System

Abstract

A system is provided.SOLUTION: The system includes a means for monitoring the progress of the conference, a means for analyzing the statement contents of the participants, a means for generating appropriate reactions and questions on the basis of the analysis result, and a means for displaying and reproducing the generated reactions and questions.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In recent years, video chat has become commonplace with the spread of telework and online learning. However, compared to face-to-face meetings, online meetings often make it difficult to gauge the mood of the room, resulting in few participants speaking and stalling progress. This can lead to reduced meeting efficiency and important opinions not being shared. Other reasons participants may refrain from speaking include differences in the environment and the risk of simultaneous comments. The present invention aims to solve these issues and make online meetings smoother and more lively. [Means for solving the problem]

[0005] The present invention solves the above problems with a system that includes a means for monitoring the progress of a meeting, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions. This system generates backchannels and empathetic comments, reducing silent periods during meetings and encouraging participants to speak up. It also has a function that encourages participants to speak at appropriate times based on their tendency to speak, allowing the discussion to proceed smoothly without stalling.

[0006] "Means for monitoring the progress of a meeting" refers to a means for monitoring the progress of an online meeting in real time and understanding participants' comments, periods of silence, and the progress of the discussion.

[0007] The "means for analyzing the content of participants' statements" refers to a means for converting the content of participants' statements during an online conference from audio to text and analyzing the text data using natural language processing technology.

[0008] "Means for generating appropriate responses and questions based on analysis results" refers to means for generating responses, empathetic comments, and questions based on what participants say and the progress of the meeting.

[0009] "Means for displaying and playing back generated responses and questions" refers to means for displaying the generated responses and questions on the video chat screen of an online conference and playing them back as audio.

[0010] "Agreements and empathetic comments" are reactions and comments inserted at appropriate times to show empathy for what the participants are saying.

[0011] "Participant speech trends" refer to speech patterns and characteristics of specific participants that are estimated based on past speech history and reactions.

[0012] The "function to encourage participants to speak at the appropriate time" is a function that generates appropriate questions and prompting comments, taking into consideration the progress of the meeting and the speaking tendencies of the participants, to make it easier for participants to speak. [Brief explanation of the drawings]

[0013] [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

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] 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).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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."

[0034] The present invention is a system for making online conferences proceed smoothly and lively. The system includes, as its main means, a means for monitoring the progress of the conference, a means for analyzing the content of participants' remarks, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[0035] The specific program processing will be explained below.

[0036] Program processing overview

[0037] 1. Server: Conference preparation and connection

[0038] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[0039] The server checks the connections of the users and starts the conference after confirming that everyone has joined.

[0040] 2. Server: Collecting and analyzing voice data

[0041] The server collects voice data from each user in real time.

[0042] The server analyzes the voice data, converts it into text data, and analyzes the content of the speech using natural language processing (NLP) technology.

[0043] 3. Server: Generates responses and questions

[0044] The server generates appropriate responses and questions based on the analyzed content of the speech.

[0045] For example, if there is silence after a statement, the AI ​​will generate a response such as, "I see, that's interesting."

[0046] Also, when the discussion stagnates, it generates questions such as, "Mr. / Ms. XX, what do you think about this?"

[0047] 4. Device: Display and playback

[0048] The user's terminal displays the responses and questions sent from the server on the screen.

[0049] The device provides visual and auditory feedback to the user, accompanied by animations of AI characters and voice output.

[0050] Example 1: Starting a meeting and responding

[0051] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[0052] 2. The user begins to introduce themselves and the system detects periods of silence.

[0053] 3. The server uses AI to generate a response such as "That's great."

[0054] 4. The user's device displays the backchannel and plays it back aloud.

[0055] Example 2: Facilitation support

[0056] 1. The server detects that the discussion is stalled.

[0057] 2. The server generates a question for a particular user, such as "What is your opinion on this issue?"

[0058] 3. The user's device displays the question on the screen and plays it aloud.

[0059] 4. Users answer questions, getting the discussion moving again.

[0060] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving the efficiency of meetings. Even when there are few participants to speak or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and discussion to proceed in a manner that is in line with the purpose.

[0061] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[0062] The processing flow will be explained below.

[0063] Program processing flow

[0064] Step 1:

[0065] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[0066] Step 2:

[0067] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[0068] Step 3:

[0069] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[0070] Step 4:

[0071] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[0072] Step 5:

[0073] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[0074] Step 6:

[0075] The server generates appropriate responses and questions based on the analysis results. For example, if the server detects silence, the AI ​​will generate a response such as "I see, that's interesting."

[0076] Step 7:

[0077] The server sends the generated responses and questions to each user's device.

[0078] Step 8:

[0079] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0080] Step 9:

[0081] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0082] Step 10:

[0083] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[0084] Step 11:

[0085] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0086] Step 12:

[0087] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[0088] Step 13:

[0089] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[0090] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[0091] Example 1

[0092] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0093] Online meetings require smooth communication between participants, but in reality, progress often stalls. For example, periods of silence and stalled discussions are major factors that reduce meeting efficiency. Furthermore, if some participants do not actively speak up, there is a lack of appropriate encouragement, which can lead to a lack of depth in the overall discussion. It is necessary to resolve these issues, facilitate smooth online meetings, and provide an environment in which all participants can actively participate.

[0094] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0095] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology, means for analyzing the text data using natural language processing technology, means for generating appropriate reactions and queries using a generative AI model, and means for converting the generated text into voice data using speech synthesis technology. This makes it possible to automatically detect periods of silence and stagnation in the discussion, and to generate, display, and play back appropriate backchannels and questions. This facilitates the progress of the conference and provides a conference environment in which all participants are actively involved.

[0096] "Means for monitoring the progress of meetings" refers to technology for monitoring the overall progress of online meetings and detecting periods of silence or stagnation in discussions.

[0097] The "means for analyzing the content of participants' statements" refers to a technology that converts the content of each participant's statements into text data based on the collected voice data, and analyzes the text.

[0098] "Means for generating appropriate responses and questions based on analysis results" refers to technology that automatically generates responses and questions based on analyzed text data to facilitate the progress of meetings.

[0099] "Means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology" refers to technology for collecting voices during a meeting in real time and instantly converting them into text using automatic speech recognition technology.

[0100] "Means for analyzing text data using natural language processing technology" refers to technology that uses natural language processing technology to analyze converted text data and understand its content and intent.

[0101] "Means for generating appropriate reactions and queries using a generative AI model" refers to a technology that uses a generative AI model to generate natural responses and questions based on text data.

[0102] "Means for converting text generated using speech synthesis technology into speech data" refers to a technology that uses speech synthesis technology to convert generated text into speech data that provides visual and auditory feedback to the user.

