] 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."

[1155] In online events, there is a need for a method to maintain viewer interest, promote interactive communication, and collect event data to improve future events. In particular, there are insufficient methods for generating appropriate topics and questions in real time to encourage active viewer participation, and it is difficult to effectively collect feedback information after the event has ended.

[1156] 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.

[1157] In this invention, the server includes means for using a generative AI model to generate and distribute chat content based on viewer interests during an online event, means for collecting feedback after the event ends and sending it to the server, means for generating product-related questions and comments, a terminal for displaying the generated chat content, and means for collecting reactions from viewers and feeding them back to the AI ​​model in real time. This allows viewers to receive interesting chat content in real time, stimulating two-way communication, and effectively collecting and saving feedback from viewers, making it possible to identify areas for improvement for the next event.

[1158] A "generative AI model" is an artificial intelligence model that generates new text and content based on large amounts of data.

[1159] An "online event" is an event that takes place in real time over the Internet.

[1160] "Viewers" are users who participate in online events, view content, and ask questions.

[1161] "Chat content" refers to text generated by a generative AI model and used for two-way communication with viewers.

[1162] A "terminal" is a device that displays the generated chat content and receives input from viewers.

[1163] "Feedback Form" means an electronic form used to collect opinions and feedback from viewers after an event has concluded.

[1164] "Persona" refers to profile information of a typical audience for a particular event or service.

[1165] The "server" is a central computer system that loads event data, generates chat content, stores viewer reaction data, and so on.

[1166] "Products" means any products or services featured during the Online Event.

[1167] "Real time" means that various processes are carried out at the same time as the event is progressing.

[1168] "Response data" refers to information about interactions, such as comments and questions, that viewers make during an online event.

[1169] This invention is a system that uses generative AI models to generate chat content based on viewer interests at online events, promoting real-time two-way communication, and collecting and saving viewer feedback to help improve future events. The system consists of three main elements: a server, terminals, and users.

[1170] Processing performed by the server

[1171] 1. Loading event data

[1172] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[1173] 2. ChatGPT model initialization and context preparation

[1174] The server initializes a generative AI model (e.g., ChatGPT) and passes the loaded event details and persona configuration information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[1175] 3. Generating and delivering the first chat

[1176] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[1177] 4. Real-time chat generation

[1178] During the event, the server collects viewers' reactions and feeds them back to the ChatGPT model in real time to generate new questions and comments, which can continuously stimulate viewers' interest. It also generates product-related questions and comments, promoting two-way communication.

[1179] 5. Data storage and feedback collection

[1180] After the event is over, the server stores all viewer response data. After the event is over, a feedback form is sent to the viewer to collect suggestions for improvements to the next event. The collected feedback is stored on the server and used for the next event.

[1181] Processing performed by the device

[1182] 1. Displaying chat contents

[1183] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[1184] 2. Receiving and sending user input

[1185] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[1186] 3. Providing a feedback form

[1187] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1188] User actions

[1189] 1. View and respond to chat

[1190] Users can view the chat content generated by the generative AI model on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[1191] 2. Submit your own questions or comments

[1192] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[1193] 3. Providing Feedback

[1194] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[1195] Specific examples

[1196] Example 1: Starting an event

[1197] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1198] Terminal: The terminal displays the questions sent from the server in real time.

[1199] User: The user is interested in the question displayed and enters an additional question.

[1200] Example 2: Interaction during an event

[1201] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[1202] Terminal: The terminal will immediately display the new question as it is generated.

[1203] User: The user responds again to the new question and sends a chat.

[1204] Example 3: After the event

[1205] Server: After the event, the server stores all the response data and delivers the feedback form.

[1206] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1207] User: The user fills out a feedback form and submits it.

[1208] Specific examples of use

[1209] Example prompt sentence:

[1210] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[1211] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1212] Step 1:

[1213] Before the event begins, the server loads the event details and persona setting information from the database. Specifically, it obtains the event theme, speaker information, target topics, etc. This information serves as the basic data for the generative AI model to generate appropriate chat content. The input is the event ID, and the output is the event details and persona setting information.

[1214] Step 2:

[1215] The server initializes the ChatGPT model based on the acquired event details and persona configuration information. This information is input into the model as context data, and the generated chat content is prepared to be event-related. The input is the event details and persona configuration information, and the output is the initialized ChatGPT model.

[1216] Step 3:

[1217] When an event starts, the server sends a request to the ChatGPT model to generate the initial question and comment. The generative AI model generates appropriate questions and comments based on the provided context and returns them to the server. The server then distributes the generated content to the chat system so that viewers can view it. The input is the initialized ChatGPT model, and the output is the initial question and comment.

[1218] Step 4:

[1219] The terminal displays the chat content sent from the server in real time. Viewers can check the generated questions and comments on the terminal and make appropriate responses. The input is the chat content sent from the server, and the output is the displayed chat content.

[1220] Step 5:

[1221] Users input questions and comments in response to the chat content displayed on their devices. The input data is sent from the device to the server. The input is the user's question or comment, and the output is the data sent from the device to the server.

[1222] Step 6:

[1223] The server collects user responses in real time and feeds them back to the generative AI model. The model generates new questions and comments, promoting two-way communication. The input is user response data, and the output is the generated new questions and comments.

[1224] Step 7:

[1225] When the event ends, the server saves all the reaction data. This records what interactions took place during the event and serves as a reference for the next event. The input is all the reaction data after the event ends, and the output is the saved reaction data.

[1226] Step 8:

[1227] After the event, the server distributes a feedback form to viewers. Viewers fill out the form with their opinions and thoughts and send it to the server. The input is the feedback form distributed to the viewer, and the output is the filled-in feedback.

[1228] Step 9:

[1229] The device passes the feedback sent by the viewer to the server, which stores it. This feedback is used to identify improvements for the next event. The input is the viewer feedback, and the output is the stored feedback data.

[1230] For example:

[1231] Regarding prompt generation, by inputting the following prompt into the model, appropriate chat content is generated:

[1232] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[1233] 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.

[1234] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[1235] Processing performed by the server

[1236] 1. Loading event data

[1237] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[1238] 2. ChatGPT model initialization and context preparation

[1239] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[1240] 3. Generating and delivering the first chat

[1241] When an event starts, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different perspectives (e.g., persona A, B, C), and are delivered to the chat system in real time.

[1242] 4. Emotion engine integration

[1243] The server collects viewer reaction data in real time and passes it to the emotion engine to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions.

[1244] 5. Real-time chat generation

[1245] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the emotion engine, and distributes the generated content in real time.

[1246] 6. Data storage and feedback collection

[1247] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[1248] Processing performed by the device

[1249] 1. Displaying chat contents

[1250] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[1251] 2. Receiving and sending user input

[1252] When a user enters a chat or a response to a question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[1253] 3. Providing a feedback form

[1254] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1255] User actions

[1256] 1. View and respond to chat

[1257] Users can view the chat content generated by ChatGPT on their device and respond to questions or comments that interest them.

[1258] 2. Submit your own questions or comments

[1259] Users can input their own concerns or questions and send them to the chat system.

[1260] 3. Providing Feedback

[1261] After the event, users can enter their opinions and thoughts in a feedback form and submit it, which will help improve the quality of future events.

[1262] Specific examples

[1263] Example 1: Starting an event

[1264] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1265] Terminal: The terminal displays the questions sent from the server in real time.

[1266] User: A user is interested in a question displayed and enters a new question or comment.

[1267] Example 2: Interaction during an event

[1268] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[1269] Terminal: The terminal will immediately display the new question as it is generated.

[1270] User: A user responds to new questions and sends chats.

[1271] Example 3: After the event

[1272] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[1273] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1274] User: The user fills out a feedback form and submits it.

[1275] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[1276] The processing flow will be explained below.

[1277] Step 1:

[1278] Before the online event starts, the server loads the event details and persona configuration information from the database, including the event theme, speakers, and topics covered.

[1279] Step 2:

[1280] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the questions and comments generated are appropriate for the event.

[1281] Step 3:

[1282] At the start of an event, the server sends a request to the ChatGPT model to generate the initial questions and comments, which include content based on different personas (e.g., personas A, B, C).

[1283] Step 4:

[1284] The server then broadcasts the generated questions and comments to a real-time chat system, allowing viewers to see the original chat transcript.

[1285] Step 5:

[1286] The terminal displays the chat contents sent from the server to the user, allowing the user to view questions and comments generated in real time.