[0103] The present invention provides a system for making online conferences proceed smoothly and lively. The system includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[0104] System configuration

[0105] Hardware and software used

[0106] Hardware: Servers (high-performance central processing units (CPUs), memory devices (RAM), large-capacity storage), user devices (personal computers (PCs), tablets, smartphones)

[0107] Software: natural language processing engines (e.g., general-purpose cloud natural language processing APIs or commercial natural language processing engines), automatic speech recognition engines (e.g., cloud-based speech recognition services), speech synthesis engines (e.g., cloud-based text-to-speech services), generative AI models (e.g., general-purpose chatbot models)

[0108] Program processing

[0109] 1. Server: Prepares the meeting and sends the meeting invitation link to users via email or notification at the specified date and time. The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[0110] 2. Server: When a meeting starts, the WebRTC protocol is used to collect each user's voice in real time, and the collected voice data is converted to text data using a cloud-based automatic speech recognition service.

[0111] 3. Server: The converted text data is sent to a natural language processing engine, where its content is analyzed. Based on this analyzed data, the meaning of what the participants said is understood.

[0112] 4. Server: Uses a generative AI model to generate appropriate responses and questions from the analyzed data. For example, if there is silence, it generates a response such as "I see, that's interesting," and if the discussion stagnates, it generates a question such as "What is your opinion on this issue?" An example of a prompt is, "If the user finishes speaking and there is silence, generate an appropriate response."

[0113] 5. Server: The generated text of the interjections and questions is sent to a speech synthesis engine, where it is converted into audio data. A cloud-based text-to-speech service is used to output the generated text as audio.

[0114] 6. Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio. The device uses HTML and JavaScript to display the text in the interface and the Web Audio API to play the audio.

[0115] Specific examples

[0116] Example 1: Starting a meeting and responding

[0117] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[0118] 2. The user begins to introduce themselves and the system detects periods of silence.

[0119] 3. The server uses AI to generate a response such as "That's great."

[0120] 4. The user's device displays the response and plays it aloud.

[0121] Example 2: Facilitation support

[0122] 1. The server detects that the discussion is stalled.

[0123] 2. The server generates a question for a specific user, such as "What is your opinion on this issue?"

[0124] 3. The user's device displays the question on the screen and plays it aloud.

[0125] 4. Users answer questions, getting the discussion moving again.

[0126] As described above, this invention can provide an environment in which participants can speak more actively in online meetings, thereby improving the efficiency of meetings. Even when there are few participants or the discussion stagnates, AI can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is in line with the purpose.

[0127] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0128] Step 1:

[0129] Server: Prepares the meeting and sends the meeting invitation link to the user at the specified date and time.

[0130] What happens: The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[0131] Input: A list of pre-registered user email addresses and / or notification system addresses.

[0132] Output: An email or notification containing the meeting invite link.

[0133] Step 2:

[0134] Users: Click the invitation link they receive to join the meeting.

[0135] Specific operation: The user clicks on the link in the email and enters the conference room via a web browser or dedicated app.

[0136] Input: The invitation link the user received.

[0137] Output: Connection status information to the conference platform.

[0138] Step 3:

[0139] Server: Checks user connection status in real time, confirms that everyone is participating, and starts the meeting.

[0140] What it does: The server monitors the user's online status via a WebSocket or HTTP connection.

[0141] Input: Connection log of the conference platform.

[0142] Output: Trigger to start the conference.

[0143] Step 4:

[0144] Server: Collects real-time audio streams from each user during the conference.

[0145] What happens: The server receives the audio data using the WebRTC protocol.

[0146] Input: The audio stream during the conference.

[0147] Output: Collected audio data.

[0148] Step 5:

[0149] Server: The collected voice data is converted into text data using automatic speech recognition (ASR) technology.

[0150] What it does: The server calls a cloud-based automatic speech recognition service to convert the audio data into text.

[0151] Input: Audio data.

[0152] Output: Text data.

[0153] Step 6:

[0154] Server: Analyzes the converted text data using natural language processing (NLP) technology to understand what is being said.

[0155] Specific operation: The server uses a natural language processing engine to analyze the content and intent of the text data.

[0156] Input: Text data.

[0157] Output: Analysis results (understanding of what was said).

[0158] Step 7:

[0159] Server: Uses generative AI models to generate appropriate reactions and queries based on the analysis results.

[0160] Specific operation: The server inputs a prompt sentence into the generative AI model and generates appropriate responses or questions.

[0161] Input: Analysis results and prompt statement.

[0162] Output: Generated text (backchannel or question).

[0163] Step 8:

[0164] Server: The generated text is passed to a speech synthesis engine and converted into voice data.

[0165] What it does: The server uses a cloud-based text-to-speech service to convert the generated text into audio.

[0166] Input: The generated text.

[0167] Output: Audio data.

[0168] Step 9:

[0169] Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio.

[0170] What it does: The device uses HTML and JavaScript to display text and the Web Audio API to play audio.

[0171] Input: Audio and text data sent from the server.

[0172] Output: Screen display, audio playback.

[0173] Step 10:

[0174] User: The user answers the displayed questions by voice, and the voice data is collected and analyzed by the server again.

[0175] Specific operation: The user speaks through the device's microphone, and the voice is sent to the server.

[0176] Input: The user's spoken response.

[0177] Output: New audio data is sent to the server.

[0178] (Application example 1)

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

[0180] Conventional online conference systems analyze participants' comments and generate appropriate responses and questions to facilitate smooth progress in the meeting. However, they have limitations in generating, displaying, and playing back responses in real time. Furthermore, when dealing with customers in virtual stores, there are cases where the conversation with the customer is interrupted or an appropriate response is not given. This can lead to a decrease in customer satisfaction and the loss of business opportunities.

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

[0182] In this invention, the server includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, and a means for generating appropriate responses and questions based on the analysis results. In addition, the server includes a means for monitoring interactions with customers in the virtual space and generating responses and questions at appropriate times, and a means for visually and audibly displaying and playing back the generated responses and questions through avatars. This not only facilitates the progress of online conferences, but also enables more effective customer service in virtual stores.

[0183] "Meeting proceedings" refers to the process by which participants come together to discuss and negotiate a specific topic or theme.

[0184] "Monitoring means" are methods or tools that allow a system to constantly monitor and collect data on specific conditions or events.

[0185] "Means for analyzing speech content" refers to methods and tools for collecting participants' speech and understanding and analyzing its content using technologies such as natural language processing.

[0186] "Means for generating appropriate responses or questions based on the analysis results" refers to methods or tools for automatically generating appropriate responses or questions based on the analyzed utterance content.

[0187] "Means for displaying and playing generated responses and questions" means methods and tools for visually and audibly presenting generated responses and questions to the user.

[0188] "Virtual customer interaction" refers to the act of interacting and communicating with customers in virtual reality or other digital platforms.

[0189] An "avatar" is a digital character that functions as a user's alter ego in a virtual space.

[0190] "Displaying and playing visually and audibly" means providing information to the user both visually and audibly by not only displaying the generated responses and questions on the screen but also playing them as audio.

[0191] A "system" is a collection of integrated devices and software that combines multiple components to achieve specific functions or services.

[0192] This invention is a system for facilitating customer service in online conferences and virtual stores. This system includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for monitoring interactions with customers in a virtual space, means for generating responses and questions at appropriate times, and means for visually and audibly displaying and playing back the generated responses and questions through an avatar.

[0193] Program processing overview

[0194] The server manages meetings and customer service in virtual stores in the following steps: First, the server collects voice data in real time and converts it into text data. Next, it uses natural language processing technology to analyze what is being said and generates appropriate responses and questions based on the analysis results. The generated responses and questions are sent to the user's device and displayed and played back visually and audibly through the avatar.

[0195] Hardware and software used

[0196] Hardware:

[0197] Smartphone

[0198] Head-mounted display (HMD)

[0199] server

[0200] software:

[0201] Python

[0202] Speech Recognition Library

[0203] Natural language processing library (spaCy)

[0204] Generative AI model (OpenAI GPT-3)

[0205] Program processing

[0206] The server uses a speech recognition library to collect voice data and convert it into text data. It then uses a natural language processing library to analyze the text data and generate appropriate responses and questions based on the content. This generation process uses a generative AI model such as OpenAI GPT-3.

[0207] The generated responses and questions are sent to the user's device, where they are displayed visually and audibly through an avatar. For example, if there is silence after a user makes a statement, the system generates a response such as "I see, that's interesting." If the discussion reaches a deadlock, the system generates a question such as "What do you think about this?"

[0208] Examples and prompts

[0209] Example 1: Starting a meeting and responding

[0210] The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting. When the user begins to introduce themselves, if the system detects a period of silence, the server uses AI to generate a backchannel such as "That's great." The user's device displays the backchannel and plays it back aloud.

[0211] Example 2: Facilitation support

[0212] The server detects when the discussion has stalled and generates a question for a specific user, such as "What is your opinion on this issue?" The user's device displays the question on the screen and plays it back aloud. When the user answers the question, the discussion resumes its smooth flow.

[0213] Examples of prompt statements

[0214] "The customer is experiencing periods of silence. Please generate an appropriate response."

[0215] Based on this prompt, the generative AI model generates appropriate responses and questions, which are then provided to the user through the system, making customer interactions in online meetings and virtual stores smoother and more effective.

[0216] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0217] Step 1:

[0218] The server collects voice data using a voice recognition library. Specifically, it records the user's speech in real time and obtains the voice data. The input of this step is the user's raw voice, and the output is voice data.

[0219] Step 2:

[0220] The server converts the collected voice data into text data. Specifically, it analyzes the voice data using a speech recognition library (SpeechRecognition) and converts the content into text format. The input of this step is voice data, and the output is the converted text data.

[0221] Step 3:

[0222] The server analyzes the text data using a natural language processing library (e.g., spaCy). Specifically, it analyzes the context and meaning of the text data and identifies situations such as periods of silence and stalled discussions. The input for this step is the text data, and the output is the analysis results.

[0223] Step 4:

[0224] The server generates appropriate responses and questions based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-3) to generate responses and questions corresponding to the analysis results. The input of this step is the analysis results, and the output is the generated responses and questions.

[0225] Step 5:

[0226] The server sends the generated responses and questions to the user's device. Specifically, it sends the responses and questions to the device via data communication. The input to this step is the generated responses and questions, and the output is the sent data.

[0227] Step 6:

[0228] The device displays the transmitted responses and questions through an avatar and plays them back as audio. Specifically, it uses the avatar's movements and speech synthesis technology to present the responses and questions to the user visually and audibly. The input of this step is the transmitted data, and the output is the display and audio.

[0229] Step 7:

[0230] The user responds to the responses and questions presented through the terminal. Specifically, the user makes a new statement, which is then sent back to the server. The input of this step is the user's response, and the output is new voice data.

[0231] The above is the processing steps and specific operational flow of the system that realizes this application example. This series of processes is expected to improve the user experience by enabling smooth customer service in online meetings and virtual stores.

[0232] 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.

[0233] This invention is a system for making online conferences proceed smoothly and lively, and provides more human-like interactions by combining an emotion engine that recognizes user emotions. The system mainly includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, a means for displaying and playing back the generated responses and questions, and a means for emotion analysis.

[0234] The specific program processing will be explained below.

[0235] Program processing overview

[0236] 1. Server: Conference preparation and connection

[0237] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[0238] The server checks the users' connections and ensures that everyone has joined.

[0239] 2. Server: Collecting and analyzing voice data

[0240] The server collects voice data from each user in real time.

[0241] The server converts the voice data into text data using a voice analysis engine and analyzes the content of the speech using natural language processing (NLP) technology.

[0242] 3. Server: Applying sentiment analysis methods

[0243] The server applies emotion analysis means based on the collected user voice and video data to recognize the user's emotional state.

[0244] The server adjusts responses and questions using the emotion data obtained by the emotion analysis means.

[0245] 4. Server: Generates backchannels and questions

[0246] The server generates appropriate responses and questions based on the analyzed speech content and emotional data. For example, if the user is nervous, the AI ​​will generate a response such as "Please relax. I'm speaking clearly."

[0247] Additionally, if the AI ​​detects that the user is confused, it will generate questions such as, "Could you please explain this in more detail?"

[0248] 5. Device: Display and playback

[0249] The user's device displays the responses and questions sent from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0250] 6. Server: Conference progress management

[0251] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0252] 7. Server: Meeting summary creation

[0253] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0254] The server summarizes the main points of the meeting and generates a summary comment, which the user's terminal displays on its screen and plays back audibly.

[0255] 8. Server: End the conference

[0256] The server confirms that the conference has ended and sends an end notice to all users, terminates the connections of all users, and saves the conference log data.

[0257] Example 1: Meeting initiation and emotion recognition

[0258] 1. When a user joins a conference, the server uses emotion analysis means to recognize the user's initial emotional state.

[0259] 2. The user begins to introduce themselves, and the system detects the user's nervousness.

[0260] 3. The server uses AI to generate a response such as "Relax, this is a good start."