[1287] Step 6:

[1288] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[1289] Step 7:

[1290] The device sends the user's input responses and questions to the server, and simultaneously sends the user's facial expressions and voice data to the emotion engine.

[1291] Step 8:

[1292] The emotion engine analyzes the user's facial and voice data to identify the user's emotional state, and sends the recognized emotion data to the server.

[1293] Step 9:

[1294] The server receives reaction data sent by users and emotion data from the emotion engine, and feeds this back to the ChatGPT model to improve the accuracy of the next questions and comments it generates.

[1295] Step 10:

[1296] The server again sends requests to the ChatGPT model to generate new questions and comments, and delivers them to the chat system in real time. This process is repeated throughout the event.

[1297] Step 11:

[1298] After the event, the server stores all reaction and sentiment data in a database, which allows for analysis and improvement for future events.

[1299] Step 12:

[1300] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[1301] Step 13:

[1302] The terminal displays a feedback form to the user for easy access and completion.

[1303] Step 14:

[1304] Users can enter their opinions and thoughts in the feedback form and press the submit button to send the data to the server, which will help improve the quality of the next event.

[1305] Step 15:

[1306] The server stores the submitted feedback data in a database and uses it for analysis to identify areas for improvement for the next event.

[1307] Example 2

[1308] 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."

[1309] To promote viewer engagement and interactive communication at online events, a system is needed that can generate chat content in real time based on viewer interests and emotions and instantly reflect their feedback. However, conventional systems do not adequately generate content based on viewer emotions or generate chat content from diverse perspectives using personas, resulting in a decline in viewer participation.

[1310] 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 a means for generating and distributing chat content according to viewer interests during an online event using a generative AI model, a terminal for displaying the generated chat content, a means for collecting reactions from viewers and feeding them back to the AI ​​model in real time, an engine for analyzing viewer emotions, and a means for improving the chat content based on the viewer emotions. This enables real-time chat generation and feedback according to viewer emotions and reactions.

[1311] A "generative AI model" is a type of artificial intelligence that automatically generates chat content based on viewers' interests and reactions.

[1312] An "online event" is an event or webinar that takes place live over the internet and allows participants to participate in it in real time.

[1313] "Viewers" are users who participate in an online event, who watch the event content, and participate through chat.

[1314] "Chat content" refers to messages such as questions and comments generated by the generative AI model, and is information used to communicate with viewers during online events.

[1315] "Terminal" refers to a device used by a viewer, such as a computer, smartphone, or tablet, that can display and input chat content.

[1316] "Reactions" refer to responses such as comments, questions, and facial expression analysis data made by viewers in response to chat content.

[1317] "Feedback" refers to the process of collecting and analyzing viewer reaction data, feeding the results back into the generative AI model, and reflecting them in the generation of the next chat content.

[1318] An "emotion engine" is a system that analyzes the viewer's facial expressions and voice data to determine their emotional state.

[1319] A "persona" is a virtual character or user profile created based on the viewer's interests and reactions, and is a factor taken into consideration when generating chat content.

[1320] "Feedback Form" means an input form provided to collect opinions and feedback from viewers after the online event has ended.

[1321] "Analysis" refers to the process of analyzing viewer reaction data and emotional data using statistical or machine learning techniques.

[1322] "Distribution" refers to the process of sending generated chat content, feedback forms, etc. to viewers in real time.

[1323] "Storage" refers to the process of storing data collected during an event in a recording medium such as a database.

[1324] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[1325] server

[1326] The server handles the primary data processing for the online event, and specifically uses the following hardware and software:

[1327] Hardware: High-performance server machine (e.g., a server equipped with an Intel Xeon processor)

[1328] Software: Cloud infrastructure using AWS or Google Cloud, database using MySQL or PostgreSQL, ChatGPT as the generative AI model, and Affectiva as the sentiment analysis engine

[1329] 1. Loading event data

[1330] Before the event starts, the server loads the event details (e.g., event name, start time, speaker information) and persona setting information from the database, thereby obtaining information about the event theme and speakers.

[1331] 2. ChatGPT model initialization and context preparation

[1332] Based on the loaded information, we initialize the ChatGPT model and set the context, which allows the model to generate questions and comments related to the event.

[1333] 3. Generating and delivering the first chat

[1334] Once the event begins, the ChatGPT model generates the initial questions and comments, which are then streamed in real time.

[1335] 4. Emotion engine integration

[1336] The server collects viewer reaction data in real time and passes it to the emotion engine for analysis. The analysis results are fed back to the ChatGPT model to refine the chat content based on viewer emotions.

[1337] 5. Real-time chat generation

[1338] Based on viewer reaction data and the analysis results of the emotion engine, the server uses the ChatGPT model to generate new questions and comments and distribute them in real time.

[1339] 6. Data storage and feedback collection

[1340] After the event, all reaction and sentiment data is stored in a database, and a feedback form is sent to viewers to collect suggestions for improvements to be made to the next event.

[1341] Terminal

[1342] A terminal is a device used by a viewer and has the following functions:

[1343] Hardware: PCs, smartphones, tablets

[1344] Software: Web browser or dedicated application

[1345] 1. Displaying chat contents

[1346] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[1347] 2. Receiving and sending user input

[1348] When a user enters a response to a chat or question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[1349] 3. Providing a feedback form

[1350] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1351] User

[1352] Users participate in online events and interact with the system using their devices, performing the following actions in the process:

[1353] 1. View and respond to chat

[1354] Users can view the chat content generated by ChatGPT on their device and respond to any content that interests them.

[1355] 2. Submit your own questions or comments

[1356] Users can input their own concerns or questions and send them to the chat system.

[1357] 3. Providing Feedback

[1358] After the event, users can enter their opinions and thoughts in a feedback form to help improve the quality of the next event.

[1359] Specific examples

[1360] Below is a concrete example of how this system can be actually operated.

[1361] Example 1: Starting an event

[1362] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1363] Terminal: The terminal displays the questions sent from the server in real time.

[1364] User: A user is interested in a question displayed and enters a new question or comment.

[1365] Example 2: Interaction during an event

[1366] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[1367] Terminal: The terminal will immediately display the new question as it is generated.

[1368] User: A user responds to new questions and sends chats.

[1369] Example 3: After the event

[1370] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[1371] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1372] User: The user fills out a feedback form and submits it.

[1373] Prompt Sentence Examples

[1374] "Thank you for joining us for our virtual event today. Please answer the following question: What do you think about the theme of the event?"

[1375] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[1376] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1377] Step 1:

[1378] Loading event data

[1379] Before an online event begins, the server loads event details and persona configuration information from a database.

[1380] Input: Event ID

[1381] Operation:

[1382] The server executes the SQL query "SELECT FROM EventDetails WHERE EventID = X".

[1383] The server obtains detailed information about the event (e.g., event name, start time, speaker information) and persona setting information (e.g., characteristics of Persona A and Persona B).

[1384] Output: Event details, persona setting information

[1385] Step 2:

[1386] ChatGPT model initialization and context preparation

[1387] The server initializes the ChatGPT model and sets the context based on the loaded event details and persona configuration information.

[1388] Input: Event details, persona setting information

[1389] Operation:

[1390] The server calls the "init_model(event_details, persona_info)" method to initialize the ChatGPT model.

[1391] After initialization, the model is provided with an initial prompt appropriate to the event content to set the context.

[1392] Output: Initialized ChatGPT model

[1393] Step 3:

[1394] First chat generation and distribution

[1395] When an event starts, the server will generate the initial questions and comments from the ChatGPT model, and the chat will be streamed in real time.

[1396] Input: Initialized ChatGPT model

[1397] Operation:

[1398] The server calls the "generate_initial_chat(model, context)" method to generate the initial chat content.

[1399] The generated chat content is sent to the chat system using the "send_to_chat_system(chat_content)" method.

[1400] Output: First chat

[1401] Step 4:

[1402] Collecting reactions from viewers

[1403] The device collects chat content and reactions from viewers in real time and sends them to the server.

[1404] Input: Viewer input data (e.g., chat messages, facial expression analysis data)

[1405] Operation:

[1406] The terminal executes the "receive_user_input()" method to collect input from the user.

[1407] The collected data is sent to the server using the "send_to_server(user_input)" method.

[1408] Output: Viewer input data

[1409] Step 5:

[1410] Emotion engine collaboration

[1411] The server passes viewer reaction data to the emotion engine in real time to analyze user emotions.