[0261] 4. The user's device displays the backchannel and plays it back aloud.

[0262] Example 2: Facilitation and emotional response

[0263] 1. The server detects that the discussion is stagnating and uses sentiment analysis to recognize that a particular user is confused.

[0264] 2. The server generates a question for a specific user, such as "Could you please be more specific about this problem?"

[0265] 3. The user's device displays the question on the screen and plays it aloud.

[0266] 4. Users answer questions, getting the discussion moving again.

[0267] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[0268] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[0269] The processing flow will be explained below.

[0270] Program processing flow

[0271] Step 1:

[0272] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[0273] Step 2:

[0274] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[0275] Step 3:

[0276] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[0277] Step 4:

[0278] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[0279] Step 5:

[0280] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[0281] Step 6:

[0282] The server applies emotion analysis means based on the collected audio and video data to recognize the user's emotional state.

[0283] Step 7:

[0284] The server uses the emotional data obtained by the emotion analysis means to adjust the content of the interjections and questions. For example, if the user is nervous, the AI ​​will generate an interjection such as "Please relax, I'm speaking clearly." If the user is confused, it will generate a question such as "Could you explain this point in more detail?"

[0285] Step 8:

[0286] The server sends the generated responses and questions to each user's device.

[0287] Step 9:

[0288] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0289] Step 10:

[0290] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0291] Step 11:

[0292] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[0293] Step 12:

[0294] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0295] Step 13:

[0296] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[0297] Step 14:

[0298] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[0299] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[0300] Example 2

[0301] 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."

[0302] In online meetings, it is necessary to encourage participants to speak up, prevent discussions from stalling, and ensure the meeting progresses smoothly and actively. However, current online meeting systems have limitations in their ability to recognize stalls in speech and participants' emotional states and respond appropriately, resulting in reduced meeting efficiency. The objective of this invention is to solve these problems and provide more effective, human-like interactions.

[0303] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for monitoring the progress of the conference, a means for collecting voice data from participants in real time, a means for analyzing the voice data and converting it into text data, a means for analyzing the content of remarks based on the text data, a means for analyzing the emotions of the participants, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions. This allows the conference to proceed smoothly and enables more lively discussions by recognizing the emotional states of the participants and taking appropriate measures. The "means for monitoring the progress of the conference" refers to a technology for monitoring the progress of the conference in real time and grasping the progress of the discussion.

[0304] The "means for collecting voice data from participants in real time" is a technology for capturing and collecting the voices of participants in real time during a conference.

[0305] "Means for analyzing voice data and converting it into text data" refers to technology for analyzing collected voice data and converting it into text data.

[0306] "Means for analyzing speech content based on text data" refers to a technology that uses converted text data to analyze the content of speech and use the results to help progress the meeting.

[0307] "Means for analyzing participants' emotions" refers to technology for recognizing and analyzing participants' emotional states based on collected audio and video data.

[0308] "Means for generating appropriate responses and questions based on analysis results" refers to technology for generating responses and questions appropriate to the progress of a meeting based on the results of analyzing the content of statements and the emotional state of participants.

[0309] The "means for displaying and playing back generated responses and questions" refers to technology for visually displaying generated responses and questions and playing them back as audio. This invention is a system for encouraging participants to speak up in online meetings and promoting smooth and lively discussions. This system has the ability to recognize the user's emotional state and provide appropriate responses and questions.

[0310] The specific details of the system and the program processing are explained below.

[0311] System configuration

[0312] This system consists of the following main means:

[0313] A means of monitoring the progress of a meeting

[0314] A means of collecting real-time audio data from participants

[0315] A means of analyzing voice data and converting it into text data

[0316] A method for analyzing speech content based on text data

[0317] A means of analyzing participants' emotions

[0318] A means of generating appropriate responses and questions based on the analysis results

[0319] A means to view and play generated responses and questions

[0320] Program processing overview

[0321] 1. Server: Conference preparation and connection

[0322] The server sends the meeting invitation link to the user at the specified date and time as an email or notification, for example, using a calendar management system.

[0323] The server checks the connections of all users and confirms that everyone has joined. It uses WebSocket to monitor the connection status in real time.

[0324] 2. Server: Collecting and analyzing voice data

[0325] The server collects voice data from each user in real time, which is streamed using WebRTC technology.

[0326] The server converts the voice data into text data using a voice analysis engine, for example, a voice recognition API.

[0327] The text data is analyzed using natural language processing (NLP) techniques, which can be natural language understanding APIs.

[0328] 3. Server: Applying sentiment analysis methods

[0329] The server uses an emotion analysis engine to recognize the user's emotional state based on the collected audio and video data. For example, it uses an image analysis API for emotion analysis.

[0330] The server analyzes the recognized emotion data and uses it to generate appropriate responses and questions.

[0331] 4. Server: Generates backchannels and questions

[0332] The server uses a generative AI model to generate appropriate responses and questions based on the analyzed speech content and emotion data. The generative AI model uses a natural language generation API.

[0333] For example, if the user is nervous, a back-channel response such as "Relax, I'm speaking clearly" can be generated.

[0334] 5. Device: Display and playback

[0335] The user's terminal displays the responses and questions sent from the server on the screen.

[0336] The device plays back responses and questions aloud along with the animation of the AI ​​character, using a speech synthesis API as its speech synthesis engine.

[0337] 6. Server: Conference progress management

[0338] The server continuously monitors the progress of the conference and detects when the discussion stagnates.

[0339] When a stall is identified, generate appropriate questions for specific users, such as "Could you explain this in more detail?"

[0340] 7. Server: Meeting summary creation

[0341] The server monitors the end time of the meeting and notifies everyone when the end time is approaching.

[0342] The server summarizes the main points of the meeting and generates summary comments using a generative AI model, then sends the results to the user's device.

[0343] 8. Server: End the conference

[0344] The server confirms that the time for the conference to end has come and sends an end notice to all users.

[0345] All users are terminated and the log data of the conference is saved, which can be saved in cloud storage, for example.

[0346] Examples and prompts

[0347] Example 1: Meeting initiation and emotion recognition

[0348] The server uses emotion analysis means to recognize the initial emotional state of the user when the user joins the conference.

[0349] The user begins to introduce himself, and the server detects the user's nervousness.

[0350] The server uses AI to generate responses such as "Relax, this is a good start."

[0351] The device displays the backchannel and plays it back aloud.

[0352] Example 2: Facilitation and emotional response

[0353] The server detects that the discussion is stagnating and uses emotion analysis means to recognize that a particular user is confused.

[0354] The server generates a question for a particular user, such as "Could you please explain this problem in more detail?"