[1412] Input: Viewer input data

[1413] Operation:

[1414] The server sends reaction data to the emotion engine using the "send_to_emotion_engine(reaction_data)" method.

[1415] The emotion engine analyzes the data and returns emotion results to the server.

[1416] Output: Emotion analysis results

[1417] Step 6:

[1418] Real-time chat generation

[1419] The server uses the ChatGPT model to generate new questions and comments based on viewer reaction data and the analysis results of the emotion engine, and distributes them in real time.

[1420] Input: Sentiment analysis results, viewer input data

[1421] Operation:

[1422] The server executes the "generate_new_chat(model, feedback)" method to generate new chat content.

[1423] The generated new chat content is distributed using "send_to_chat_system(chat_content)".

[1424] Output: New chat transcript

[1425] Step 7:

[1426] Data storage and feedback collection

[1427] After the event, the server stores all reaction and emotion data in a database and delivers a feedback form to viewers.

[1428] Input: All data after the event

[1429] Operation:

[1430] The server saves all data to the database using "save_data_to_database(data)".

[1431] Send the feedback form to viewers using "send_feedback_form()".

[1432] Output: Saved data, delivered feedback form

[1433] (Application example 2)

[1434] 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."

[1435] Conventional online events lacked mechanisms to effectively promote interactive communication with viewers. As a result, it was difficult to respond immediately to viewers' interests and reactions, resulting in a decline in viewer engagement. Furthermore, there was no technology that could analyze viewers' emotions in real time and dynamically improve chat content using generative AI models based on that analysis. This made it difficult to provide optimal content that met viewers' expectations.

[1436] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for using a generative AI model to generate and distribute chat content tailored to viewer interests during an online event; a terminal for displaying the generated chat content; means for collecting viewer reactions and providing feedback to the AI ​​model in real time; means for analyzing viewer emotions using an emotion analysis engine and improving the generated chat content based on the analysis results; means for constantly generating new questions and comments based on viewer reactions and distributing them in real time; and means for distributing a feedback form to viewers after the event ends and identifying areas for improvement for the next event. This makes it possible to significantly increase viewer engagement by immediately responding to viewer emotions and reactions and providing optimal chat content.

[1437] A "generative AI model" is an artificial intelligence model that takes data as input and automatically generates content based on specific contexts and conditions.

[1438] "Chat content" refers to text messages such as questions and comments that are generated by a generative AI model during an online event and distributed to viewers.

[1439] A "terminal" is a device that displays the generated chat content, accepts input from viewers, and transmits it to the server.

[1440] "Viewer reactions" refers to feedback such as facial expressions, gestures, voice, and text comments given by viewers during an online event.

[1441] "Real-time feedback" refers to a mechanism that instantly senses viewer reactions and sends them to a generative AI model for use in improving chat content.

[1442] An "emotion analysis engine" is software or hardware that analyzes emotions from viewers' facial expressions and voices and provides the results to a generative AI model.

[1443] A "persona" is a virtual character based on a specific audience or user profile, which the generative AI model uses as a reference when generating chat content.

[1444] A "feedback form" is a form provided to viewers after an online event to allow them to enter their evaluation of the event and suggestions for improvement.

[1445] This invention is a system that uses a generative AI model and an emotion analysis engine to generate chat content in real time during online events based on viewer interests, promoting two-way communication with viewers. This system consists of four main components: a server, terminals, viewers, and an emotion analysis engine.

[1446] Processing performed by the server

[1447] 1. Loading event data

[1448] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[1449] 2. ChatGPT model initialization and context preparation

[1450] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[1451] 3. Generating and delivering the first chat

[1452] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different personas (e.g., persona A, persona B, etc.) and are distributed to the chat system in real time.

[1453] 4. Emotion engine integration

[1454] The server collects viewer reaction data in real time and passes it to a sentiment analysis engine to analyze user emotions. The analysis results are fed back to the ChatGPT model, which refines the generated chat content according to viewer emotions.

[1455] 5. Real-time chat generation

[1456] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the sentiment analysis engine, and distributes the generated content in real time.

[1457] 6. Data storage and feedback collection

[1458] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[1459] Processing performed by the device

[1460] 1. Displaying chat contents

[1461] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[1462] 2. Receiving and sending user input

[1463] When viewers chat or respond to questions, the device sends that information to the server, along with facial expression and voice analysis data from the viewers to the emotion analysis engine.

[1464] 3. Providing a feedback form

[1465] After the event, the device will display a feedback form, allowing viewers to easily enter their opinions and thoughts.

[1466] User actions

[1467] 1. View and respond to chat

[1468] Viewers can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them.

[1469] 2. Submit your own questions or comments

[1470] Viewers can enter their own concerns or questions and send them to the chat system.

[1471] 3. Providing Feedback

[1472] After the event, viewers can fill out a feedback form and submit their thoughts and opinions, which will help improve the quality of future events.

[1473] Hardware / Software used

[1474] Server: AWS EC2 instance, or other cloud-based virtual machine

[1475] Sentiment Analysis Engine: Affectiva SDK, or other sentiment analysis platform

[1476] Chat display devices: smartphones, tablets, PCs

[1477] Real-time databases: Firebase Realtime Database, MongoDB

[1478] Generative AI model: OpenAI GPT-4 API or other advanced generative AI models

[1479] Prompt Sentence Examples

[1480] "User emotion is {emotion}. Event data is:\n{event_data}\nPlease generate an appropriate question."

[1481] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1482] Step 1:

[1483] Before the online event starts, the server loads the event details and persona information from the database, which provides basic information about the event theme, speakers, and topics covered. The input is the event ID, and the output is the event details and persona information.

[1484] Step 2:

[1485] The server initializes the ChatGPT model based on the loaded event details and persona settings, and prepares the context. This provides the context for the generative AI model to generate questions and comments related to the event. The input is the event details and persona settings, and the output is the initialized ChatGPT model.

[1486] Step 3:

[1487] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments. The generated questions and comments reflect different personas (e.g., Persona A, Persona B) and are delivered to the chat system in real time. The input is the initialized ChatGPT model, and the output is the generated initial questions and comments.

[1488] Step 4:

[1489] The terminal displays the chat content sent from the server in real time. Viewers view and respond to generated questions and comments. The input is the chat content sent from the server, and the output is the chat content that viewers see on their terminal.

[1490] Step 5:

[1491] When viewers enter responses to chats or questions, the device sends that information to the server. It also sends the viewer's facial expressions and voice analysis data to the emotion analysis engine. The input is the viewer's responses and emotion data, and the output is the data sent to the server and the emotion analysis engine.

[1492] Step 6:

[1493] The server sends viewer reaction data to the sentiment analysis engine in real time to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions. The input is viewer reaction data and sentiment analysis results, and the output is refined chat content.

[1494] Step 7:

[1495] The server sends a request to the ChatGPT model to generate new questions and comments based on the viewer reaction data and the analysis results from the sentiment analysis engine, and distributes the generated content in real time. The input is the analysis results and the ChatGPT model, and the output is new questions and comments generated in real time.

[1496] Step 8:

[1497] After the event, the server stores all the reaction and emotion data in a database and sends a feedback form to the viewers to collect improvements for the next event. The input is the viewer's reaction and emotion data, and the output is the stored data and the feedback form.

[1498] 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.

[1499] 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.

[1500] 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.

[1501] [Fourth embodiment]

[1502] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1503] 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.

[1504] 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).

[1505] 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.

[1506] 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.

[1507] 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).

[1508] 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.

[1509] 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.

[1510] 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.

[1511] 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.

[1512] 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.

[1513] 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.

[1514] 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."

[1515] This invention is a system that uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[1516] Processing performed by the server

[1517] 1. Loading event data

[1518] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[1519] 2. ChatGPT model initialization and context preparation

[1520] The server initializes the generative AI model (ChatGPT) and passes the loaded event details and persona setting information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[1521] 3. Generating and delivering the first chat

[1522] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[1523] 4. Real-time chat generation

[1524] During the event, the server collects audience reactions and feeds them back to the ChatGPT model in real time, generating new questions and comments that can continuously increase audience engagement.

[1525] 5. Data storage and feedback collection

[1526] After the event is over, the server stores all viewer response data and sends a feedback form to the viewer to collect suggestions for improving the next event.