[0355] The device displays the question on the screen and plays it aloud.

[0356] Users can answer questions to get the discussion moving again.

[0357] Prompt Sentence Examples

[0358] "Meeting start prompt": "Analyze the user's current emotional state, determine if they are nervous, and generate appropriate responses."

[0359] "Prompt for when the discussion stalls": "Monitor the progress of the discussion and generate appropriate questions when the discussion stalls."

[0360] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[0361] The flow of the identification process in the second embodiment will be described with reference to Fig. 13. Step 1: Conference invitation and connection confirmation

[0362] The server sends a meeting invitation link to the user via email or notification at the specified date and time. The input is the user's email address and notification information, and the output is a notification that the invitation link will be sent. Specifically, the server manages the meeting schedule using a calendar management system and sends notifications at specific times.

[0363] The server confirms that the user clicked the link to join the conference. The input is the user's connection request, and the output is connection establishment information. WebSocket is used to monitor the connection status in real time.

[0364] Step 2: Collect and convert audio data

[0365] The server collects voice data from each user in real time. The input is the user's voice stream, and the output is the collected voice data. The voice data is streamed using WebRTC technology.

[0366] The server converts the collected voice data into text data using a voice analysis engine. The input is voice data, and the output is the converted text data. The specific process of converting voice data into text data is performed using a voice recognition API.

[0367] Step 3: Analyzing the speech

[0368] The server analyzes the text data using natural language processing (NLP) technology. The input is the converted text data, and the output is the analysis result of the speech content. Using a natural language understanding API, the intent and content of the speech are analyzed from the text data.

[0369] Step 4: Sentiment Analysis

[0370] The server uses an emotion analysis engine based on audio and video data to recognize the user's emotional state. The input is the user's audio and video data, and the output is the emotion analysis results. Specific processing to identify the emotional state is performed using image analysis APIs, etc.

[0371] Step 5: Generate responses and questions

[0372] The server uses a generative AI model to generate appropriate backchannels and questions based on the analyzed utterance content and emotional data. The input is the analysis results of the utterance content and the emotional analysis results, and the output is the generated backchannels and questions. Using the generative AI model, it is possible to generate a backchannel such as "Relax, that's a good start."

[0373] Step 6: View and play back responses and questions

[0374] The terminal displays the backchannels and questions sent from the server on the screen. The input is the backchannels and questions sent from the server, and the output is a text message displayed on the terminal.

[0375] The device plays back responses and questions aloud along with the animation of the AI ​​character. The input is voice data sent from the server, and the output is voice playback. The specific playback processing is performed using a speech synthesis API.

[0376] Step 7: Managing the meeting

[0377] The server continuously monitors the progress of the conference and detects when the discussion stagnates. The input is conference progress data, and the output is information on the detection of stagnation of the discussion.

[0378] The server generates an appropriate question for a specific user when the discussion stalls. The input is the stall detection information, and the output is the generated question. For example, it generates a question such as, "Could you please explain this point in more detail?"

[0379] Step 8: Create a meeting summary

[0380] The server monitors the end of the meeting and notifies everyone when the end is approaching. The input is the current time and the scheduled end time of the meeting, and the output is a notification that the end is imminent.

[0381] The server summarizes the main points of the meeting and generates summary comments. The input is a text log of the meeting content, and the output is summary comments. A generative AI model is used to create a meeting summary.

[0382] Step 9: End the meeting

[0383] The server confirms that the conference has ended and sends an end notification to all users. The input is the current time and the scheduled end time, and the output is the end notification.

[0384] The server terminates all user connections and saves the conference log data. The input is the participant connection information and conference log data, and the output is the saved log file. The server then processes the log to save it in cloud storage.

[0385] Through these steps, the system supports the progress of online meetings and provides an environment in which participants can be more actively involved.

[0386] (Application example 2)

[0387] 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."

[0388] Online meetings often lack interaction between participants, leading to stagnation in discussions. In particular, ignoring participants' emotional states can lead to tension and confusion, preventing meaningful dialogue. Furthermore, manually managing meeting progress is cumbersome and reduces efficiency. Meanwhile, virtual stores lack a system for providing appropriate product recommendations and support based on users' emotional states, making it difficult to increase purchasing motivation or streamline customer support.

[0389] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0390] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing participants' comments, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for recognizing the user's emotional state, and means for adjusting the responses and questions based on the user's emotional state. This provides an environment in online conferences where participants can speak more actively, improving the efficiency of the conference. Furthermore, in virtual stores, interactions tailored to the user's emotional state can be tailored to increase purchasing motivation and provide efficient customer support. The "means for monitoring the progress of the conference" is a tool for monitoring the progress of the conference in real time and proposing the next action at the appropriate time.

[0391] The "means for analyzing the content of comments" is a tool that converts participants' comments into text data and uses natural language processing technology to understand the content.

[0392] The "means for generating responses and questions" is a system for automatically generating appropriate responses and follow-up questions based on the analyzed content of statements.

[0393] The "means for displaying and playing" is a system for displaying the generated responses and questions on a user interface and playing them back aloud if necessary.

[0394] The "means for recognizing emotional states" is a system for analyzing a user's voice and video data and determining the user's emotional state in real time.

[0395] The "means for adjusting responses and questions based on emotional state" refers to a system for appropriately adjusting the content and tone of generated responses and questions in accordance with the recognized emotional state of the user. The present invention is a system for recognizing the emotional state of participants or users in online meetings or virtual stores and providing appropriate responses and questions based on that state. This system includes the following means.

[0396] 1. System program generation

[0397] This system is configured using the following hardware and software:

[0398] Sentiment analysis engine: Uses Amazon Rekognition and Google Cloud Speech-to-Text to analyze audio and video data.

[0399] Natural Language Processing Engine: Uses OpenAI GPT-4 to analyze what participants and users say and generate appropriate responses and questions.

[0400] User interface: We will use Unity to build interfaces that run on smartphones and head-mounted displays (HMDs).

[0401] Data collection: Real-time audio and video streaming is performed using WebRTC.

[0402] 2. Program processing explanation

[0403] a. Data Collection and Connections

[0404] When a user accesses a virtual store or online conference, the server uses WebRTC to collect real-time audio and video data. The user's device then sends this data to the server, where analysis begins.

[0405] b. Speech and emotion analysis

[0406] The server converts the collected voice data into text data using Google Cloud Speech-to-Text, and analyzes the video data in real time using Amazon Rekognition to detect the user's emotional state.

[0407] c. Natural Language Processing and Generation

[0408] The analyzed text data undergoes natural language processing using OpenAI GPT-4, which automatically generates appropriate responses and questions based on the content of the speech. For example, if the user does not understand a question, the system generates a question such as, "Can you explain more about the features of this product?"

[0409] d. Display / playback

[0410] The generated responses and questions are displayed in a Unity-based user interface, and the voice comments are played back using a TTS engine, providing the user with both visual and auditory feedback.

[0411] 3. Specific Examples

[0412] Example 1: Prompts for when the user is confused

[0413] It seems like the user doesn't understand the information on the product page. Suggest, "Can you explain more about this product's features?"

[0414] Example 2: Prompts for when the user is anxious

[0415] It seems like the user is hesitant to make a purchase. Suggest that you can find other products that fit your needs and budget.

[0416] effect

[0417] This system will improve the efficiency of online meetings by providing an environment in which participants can speak more actively. In addition, in virtual stores, it will enable interactions that are tailored to the user's emotional state, which will increase purchasing motivation and provide more efficient customer support.

[0418] The flow of the specific processing in Application Example 2 will be explained with reference to Figure 14. Step 1:

[0419] When a user accesses a virtual store or online conference, the server uses WebRTC to collect audio and video data in real time. This data is sent from the user's device to the server. The input is audio and video data, and the output is that this data is streamed to the server. Specifically, the moment the user accesses the server, WebRTC establishes a connection and starts streaming data.

[0420] Step 2:

[0421] To analyze the collected voice data, the server converts it into text data using the Google Cloud Speech-to-Text API. The input is voice data, and the output is text data. Once the text data is obtained, the server proceeds to the next step. Specifically, the server sends the voice data to the API, receives the returned text, and saves it in storage.

[0422] Step 3:

[0423] The server analyzes the collected video data using Amazon Rekognition to recognize the user's emotional state. The input is video data and the output is the user's emotional state. Specifically, the server sends the video data to the API, receives the emotional data returned as the analysis result, and stores it.

[0424] Step 4:

[0425] The server analyzes the text data and emotion data using the OpenAI GPT-4 model and generates appropriate responses and questions based on the content of the speech. The input is text data and emotion data, and the output is the generated response or question. Specifically, the server inputs the text data and emotion data as prompts for the generative AI model and obtains the generated sentence.

[0426] Step 5:

[0427] The server sends the generated responses and questions to the user interface. The input is the text of the responses and questions, and the output is the text displayed on the user interface. Specifically, the server sends the responses and questions in text format and displays them on the user's terminal.

[0428] Step 6:

[0429] The device displays the received responses and questions on the screen and plays them aloud if necessary. The input is text data from the server, and the output is visual and audio information presented to the user. Specifically, the text data is sent to the display component, and then played back as audio using the audio playback engine.

[0430] Step 7:

[0431] The server monitors the user's responses and continues to generate additional questions and responses as needed. The input is the user's new audio and video data, and the output is newly generated responses and questions. Specifically, it receives new data in real time via WebRTC and processes it again.

[0432] 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.

[0433] 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.

[0434] 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.

[0435] [Second embodiment]

[0436] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0437] 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.

[0438] 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).

[0439] 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.

[0440] 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.

[0441] 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).

[0442] 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.

[0443] 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.

[0444] 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.

[0445] 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.

[0446] 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.

[0447] 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."

[0448] The present invention is a system for making online conferences proceed smoothly and lively. The system includes, as its main means, a means for monitoring the progress of the conference, a means for analyzing the content of participants' remarks, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[0449] The specific program processing will be explained below.

[0450] Program processing overview

[0451] 1. Server: Conference preparation and connection

[0452] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[0453] The server checks the connections of the users and starts the conference after confirming that everyone has joined.

[0454] 2. Server: Collecting and analyzing voice data

[0455] The server collects voice data from each user in real time.

[0456] The server analyzes the voice data, converts it into text data, and analyzes the content of the speech using natural language processing (NLP) technology.

[0457] 3. Server: Generates responses and questions

[0458] The server generates appropriate responses and questions based on the analyzed content of the speech.

[0459] For example, if there is silence after a statement, the AI ​​will generate a response such as, "I see, that's interesting."

[0460] Also, when the discussion stagnates, it generates questions such as, "Mr. / Ms. XX, what do you think about this?"

[0461] 4. Device: Display and playback

[0462] The user's terminal displays the responses and questions sent from the server on the screen.

[0463] The device provides visual and auditory feedback to the user, accompanied by animations of AI characters and voice output.

[0464] Example 1: Starting a meeting and responding

[0465] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[0466] 2. The user begins to introduce themselves and the system detects periods of silence.

[0467] 3. The server uses AI to generate a response such as "That's great."

[0468] 4. The user's device displays the backchannel and plays it back aloud.

[0469] Example 2: Facilitation support

[0470] 1. The server detects that the discussion is stalled.

[0471] 2. The server generates a question for a particular user, such as "What is your opinion on this issue?"

[0472] 3. The user's device displays the question on the screen and plays it aloud.

[0473] 4. Users answer questions, getting the discussion moving again.

[0474] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving the efficiency of meetings. Even when there are few participants to speak or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and discussion to proceed in a manner that is in line with the purpose.

[0475] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[0476] The processing flow will be explained below.

[0477] Program processing flow

[0478] Step 1:

[0479] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[0480] Step 2:

[0481] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[0482] Step 3:

[0483] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[0484] Step 4:

[0485] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[0486] Step 5:

[0487] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[0488] Step 6:

[0489] The server generates appropriate responses and questions based on the analysis results. For example, if the server detects silence, the AI ​​will generate a response such as "I see, that's interesting."

[0490] Step 7:

[0491] The server sends the generated responses and questions to each user's device.

[0492] Step 8:

[0493] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0494] Step 9:

[0495] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0496] Step 10:

[0497] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[0498] Step 11:

[0499] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0500] Step 12:

[0501] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[0502] Step 13:

[0503] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[0504] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[0505] Example 1

[0506] 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."

[0507] Online meetings require smooth communication between participants, but in reality, progress often stalls. For example, periods of silence and stalled discussions are major factors that reduce meeting efficiency. Furthermore, if some participants do not actively speak up, there is a lack of appropriate encouragement, which can lead to a lack of depth in the overall discussion. It is necessary to resolve these issues, facilitate smooth online meetings, and provide an environment in which all participants can actively participate.

[0508] 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.

[0509] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology, means for analyzing the text data using natural language processing technology, means for generating appropriate reactions and queries using a generative AI model, and means for converting the generated text into voice data using speech synthesis technology. This makes it possible to automatically detect periods of silence and stagnation in the discussion, and to generate, display, and play back appropriate backchannels and questions. This facilitates the progress of the conference and provides a conference environment in which all participants are actively involved.