[1527] Processing performed by the device

[1528] 1. Displaying chat contents

[1529] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[1530] 2. Receiving and sending user input

[1531] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[1532] 3. Providing a feedback form

[1533] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1534] User actions

[1535] 1. View and respond to chat

[1536] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[1537] 2. Submit your own questions or comments

[1538] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[1539] 3. Providing Feedback

[1540] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[1541] Specific examples

[1542] Example 1: Starting an event

[1543] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1544] Terminal: The terminal displays the questions sent from the server in real time.

[1545] User: The user is interested in the question displayed and enters an additional question.

[1546] Example 2: Interaction during an event

[1547] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[1548] Terminal: The terminal will immediately display the new question as it is generated.

[1549] User: The user responds again to the new question and sends a chat.

[1550] Example 3: After the event

[1551] Server: After the event, the server stores all the response data and delivers the feedback form.

[1552] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1553] User: The user fills out a feedback form and submits it.

[1554] This allows the system of the present invention to facilitate two-way communication at online events and increase audience engagement.

[1555] The processing flow will be explained below.

[1556] Step 1:

[1557] The server loads the event details and persona configuration information from the database a few minutes before the start of the online event, capturing basic information about the event theme, speakers, and topics covered.

[1558] Step 2:

[1559] The server initializes the ChatGPT model with the loaded event details and persona configuration information to prepare the context, so that the AI ​​understands the context to generate questions and comments about the event.

[1560] Step 3:

[1561] At the start of an event, the server sends a request to the ChatGPT model to generate initial questions and comments, including those from different perspectives such as personas A, B, and C.

[1562] Step 4:

[1563] The server distributes the generated questions and comments to a real-time chat system, which displays the initial chat content on the viewers' devices.

[1564] Step 5:

[1565] The terminal displays the chat contents delivered from the server to the user, allowing the user to view questions and comments generated in real time.

[1566] Step 6:

[1567] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[1568] Step 7:

[1569] The device sends the user's input responses and questions to the server, and the data sent becomes important information that enhances the interactivity of the event.

[1570] Step 8:

[1571] The server receives the reaction data sent by users and inputs it into the ChatGPT model as feedback, which allows the AI ​​to improve the accuracy of the questions and comments it generates next.

[1572] Step 9:

[1573] The server then sends a request to the ChatGPT model to generate new questions and comments, and distributes the generated content to the real-time chat system. This process is repeated throughout the event.

[1574] Step 10:

[1575] After the event, the server stores all reaction data in a database, which helps identify improvements for future events and audience interests.

[1576] Step 11:

[1577] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[1578] Step 12:

[1579] The terminal causes the feedback form to be displayed and easily accessible for the user to complete.

[1580] Step 13:

[1581] Users can enter their opinions and thoughts in the feedback form and press the submit button. The feedback will help improve the quality of the next event.

[1582] Step 14:

[1583] The device forwards the feedback sent by the user to the server, which stores it in a database and uses it for analysis.

[1584] Example 1

[1585] 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."

[1586] With conventional online events, it was difficult to engage with viewers in two-way communication that matched their interests, making it difficult to encourage active participation. Furthermore, there was a lack of a system for effectively collecting viewers' reactions and feedback and reflecting that information in future events.

[1587] 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.

[1588] In this invention, the server includes means for generating and distributing chat content according to viewer interests during an online event, a display device for displaying the generated chat content, means for collecting viewer reactions and providing real-time feedback to the generative AI model, means for acquiring detailed event information and persona setting information in advance, and means for generating new questions and comments in real-time based on viewer reactions during the online event, thereby attracting viewer interest, effectively promoting two-way communication, and improving the quality of future events.

[1589] A "generative AI model" is an artificial intelligence system designed to perform tasks such as text generation.

[1590] An "online event" is an event such as a lecture, seminar, or conference that is held in real time via the Internet.

[1591] "Viewer" means a person who participates in an online event and watches or listens to the content in real time.

[1592] "Chat content" is a collection of statements, questions, and comments displayed in text format.

[1593] A "display device" is a device used by a user to view the generated chat content, and includes a personal computer, a smartphone, a tablet, and the like.

[1594] "Reactions" are reactions such as comments, questions, and opinions that viewers show during the event.

[1595] "Feedback" refers to information such as opinions, ratings, and impressions collected from viewers.

[1596] "Event details" refers to information about the online event, such as the theme, speakers, and topics.

[1597] "Persona setting information" is information about a virtual character based on the viewer's interests and attributes.

[1598] "New questions and comments" are new textual statements or questions that are generated based on viewer reactions.

[1599] This system uses a generative AI model to generate chat content based on the interests of viewers during online events, promoting interactive communication among viewers. This system consists of three main components: a server, a terminal, and a user.

[1600] Server Roles

[1601] The server loads online event details and persona configuration information from a database such as MySQL. This information includes the event theme, speakers, planned topics, and audience persona information. The server then initializes OpenAI's ChatGPT model and passes it the event details and persona information. Once this is ready, the server generates the initial chat content and distributes it to the chat system in real time.

[1602] For example, the server loads information about a webinar on AI technology and sends a prompt to ChatGPT to ask viewers what topic in AI technology they are most interested in. The generated questions are then sent to the display device in real time.

[1603] Device Role

[1604] The device displays chat content sent from the server in real time, allowing viewers to easily view the questions and comments that have been generated. When viewers type responses to chats or questions, the device sends the information to the server. After the event ends, a feedback form is displayed, allowing viewers to easily enter their opinions and thoughts.

[1605] As a specific example, a viewer types a response into their smartphone, such as "I'm interested in image recognition using AI," and this is sent to the server.

[1606] User Roles

[1607] Users can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them, stimulating two-way communication. Users can also enter their own concerns or questions and submit them to the chat system. After the event, users can enter their opinions and impressions in a feedback form and submit it to help improve the quality of the next event.

[1608] As a concrete example, a user types the question "How does AI learn?" and sends it to the chat.

[1609] Prompt Sentence Examples

[1610] Below is an example of a prompt sentence.

[1611] "What topic in AI technology interests you most?"

[1612] "Please tell us about the latest technological trends using AI."

[1613] What was the most memorable aspect of what was said at the event?

[1614] In this system, the server collects viewer reactions and feeds them back into the ChatGPT model to generate new questions and comments in real time. After the event ends, the server saves all viewer reaction data and sends a feedback form. Through this process, the system of the present invention can increase viewer interest and effectively promote two-way communication.

[1615] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1616] Step 1: Loading event data

[1617] Specifically: Before the start of an online event, the server loads the event details (theme, speakers, topics, etc.) and persona configuration information from a database, which is necessary to generate questions and comments related to the event content.

[1618] Input: Database connection information (e.g. MySQL query), event identification information.

[1619] Data processing: Executes SQL queries to retrieve data about events and loads it into memory.

[1620] Output: A data object containing event details and persona configuration information.

[1621] Specific operation: The server executes an SQL query such as "SELECT FROM events WHERE id=1" to retrieve information about the "2023 AI Technology Webinar."

[1622] Step 2: Initializing the ChatGPT model and preparing the context

[1623] Specifically: The server initializes the ChatGPT model with the retrieved event details and persona configuration information, allowing the model to have the right context to generate questions and comments based on the event content.

[1624] Input: Event details and persona configuration information.

[1625] Data processing: Obtaining the API key and configuration parameters required for initialization, and passing context data to the ChatGPT model.

[1626] Output: The initialized ChatGPT model.

[1627] What happens: The server uses the openai.ChatCompletion.create method in Python to pass event information (e.g., the speaker is John Doe, and the topic is AI and medical technology) to the model.

[1628] Step 3: Generate and deliver the first chat

[1629] Specifically: When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate immediately.

[1630] Input: An initialized ChatGPT model, a trigger to start the event.

[1631] Data processing: Enter the prompt text into ChatGPT and format the generated data.

[1632] Output: Data from the initial question and comment.

[1633] How it works: The server generates a question from the ChatGPT model: "What topic are you most interested in in AI technology?" and sends it to the chat system via WebSocket or HTTP API.

[1634] Step 4: Real-time chat generation

[1635] During the event, the server collects audience reactions in real time and feeds them back to the ChatGPT model, allowing it to continuously generate new questions and comments.

[1636] Input: Viewer response data.

[1637] Data processing: Collect and analyze viewer response data and provide it to ChatGPT as feedback.

[1638] Output: Question and comment data updated in real time.

[1639] How it works: A viewer responds by saying, "I'm interested in AI image recognition," and the server then tells ChatGPT, "I'd like to know more about applications in the medical field." The model then generates a new question and distributes it to the chat system.

[1640] Step 5: Store data and collect feedback

[1641] After the event, the server will save all viewer response data and send a feedback form to the viewer, allowing us to collect their opinions for improving the next event.

[1642] Input: Event end trigger, viewer response data.

[1643] Data processing: Response data storage, feedback form generation and distribution.

[1644] Output: Saved data object, feedback form delivered to viewers.

[1645] What it does: The server saves all chat data from viewers in a MySQL database and sends feedback forms such as "How satisfied were you with the event?" via Google Forms or a dedicated application.

[1646] This is the specific processing flow of the program for this system. At each step, the necessary data processing and calculations are carried out based on clearly defined input data, and appropriate output is generated, promoting two-way communication with viewers.

[1647] (Application example 1)

[1648] 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."

[1649] In online events, there is a need for a method to maintain viewer interest, promote interactive communication, and collect event data to improve future events. In particular, there are insufficient methods for generating appropriate topics and questions in real time to encourage active viewer participation, and it is difficult to effectively collect feedback information after the event has ended.

[1650] 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.

[1651] In this invention, the server includes means for using a generative AI model to generate and distribute chat content based on viewer interests during an online event, means for collecting feedback after the event ends and sending it to the server, means for generating product-related questions and comments, a terminal for displaying the generated chat content, and means for collecting reactions from viewers and feeding them back to the AI ​​model in real time. This allows viewers to receive interesting chat content in real time, stimulating two-way communication, and effectively collecting and saving feedback from viewers, making it possible to identify areas for improvement for the next event.

[1652] A "generative AI model" is an artificial intelligence model that generates new text and content based on large amounts of data.

[1653] An "online event" is an event that takes place in real time over the Internet.

[1654] "Viewers" are users who participate in online events, view content, and ask questions.

[1655] "Chat content" refers to text generated by a generative AI model and used for two-way communication with viewers.

[1656] A "terminal" is a device that displays the generated chat content and receives input from viewers.

[1657] "Feedback Form" means an electronic form used to collect opinions and feedback from viewers after an event has concluded.

[1658] "Persona" refers to profile information of a typical audience for a particular event or service.

[1659] The "server" is a central computer system that loads event data, generates chat content, stores viewer reaction data, and so on.

[1660] "Products" means any products or services featured during the Online Event.

[1661] "Real time" means that various processes are carried out at the same time as the event is progressing.

[1662] "Response data" refers to information about interactions, such as comments and questions, that viewers make during an online event.

[1663] This invention is a system that uses generative AI models to generate chat content based on viewer interests at online events, promoting real-time two-way communication, and collecting and saving viewer feedback to help improve future events. The system consists of three main elements: a server, terminals, and users.

[1664] Processing performed by the server

[1665] 1. Loading event data

[1666] Before the online event starts, the server loads the event details and persona configuration information from the database, which allows it to know information such as the event theme, speakers, and topics covered.

[1667] 2. ChatGPT model initialization and context preparation

[1668] The server initializes a generative AI model (e.g., ChatGPT) and passes the loaded event details and persona configuration information as input, allowing the AI ​​model to generate questions and comments related to the event content.

[1669] 3. Generating and delivering the first chat

[1670] When an event starts, the server sends a request to the ChatGPT model to generate the initial questions and comments, which are then distributed to the chat system in real time, allowing viewers to participate.

[1671] 4. Real-time chat generation

[1672] During the event, the server collects viewers' reactions and feeds them back to the ChatGPT model in real time to generate new questions and comments, which can continuously stimulate viewers' interest. It also generates product-related questions and comments, promoting two-way communication.

[1673] 5. Data storage and feedback collection

[1674] After the event is over, the server stores all viewer response data. After the event is over, a feedback form is sent to the viewer to collect suggestions for improvements to the next event. The collected feedback is stored on the server and used for the next event.

[1675] Processing performed by the device

[1676] 1. Displaying chat contents

[1677] The terminal displays the chat contents sent from the server in real time, allowing viewers to easily view generated questions and comments.

[1678] 2. Receiving and sending user input

[1679] When a user enters a chat or a response to a question, the terminal sends the information to the server.

[1680] 3. Providing a feedback form

[1681] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1682] User actions

[1683] 1. View and respond to chat

[1684] Users can view the chat content generated by the generative AI model on their devices and respond to questions or comments that interest them, thereby stimulating two-way communication.

[1685] 2. Submit your own questions or comments

[1686] Users can input their own concerns or questions and send them to the chat system, which allows for deeper communication.

[1687] 3. Providing Feedback

[1688] After the event, users can enter their opinions and impressions in a feedback form and submit it, which will help improve the quality of the next event.

[1689] Specific examples

[1690] Example 1: Starting an event

[1691] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1692] Terminal: The terminal displays the questions sent from the server in real time.

[1693] User: The user is interested in the question displayed and enters an additional question.

[1694] Example 2: Interaction during an event

[1695] Server: The server collects user responses in real time and gives feedback to ChatGPT to generate new questions.

[1696] Terminal: The terminal will immediately display the new question as it is generated.

[1697] User: The user responds again to the new question and sends a chat.

[1698] Example 3: After the event

[1699] Server: After the event, the server stores all the response data and delivers the feedback form.

[1700] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1701] User: The user fills out a feedback form and submits it.

[1702] Specific examples of use

[1703] Example prompt sentence:

[1704] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[1705] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1706] Step 1:

[1707] Before the event begins, the server loads the event details and persona setting information from the database. Specifically, it obtains the event theme, speaker information, target topics, etc. This information serves as the basic data for the generative AI model to generate appropriate chat content. The input is the event ID, and the output is the event details and persona setting information.

[1708] Step 2:

[1709] The server initializes the ChatGPT model based on the acquired event details and persona configuration information. This information is input into the model as context data, and the generated chat content is prepared to be event-related. The input is the event details and persona configuration information, and the output is the initialized ChatGPT model.

[1710] Step 3:

[1711] When an event starts, the server sends a request to the ChatGPT model to generate the initial question and comment. The generative AI model generates appropriate questions and comments based on the provided context and returns them to the server. The server then distributes the generated content to the chat system so that viewers can view it. The input is the initialized ChatGPT model, and the output is the initial question and comment.

[1712] Step 4:

[1713] The terminal displays the chat content sent from the server in real time. Viewers can check the generated questions and comments on the terminal and make appropriate responses. The input is the chat content sent from the server, and the output is the displayed chat content.

[1714] Step 5:

[1715] Users input questions and comments in response to the chat content displayed on their devices. The input data is sent from the device to the server. The input is the user's question or comment, and the output is the data sent from the device to the server.

[1716] Step 6:

[1717] The server collects user responses in real time and feeds them back to the generative AI model. The model generates new questions and comments, promoting two-way communication. The input is user response data, and the output is the generated new questions and comments.

[1718] Step 7:

[1719] When the event ends, the server saves all the reaction data. This records what interactions took place during the event and serves as a reference for the next event. The input is all the reaction data after the event ends, and the output is the saved reaction data.

[1720] Step 8:

[1721] After the event, the server distributes a feedback form to viewers. Viewers fill out the form with their opinions and thoughts and send it to the server. The input is the feedback form distributed to the viewer, and the output is the filled-in feedback.

[1722] Step 9:

[1723] The device passes the feedback sent by the viewer to the server, which stores it. This feedback is used to identify improvements for the next event. The input is the viewer feedback, and the output is the stored feedback data.

[1724] For example:

[1725] Regarding prompt generation, by inputting the following prompt into the model, appropriate chat content is generated:

[1726] "You are an AI assistant in a virtual store. The event theme is 'New Product Launch' and the speaker is 'Technologist'. Your persona is a technology enthusiast."

[1727] 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.

[1728] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[1729] Processing performed by the server

[1730] 1. Loading event data

[1731] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[1732] 2. ChatGPT model initialization and context preparation

[1733] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[1734] 3. Generating and delivering the first chat

[1735] When an event starts, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different perspectives (e.g., persona A, B, C), and are delivered to the chat system in real time.

[1736] 4. Emotion engine integration

[1737] The server collects viewer reaction data in real time and passes it to the emotion engine to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions.

[1738] 5. Real-time chat generation

[1739] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the emotion engine, and distributes the generated content in real time.