[0510] "Means for monitoring the progress of meetings" refers to technology for monitoring the overall progress of online meetings and detecting periods of silence or stagnation in discussions.

[0511] The "means for analyzing the content of participants' statements" refers to a technology that converts the content of each participant's statements into text data based on the collected voice data, and analyzes the text.

[0512] The "means of generating appropriate responses and questions based on the analysis results" refers to a technology that automatically generates responses and questions based on analyzed text data to facilitate the smooth progress of a meeting.

[0513] "Means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology" refers to technology for collecting voices during a meeting in real time and instantly converting them into text using automatic speech recognition technology.

[0514] "Means for analyzing text data using natural language processing technology" refers to technology that uses natural language processing technology to analyze converted text data and understand its content and intent.

[0515] "Means for generating appropriate reactions and queries using a generative AI model" refers to a technology that uses a generative AI model to generate natural responses and questions based on text data.

[0516] "Means for converting text generated using speech synthesis technology into speech data" refers to a technology that uses speech synthesis technology to convert generated text into speech data that provides visual and auditory feedback to the user.

[0517] The present invention provides a system for making online conferences proceed smoothly and lively. The system includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[0518] System configuration

[0519] Hardware and software used

[0520] Hardware: Servers (high-performance central processing units (CPUs), memory devices (RAM), large-capacity storage), user devices (personal computers (PCs), tablets, smartphones)

[0521] Software: natural language processing engines (e.g., general-purpose cloud natural language processing APIs or commercial natural language processing engines), automatic speech recognition engines (e.g., cloud-based speech recognition services), speech synthesis engines (e.g., cloud-based text-to-speech services), generative AI models (e.g., general-purpose chatbot models)

[0522] Program processing

[0523] 1. Server: Prepares the meeting and sends the meeting invitation link to users via email or notification at the specified date and time. The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[0524] 2. Server: When a meeting starts, the WebRTC protocol is used to collect each user's voice in real time, and the collected voice data is converted to text data using a cloud-based automatic speech recognition service.

[0525] 3. Server: The converted text data is sent to a natural language processing engine, where its content is analyzed. Based on this analyzed data, the meaning of what the participants said is understood.

[0526] 4. Server: Uses a generative AI model to generate appropriate responses and questions from the analyzed data. For example, if there is silence, it generates a response such as "I see, that's interesting," and if the discussion stagnates, it generates a question such as "What is your opinion on this issue?" An example of a prompt is, "If the user finishes speaking and there is silence, generate an appropriate response."

[0527] 5. Server: The generated text of the interjections and questions is sent to a speech synthesis engine, where it is converted into audio data. A cloud-based text-to-speech service is used to output the generated text as audio.

[0528] 6. Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio. The device uses HTML and JavaScript to display the text in the interface and the Web Audio API to play the audio.

[0529] Specific examples

[0530] Example 1: Starting a meeting and responding

[0531] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[0532] 2. The user begins to introduce themselves and the system detects periods of silence.

[0533] 3. The server uses AI to generate a response such as "That's great."

[0534] 4. The user's device displays the response and plays it aloud.

[0535] Example 2: Facilitation support

[0536] 1. The server detects that the discussion is stalled.

[0537] 2. The server generates a question for a specific user, such as "What is your opinion on this issue?"

[0538] 3. The user's device displays the question on the screen and plays it aloud.

[0539] 4. Users answer questions, getting the discussion moving again.

[0540] As described above, this invention can provide an environment in which participants can speak more actively in online meetings, thereby improving the efficiency of meetings. Even when there are few participants or the discussion stagnates, AI can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is in line with the purpose.

[0541] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0542] Step 1:

[0543] Server: Prepares the meeting and sends the meeting invitation link to the user at the specified date and time.

[0544] What happens: The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[0545] Input: A list of pre-registered user email addresses and / or notification system addresses.

[0546] Output: An email or notification containing the meeting invite link.

[0547] Step 2:

[0548] Users: Click the invitation link they receive to join the meeting.

[0549] Specific operation: The user clicks on the link in the email and enters the conference room via a web browser or dedicated app.

[0550] Input: The invitation link the user received.

[0551] Output: Connection status information to the conference platform.

[0552] Step 3:

[0553] Server: Checks user connection status in real time, confirms that everyone is participating, and starts the meeting.

[0554] What it does: The server monitors the user's online status via a WebSocket or HTTP connection.

[0555] Input: Connection log of the conference platform.

[0556] Output: Trigger to start the conference.

[0557] Step 4:

[0558] Server: Collects real-time audio streams from each user during the conference.

[0559] What happens: The server receives the audio data using the WebRTC protocol.

[0560] Input: The audio stream during the conference.

[0561] Output: Collected audio data.

[0562] Step 5:

[0563] Server: The collected voice data is converted into text data using automatic speech recognition (ASR) technology.

[0564] What it does: The server calls a cloud-based automatic speech recognition service to convert the audio data into text.

[0565] Input: Audio data.

[0566] Output: Text data.

[0567] Step 6:

[0568] Server: Analyzes the converted text data using natural language processing (NLP) technology to understand what is being said.

[0569] Specific operation: The server uses a natural language processing engine to analyze the content and intent of the text data.

[0570] Input: Text data.

[0571] Output: Analysis results (understanding of what was said).

[0572] Step 7:

[0573] Server: Uses generative AI models to generate appropriate reactions and queries based on the analysis results.

[0574] Specific operation: The server inputs a prompt sentence into the generative AI model and generates appropriate responses or questions.

[0575] Input: Analysis results and prompt statement.

[0576] Output: Generated text (backchannel or question).

[0577] Step 8:

[0578] Server: The generated text is passed to a speech synthesis engine and converted into voice data.

[0579] What it does: The server uses a cloud-based text-to-speech service to convert the generated text into audio.

[0580] Input: The generated text.

[0581] Output: Audio data.

[0582] Step 9:

[0583] Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio.

[0584] What it does: The device uses HTML and JavaScript to display text and the Web Audio API to play audio.

[0585] Input: Audio and text data sent from the server.

[0586] Output: Screen display, audio playback.

[0587] Step 10:

[0588] User: The user answers the displayed questions by voice, and the voice data is collected and analyzed by the server again.

[0589] Specific operation: The user speaks through the device's microphone, and the voice is sent to the server.

[0590] Input: The user's spoken response.

[0591] Output: New audio data is sent to the server.

[0592] (Application example 1)

[0593] 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."