[1740] 6. Data storage and feedback collection

[1741] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[1742] Processing performed by the device

[1743] 1. Displaying chat contents

[1744] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[1745] 2. Receiving and sending user input

[1746] When a user enters a chat or a response to a question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[1747] 3. Providing a feedback form

[1748] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1749] User actions

[1750] 1. View and respond to chat

[1751] Users can view the chat content generated by ChatGPT on their device and respond to questions or comments that interest them.

[1752] 2. Submit your own questions or comments

[1753] Users can input their own concerns or questions and send them to the chat system.

[1754] 3. Providing Feedback

[1755] After the event, users can enter their opinions and thoughts in a feedback form and submit it, which will help improve the quality of future events.

[1756] Specific examples

[1757] Example 1: Starting an event

[1758] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1759] Terminal: The terminal displays the questions sent from the server in real time.

[1760] User: A user is interested in a question displayed and enters a new question or comment.

[1761] Example 2: Interaction during an event

[1762] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[1763] Terminal: The terminal will immediately display the new question as it is generated.

[1764] User: A user responds to new questions and sends chats.

[1765] Example 3: After the event

[1766] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[1767] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1768] User: The user fills out a feedback form and submits it.

[1769] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[1770] The processing flow will be explained below.

[1771] Step 1:

[1772] Before the online event starts, the server loads the event details and persona configuration information from the database, including the event theme, speakers, and topics covered.

[1773] Step 2:

[1774] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the questions and comments generated are appropriate for the event.

[1775] Step 3:

[1776] At the start of an event, the server sends a request to the ChatGPT model to generate the initial questions and comments, which include content based on different personas (e.g., personas A, B, C).

[1777] Step 4:

[1778] The server then broadcasts the generated questions and comments to a real-time chat system, allowing viewers to see the original chat transcript.

[1779] Step 5:

[1780] The terminal displays the chat contents sent from the server to the user, allowing the user to view questions and comments generated in real time.

[1781] Step 6:

[1782] Users can respond to the displayed questions and comments by typing replies or adding new questions.

[1783] Step 7:

[1784] The device sends the user's input responses and questions to the server, and simultaneously sends the user's facial expressions and voice data to the emotion engine.

[1785] Step 8:

[1786] The emotion engine analyzes the user's facial and voice data to identify the user's emotional state, and sends the recognized emotion data to the server.

[1787] Step 9:

[1788] The server receives reaction data sent by users and emotion data from the emotion engine, and feeds this back to the ChatGPT model to improve the accuracy of the next questions and comments it generates.

[1789] Step 10:

[1790] The server again sends requests to the ChatGPT model to generate new questions and comments, and delivers them to the chat system in real time. This process is repeated throughout the event.

[1791] Step 11:

[1792] After the event, the server stores all reaction and sentiment data in a database, which allows for analysis and improvement for future events.

[1793] Step 12:

[1794] The server distributes a feedback form to viewers after the event, which includes questions about the event's evaluation and areas for improvement.

[1795] Step 13:

[1796] The terminal displays a feedback form to the user for easy access and completion.

[1797] Step 14:

[1798] Users can enter their opinions and thoughts in the feedback form and press the submit button to send the data to the server, which will help improve the quality of the next event.

[1799] Step 15:

[1800] The server stores the submitted feedback data in a database and uses it for analysis to identify areas for improvement for the next event.

[1801] Example 2

[1802] 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."

[1803] To promote viewer engagement and interactive communication at online events, a system is needed that can generate chat content in real time based on viewer interests and emotions and instantly reflect their feedback. However, conventional systems do not adequately generate content based on viewer emotions or generate chat content from diverse perspectives using personas, resulting in a decline in viewer participation.

[1804] 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 a means for generating and distributing chat content according to viewer interests during an online event using a generative AI model, a terminal for displaying the generated chat content, a means for collecting reactions from viewers and feeding them back to the AI ​​model in real time, an engine for analyzing viewer emotions, and a means for improving the chat content based on the viewer emotions. This enables real-time chat generation and feedback according to viewer emotions and reactions.

[1805] A "generative AI model" is a type of artificial intelligence that automatically generates chat content based on viewers' interests and reactions.

[1806] An "online event" is an event or webinar that takes place live over the internet and allows participants to participate in it in real time.

[1807] "Viewers" are users who participate in an online event, who watch the event content, and participate through chat.

[1808] "Chat content" refers to messages such as questions and comments generated by the generative AI model, and is information used to communicate with viewers during online events.

[1809] "Terminal" refers to a device used by a viewer, such as a computer, smartphone, or tablet, that can display and input chat content.

[1810] "Reactions" refer to responses such as comments, questions, and facial expression analysis data made by viewers in response to chat content.

[1811] "Feedback" refers to the process of collecting and analyzing viewer reaction data, feeding the results back into the generative AI model, and reflecting them in the generation of the next chat content.

[1812] An "emotion engine" is a system that analyzes the viewer's facial expressions and voice data to determine their emotional state.

[1813] A "persona" is a virtual character or user profile created based on the viewer's interests and reactions, and is a factor taken into consideration when generating chat content.

[1814] "Feedback Form" means an input form provided to collect opinions and feedback from viewers after the online event has ended.

[1815] "Analysis" refers to the process of analyzing viewer reaction data and emotional data using statistical or machine learning techniques.

[1816] "Distribution" refers to the process of sending generated chat content, feedback forms, etc. to viewers in real time.

[1817] "Storage" refers to the process of storing data collected during an event in a recording medium such as a database.

[1818] This invention is a system that uses a generative AI model and an emotion engine to generate chat content in real time during online events based on the interests of viewers, promoting two-way communication with viewers. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[1819] server

[1820] The server handles the primary data processing for the online event, and specifically uses the following hardware and software:

[1821] Hardware: High-performance server machine (e.g., a server equipped with an Intel Xeon processor)

[1822] Software: Cloud infrastructure using AWS or Google Cloud, database using MySQL or PostgreSQL, ChatGPT as the generative AI model, and Affectiva as the sentiment analysis engine

[1823] 1. Loading event data

[1824] Before the event starts, the server loads the event details (e.g., event name, start time, speaker information) and persona setting information from the database, thereby obtaining information about the event theme and speakers.

[1825] 2. ChatGPT model initialization and context preparation

[1826] Based on the loaded information, we initialize the ChatGPT model and set the context, which allows the model to generate questions and comments related to the event.

[1827] 3. Generating and delivering the first chat

[1828] Once the event begins, the ChatGPT model generates the initial questions and comments, which are then streamed in real time.

[1829] 4. Emotion engine integration

[1830] The server collects viewer reaction data in real time and passes it to the emotion engine for analysis. The analysis results are fed back to the ChatGPT model to refine the chat content based on viewer emotions.

[1831] 5. Real-time chat generation

[1832] Based on viewer reaction data and the analysis results of the emotion engine, the server uses the ChatGPT model to generate new questions and comments and distribute them in real time.

[1833] 6. Data storage and feedback collection

[1834] After the event, all reaction and sentiment data is stored in a database, and a feedback form is sent to viewers to collect suggestions for improvements to be made to the next event.

[1835] Terminal

[1836] A terminal is a device used by a viewer and has the following functions:

[1837] Hardware: PCs, smartphones, tablets

[1838] Software: Web browser or dedicated application

[1839] 1. Displaying chat contents

[1840] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[1841] 2. Receiving and sending user input

[1842] When a user enters a response to a chat or question, the device sends that information to the server, along with facial expression and voice analysis data from the user to the emotion engine.

[1843] 3. Providing a feedback form

[1844] After the event, the device will display a feedback form, allowing users to easily enter their opinions and thoughts.

[1845] User

[1846] Users participate in online events and interact with the system using their devices, performing the following actions in the process:

[1847] 1. View and respond to chat

[1848] Users can view the chat content generated by ChatGPT on their device and respond to any content that interests them.

[1849] 2. Submit your own questions or comments

[1850] Users can input their own concerns or questions and send them to the chat system.

[1851] 3. Providing Feedback

[1852] After the event, users can enter their opinions and thoughts in a feedback form to help improve the quality of the next event.

[1853] Specific examples

[1854] Below is a concrete example of how this system can be actually operated.

[1855] Example 1: Starting an event

[1856] Server: When an event starts, the server loads the event details and persona information, and generates and distributes the initial questions.

[1857] Terminal: The terminal displays the questions sent from the server in real time.

[1858] User: A user is interested in a question displayed and enters a new question or comment.

[1859] Example 2: Interaction during an event

[1860] Server: The server generates new questions from ChatGPT based on the user's responses and the analysis results of the emotion engine, and distributes them.

[1861] Terminal: The terminal will immediately display the new question as it is generated.

[1862] User: A user responds to new questions and sends chats.

[1863] Example 3: After the event

[1864] Server: After the event, the server stores all the reaction and emotion data and delivers the feedback form.

[1865] Terminal: The terminal displays the feedback form and makes it easily accessible to the user.

[1866] User: The user fills out a feedback form and submits it.

[1867] Prompt Sentence Examples

[1868] "Thank you for joining us for our virtual event today. Please answer the following question: What do you think about the theme of the event?"

[1869] By utilizing a generative AI model that takes user emotions into account, the system of the present invention can highly promote two-way communication at online events and significantly increase audience engagement.

[1870] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1871] Step 1:

[1872] Loading event data

[1873] Before an online event begins, the server loads event details and persona configuration information from a database.

[1874] Input: Event ID

[1875] Operation:

[1876] The server executes the SQL query "SELECT FROM EventDetails WHERE EventID = X".

[1877] The server obtains detailed information about the event (e.g., event name, start time, speaker information) and persona setting information (e.g., characteristics of Persona A and Persona B).

[1878] Output: Event details, persona setting information

[1879] Step 2:

[1880] ChatGPT model initialization and context preparation

[1881] The server initializes the ChatGPT model and sets the context based on the loaded event details and persona configuration information.

[1882] Input: Event details, persona setting information

[1883] Operation:

[1884] The server calls the "init_model(event_details, persona_info)" method to initialize the ChatGPT model.

[1885] After initialization, the model is provided with an initial prompt appropriate to the event content to set the context.

[1886] Output: Initialized ChatGPT model

[1887] Step 3:

[1888] First chat generation and distribution

[1889] When an event starts, the server will generate the initial questions and comments from the ChatGPT model, and the chat will be streamed in real time.

[1890] Input: Initialized ChatGPT model

[1891] Operation:

[1892] The server calls the "generate_initial_chat(model, context)" method to generate the initial chat content.

[1893] The generated chat content is sent to the chat system using the "send_to_chat_system(chat_content)" method.

[1894] Output: First chat

[1895] Step 4:

[1896] Collecting reactions from viewers

[1897] The device collects chat content and reactions from viewers in real time and sends them to the server.

[1898] Input: Viewer input data (e.g., chat messages, facial expression analysis data)

[1899] Operation:

[1900] The terminal executes the "receive_user_input()" method to collect input from the user.

[1901] The collected data is sent to the server using the "send_to_server(user_input)" method.

[1902] Output: Viewer input data

[1903] Step 5:

[1904] Emotion engine collaboration

[1905] The server passes viewer reaction data to the emotion engine in real time to analyze user emotions.

[1906] Input: Viewer input data

[1907] Operation:

[1908] The server sends reaction data to the emotion engine using the "send_to_emotion_engine(reaction_data)" method.

[1909] The emotion engine analyzes the data and returns emotion results to the server.

[1910] Output: Emotion analysis results

[1911] Step 6:

[1912] Real-time chat generation

[1913] The server uses the ChatGPT model to generate new questions and comments based on viewer reaction data and the analysis results of the emotion engine, and distributes them in real time.

[1914] Input: Sentiment analysis results, viewer input data

[1915] Operation:

[1916] The server executes the "generate_new_chat(model, feedback)" method to generate new chat content.

[1917] The generated new chat content is distributed using "send_to_chat_system(chat_content)".

[1918] Output: New chat transcript

[1919] Step 7:

[1920] Data storage and feedback collection

[1921] After the event, the server stores all reaction and emotion data in a database and delivers a feedback form to viewers.

[1922] Input: All data after the event

[1923] Operation:

[1924] The server saves all data to the database using "save_data_to_database(data)".

[1925] Send the feedback form to viewers using "send_feedback_form()".

[1926] Output: Saved data, delivered feedback form

[1927] (Application example 2)

[1928] 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."

[1929] Conventional online events lacked mechanisms to effectively promote interactive communication with viewers. As a result, it was difficult to respond immediately to viewers' interests and reactions, resulting in a decline in viewer engagement. Furthermore, there was no technology that could analyze viewers' emotions in real time and dynamically improve chat content using generative AI models based on that analysis. This made it difficult to provide optimal content that met viewers' expectations.

[1930] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for using a generative AI model to generate and distribute chat content tailored to viewer interests during an online event; a terminal for displaying the generated chat content; means for collecting viewer reactions and providing feedback to the AI ​​model in real time; means for analyzing viewer emotions using an emotion analysis engine and improving the generated chat content based on the analysis results; means for constantly generating new questions and comments based on viewer reactions and distributing them in real time; and means for distributing a feedback form to viewers after the event ends and identifying areas for improvement for the next event. This makes it possible to significantly increase viewer engagement by immediately responding to viewer emotions and reactions and providing optimal chat content.

[1931] A "generative AI model" is an artificial intelligence model that takes data as input and automatically generates content based on specific contexts and conditions.

[1932] "Chat content" refers to text messages such as questions and comments that are generated by a generative AI model during an online event and distributed to viewers.

[1933] A "terminal" is a device that displays the generated chat content, accepts input from viewers, and transmits it to the server.

[1934] "Viewer reactions" refers to feedback such as facial expressions, gestures, voice, and text comments given by viewers during an online event.

[1935] "Real-time feedback" refers to a mechanism that instantly senses viewer reactions and sends them to a generative AI model for use in improving chat content.

[1936] An "emotion analysis engine" is software or hardware that analyzes emotions from viewers' facial expressions and voices and provides the results to a generative AI model.

[1937] A "persona" is a virtual character based on a specific audience or user profile, which the generative AI model uses as a reference when generating chat content.

[1938] A "feedback form" is a form provided to viewers after an online event to allow them to enter their evaluation of the event and suggestions for improvement.

[1939] This invention is a system that uses a generative AI model and an emotion analysis engine to generate chat content in real time during online events based on viewer interests, promoting two-way communication with viewers. This system consists of four main components: a server, terminals, viewers, and an emotion analysis engine.

[1940] Processing performed by the server

[1941] 1. Loading event data

[1942] Before the online event starts, the server loads the event details and persona configuration information from the database, which provides basic information about the event theme, speakers, and topics to be covered.

[1943] 2. ChatGPT model initialization and context preparation

[1944] The server initializes the ChatGPT model based on the loaded event details and persona configuration information, and prepares the context so that the generative AI model can understand the context to generate questions and comments related to the event.

[1945] 3. Generating and delivering the first chat

[1946] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments, which reflect different personas (e.g., persona A, persona B, etc.) and are distributed to the chat system in real time.

[1947] 4. Emotion engine integration

[1948] The server collects viewer reaction data in real time and passes it to a sentiment analysis engine to analyze user emotions. The analysis results are fed back to the ChatGPT model, which refines the generated chat content according to viewer emotions.

[1949] 5. Real-time chat generation

[1950] The server sends a request to the ChatGPT model to generate new questions and comments based on viewer reaction data and analysis results from the sentiment analysis engine, and distributes the generated content in real time.

[1951] 6. Data storage and feedback collection

[1952] After the event is over, the server stores all reaction and emotional data in a database and also sends a feedback form to viewers to collect improvements for the next event.

[1953] Processing performed by the device

[1954] 1. Displaying chat contents

[1955] The device displays the chat content sent from the server in real time, allowing viewers to view and respond to generated questions and comments.

[1956] 2. Receiving and sending user input

[1957] When viewers chat or respond to questions, the device sends that information to the server, along with facial expression and voice analysis data from the viewers to the emotion analysis engine.

[1958] 3. Providing a feedback form

[1959] After the event, the device will display a feedback form, allowing viewers to easily enter their opinions and thoughts.

[1960] User actions

[1961] 1. View and respond to chat

[1962] Viewers can view the chat content generated by ChatGPT on their devices and respond to questions or comments that interest them.

[1963] 2. Submit your own questions or comments

[1964] Viewers can enter their own concerns or questions and send them to the chat system.

[1965] 3. Providing Feedback

[1966] After the event, viewers can fill out a feedback form and submit their thoughts and opinions, which will help improve the quality of future events.

[1967] Hardware / Software used

[1968] Server: AWS EC2 instance, or other cloud-based virtual machine

[1969] Sentiment Analysis Engine: Affectiva SDK, or other sentiment analysis platform

[1970] Chat display devices: smartphones, tablets, PCs

[1971] Real-time databases: Firebase Realtime Database, MongoDB

[1972] Generative AI model: OpenAI GPT-4 API or other advanced generative AI models

[1973] Prompt Sentence Examples

[1974] "User emotion is {emotion}. Event data is:\n{event_data}\nPlease generate an appropriate question."

[1975] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1976] Step 1:

[1977] Before the online event starts, the server loads the event details and persona information from the database, which provides basic information about the event theme, speakers, and topics covered. The input is the event ID, and the output is the event details and persona information.

[1978] Step 2:

[1979] The server initializes the ChatGPT model based on the loaded event details and persona settings, and prepares the context. This provides the context for the generative AI model to generate questions and comments related to the event. The input is the event details and persona settings, and the output is the initialized ChatGPT model.

[1980] Step 3:

[1981] When an event is initiated, the server sends a request to the ChatGPT model to generate initial questions and comments. The generated questions and comments reflect different personas (e.g., Persona A, Persona B) and are delivered to the chat system in real time. The input is the initialized ChatGPT model, and the output is the generated initial questions and comments.

[1982] Step 4:

[1983] The terminal displays the chat content sent from the server in real time. Viewers view and respond to generated questions and comments. The input is the chat content sent from the server, and the output is the chat content that viewers see on their terminal.

[1984] Step 5:

[1985] When viewers enter responses to chats or questions, the device sends that information to the server. It also sends the viewer's facial expressions and voice analysis data to the emotion analysis engine. The input is the viewer's responses and emotion data, and the output is the data sent to the server and the emotion analysis engine.

[1986] Step 6:

[1987] The server sends viewer reaction data to the sentiment analysis engine in real time to analyze user emotions. The analysis results are fed back to the ChatGPT model to refine the generated chat content according to viewer emotions. The input is viewer reaction data and sentiment analysis results, and the output is refined chat content.

[1988] Step 7:

[1989] The server sends a request to the ChatGPT model to generate new questions and comments based on the viewer reaction data and the analysis results from the sentiment analysis engine, and distributes the generated content in real time. The input is the analysis results and the ChatGPT model, and the output is new questions and comments generated in real time.

[1990] Step 8:

[1991] After the event, the server stores all the reaction and emotion data in a database and sends a feedback form to the viewers to collect improvements for the next event. The input is the viewer's reaction and emotion data, and the output is the stored data and the feedback form.

[1992] 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.

[1993] 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.

[1994] 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.

[1995] 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.

[1996] 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.

[1997] 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.

[1998] 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).

[1999] 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.

[2000] 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."

[2001] 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.

[2002] 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).

[2003] 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.

[2004] 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.

[2005] 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.

[2006] 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.

[2007] 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.

[2008] 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.

[2009] 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.

[2010] 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.

[2011] 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.

[2012] 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.

[2013] The following is further disclosed regarding the above embodiment.

[2014] (Claim 1)

[2015] A means for generating and distributing chat content based on viewer interests during an online event using a generative AI model;

[2016] a terminal for displaying the generated chat content;

[2017] A means of collecting reactions from viewers and feeding them back to the AI ​​model in real time,

[2018] A system including:

[2019] (Claim 2)

[2020] A means to set various personas and have a generative AI model generate chat content based on them,

[2021] a means for updating the persona setting information based on viewer responses;

[2022] 10. The system of claim 1, comprising:

[2023] (Claim 3)

[2024] A means to store viewer response data and identify areas for improvement for the next event;

[2025] A means to distribute a feedback form to viewers after the event has ended,

[2026] 10. The system of claim 1, comprising:

[2027] "Example 1"

[2028] (Claim 1)

[2029] A means for generating and distributing chat content based on viewer interests during an online event using a generative AI model;

[2030] a display device that displays the generated chat content;

[2031] A means of collecting reactions from viewers and feeding them back to the generative AI model in real time,

[2032] A means to obtain detailed information about the event and persona configuration information in advance,

[2033] A means of generating new questions and comments in real time based on audience responses during online events;

[2034] A system including:

[2035] (Claim 2)

[2036] A method for setting various personas and using a generative AI model to generate chat content based on them,

[2037] a means for updating the persona setting information based on viewer responses;

[2038] A means to store all viewer response data after the event has ended,

[2039] 10. The system of claim 1.

[2040] (Claim 3)

[2041] A means to store viewer response data and identify areas for improvement for the next event;

[2042] A means to distribute a feedback form to viewers after the event has ended,

[2043] a means of gathering feedback from viewers;

[2044] 10. The system of claim 1.

[2045] "Application Example 1"

[2046] (Claim 1)

[2047] A means for generating and distributing chat content based on viewer interests during an online event using a generative AI model;

[2048] a terminal for displaying the generated chat content;

[2049] A means of collecting reactions from viewers and feeding them back to the AI ​​model in real time,

[2050] a means of generating product-related questions and comments;

[2051] A means of collecting feedback after the event and sending it to the server;

[2052] A system including:

[2053] (Claim 2)

[2054] A means to set various personas and have a generative AI model generate chat content based on them,

[2055] a means for updating the persona setting information based on viewer responses;

[2056] A means for instantly displaying the generated chat content;

[2057] 10. The system of claim 1, comprising:

[2058] (Claim 3)

[2059] A means to store viewer response data and identify areas for improvement for the next event;

[2060] A means to distribute a feedback form to viewers after the event has ended,

[2061] means for transmitting the generated feedback form to a server;

[2062] 10. The system of claim 1, comprising:

[2063] "Example 2: Combining Emotion Engines"

[2064] (Claim 1)

[2065] A means for generating and distributing chat content based on viewer interests during an online event using a generative AI model;

[2066] a terminal for displaying the generated chat content;

[2067] A means of collecting reactions from viewers and feeding them back to the AI ​​model in real time,

[2068] An engine that analyzes viewers' emotions,

[2069] A means for improving chat content based on viewer sentiment;

[2070] A system including:

[2071] (Claim 2)

[2072] A means to set various personas and have a generative AI model generate chat content based on them,

[2073] a means for updating the persona setting information based on viewer responses;

[2074] 10. The system of claim 1, comprising:

[2075] (Claim 3)

[2076] A means to store viewer response data and identify areas for improvement for the next event;

[2077] A means to distribute a feedback form to viewers after the event has ended,

[2078] 10. The system of claim 1, comprising:

[2079] "Application example 2 when combining emotion engines"

[2080] (Claim 1)

[2081] A means for generating and distributing chat content based on viewer interests during an online event using a generative AI model;

[2082] a terminal for displaying the generated chat content;

[2083] A means of collecting reactions from viewers and feeding them back to the AI ​​model in real time,

[2084] A means for analyzing viewers' emotions using an emotion analysis engine and improving chat content generated based on the analysis results;

[2085] A way to constantly generate new questions and comments based on audience reactions and distribute them in real time.

[2086] A feedback form will be sent to viewers after the event to identify areas for improvement for the next event.

[2087] A system including:

[2088] (Claim 2)

[2089] A means to set various personas and have a generative AI model generate chat content based on them,

[2090] a means for updating the persona setting information based on viewer responses;

[2091] A means of collecting and analyzing viewer emotional data in real time,

[2092] A means for dynamically improving the generated chat content according to the emotions of the viewers;

[2093] 10. The system of claim 1, comprising:

[2094] (Claim 3)

[2095] A means to store viewer response data and identify areas for improvement for the next event;

[2096] A means to distribute a feedback form to viewers after the event has ended,

[2097] A means to analyze the feedback data and use it to initialize the next generative AI model;

[2098] 10. The system of claim 1, comprising: [Explanation of symbols]

[2099] 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 for generating and distributing chat content based on viewer interests during an online event using a generative AI model; a terminal for displaying the generated chat content; A means of collecting reactions from viewers and feeding them back to the AI ​​model in real time, A system including:

2. A means to set various personas and have a generative AI model generate chat content based on them, a means for updating the persona setting information based on viewer responses; The system of claim 1 , comprising:

3. A means to store viewer response data and identify areas for improvement for the next event; A means to distribute a feedback form to viewers after the event has ended, The system of claim 1 , comprising:

Images