[0594] Conventional online conference systems analyze participants' comments and generate appropriate responses and questions to facilitate smooth progress in the meeting. However, they have limitations in generating, displaying, and playing back responses in real time. Furthermore, when dealing with customers in virtual stores, there are cases where the conversation with the customer is interrupted or an appropriate response is not given. This can lead to a decrease in customer satisfaction and the loss of business opportunities.

[0595] 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.

[0596] In this invention, the server includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, and a means for generating appropriate responses and questions based on the analysis results. In addition, the server includes a means for monitoring interactions with customers in the virtual space and generating responses and questions at appropriate times, and a means for visually and audibly displaying and playing back the generated responses and questions through avatars. This not only facilitates the progress of online conferences, but also enables more effective customer service in virtual stores.

[0597] "Meeting proceedings" refers to the process by which participants come together to discuss and negotiate a specific topic or theme.

[0598] "Monitoring means" are methods or tools that allow a system to constantly monitor and collect data on specific conditions or events.

[0599] "Means for analyzing speech content" refers to methods and tools for collecting participants' speech and understanding and analyzing its content using technologies such as natural language processing.

[0600] "Means for generating appropriate responses or questions based on the analysis results" refers to methods or tools for automatically generating appropriate responses or questions based on the analyzed utterance content.

[0601] "Means for displaying and playing generated responses and questions" means methods and tools for visually and audibly presenting generated responses and questions to the user.

[0602] "Virtual customer interaction" refers to the act of interacting and communicating with customers in virtual reality or other digital platforms.

[0603] An "avatar" is a digital character that functions as a user's alter ego in a virtual space.

[0604] "Displaying and playing visually and audibly" means providing information to the user both visually and audibly by not only displaying the generated responses and questions on the screen but also playing them as audio.

[0605] A "system" is a collection of integrated devices and software that combines multiple components to achieve specific functions or services.

[0606] This invention is a system for facilitating customer service in online conferences and virtual stores. This system includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for monitoring interactions with customers in a virtual space, means for generating responses and questions at appropriate times, and means for visually and audibly displaying and playing back the generated responses and questions through an avatar.

[0607] Program processing overview

[0608] The server manages meetings and customer service in virtual stores in the following steps: First, the server collects voice data in real time and converts it into text data. Next, it uses natural language processing technology to analyze what is being said and generates appropriate responses and questions based on the analysis results. The generated responses and questions are sent to the user's device and displayed and played back visually and audibly through the avatar.

[0609] Hardware and software used

[0610] Hardware:

[0611] Smartphone

[0612] Head-mounted display (HMD)

[0613] server

[0614] software:

[0615] Python

[0616] Speech Recognition Library

[0617] Natural language processing library (spaCy)

[0618] Generative AI model (OpenAI GPT-3)

[0619] Program processing

[0620] The server uses a speech recognition library to collect voice data and convert it into text data. It then uses a natural language processing library to analyze the text data and generate appropriate responses and questions based on the content. This generation process uses a generative AI model such as OpenAI GPT-3.

[0621] The generated responses and questions are sent to the user's device, where they are displayed visually and audibly through an avatar. For example, if there is silence after a user makes a statement, the system generates a response such as "I see, that's interesting." If the discussion reaches a deadlock, the system generates a question such as "What do you think about this?"

[0622] Examples and prompts

[0623] Example 1: Starting a meeting and responding

[0624] The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting. When the user begins to introduce themselves, if the system detects a period of silence, the server uses AI to generate a backchannel such as "That's great." The user's device displays the backchannel and plays it back aloud.

[0625] Example 2: Facilitation support

[0626] The server detects when the discussion has stalled and generates a question for a specific user, such as "What is your opinion on this issue?" The user's device displays the question on the screen and plays it back aloud. When the user answers the question, the discussion resumes its smooth flow.

[0627] Examples of prompt statements

[0628] "The customer is experiencing periods of silence. Please generate an appropriate response."

[0629] Based on this prompt, the generative AI model generates appropriate responses and questions, which are then provided to the user through the system, making customer interactions in online meetings and virtual stores smoother and more effective.

[0630] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0631] Step 1:

[0632] The server collects voice data using a voice recognition library. Specifically, it records the user's speech in real time and obtains the voice data. The input of this step is the user's raw voice, and the output is voice data.

[0633] Step 2:

[0634] The server converts the collected voice data into text data. Specifically, it analyzes the voice data using a speech recognition library (SpeechRecognition) and converts the content into text format. The input of this step is voice data, and the output is the converted text data.

[0635] Step 3:

[0636] The server analyzes the text data using a natural language processing library (e.g., spaCy). Specifically, it analyzes the context and meaning of the text data and identifies situations such as periods of silence and stalled discussions. The input for this step is the text data, and the output is the analysis results.

[0637] Step 4:

[0638] The server generates appropriate responses and questions based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-3) to generate responses and questions corresponding to the analysis results. The input of this step is the analysis results, and the output is the generated responses and questions.

[0639] Step 5:

[0640] The server sends the generated responses and questions to the user's device. Specifically, it sends the responses and questions to the device via data communication. The input to this step is the generated responses and questions, and the output is the sent data.

[0641] Step 6:

[0642] The device displays the transmitted responses and questions through an avatar and plays them back as audio. Specifically, it uses the avatar's movements and speech synthesis technology to present the responses and questions to the user visually and audibly. The input of this step is the transmitted data, and the output is the display and audio.

[0643] Step 7:

[0644] The user responds to the responses and questions presented through the terminal. Specifically, the user makes a new statement, which is then sent back to the server. The input of this step is the user's response, and the output is new voice data.

[0645] The above is the processing steps and specific operational flow of the system that realizes this application example. This series of processes is expected to improve the user experience by enabling smooth customer service in online meetings and virtual stores.

[0646] 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.

[0647] This invention is a system for making online conferences proceed smoothly and lively, and provides more human-like interactions by combining an emotion engine that recognizes user emotions. The system mainly includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, a means for displaying and playing back the generated responses and questions, and a means for emotion analysis.

[0648] The specific program processing will be explained below.

[0649] Program processing overview

[0650] 1. Server: Conference preparation and connection

[0651] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[0652] The server checks the users' connections and ensures that everyone has joined.

[0653] 2. Server: Collecting and analyzing voice data

[0654] The server collects voice data from each user in real time.

[0655] The server converts the voice data into text data using a voice analysis engine and analyzes the content of the speech using natural language processing (NLP) technology.

[0656] 3. Server: Applying sentiment analysis methods

[0657] The server applies emotion analysis means based on the collected user voice and video data to recognize the user's emotional state.

[0658] The server adjusts responses and questions using the emotion data obtained by the emotion analysis means.

[0659] 4. Server: Generates backchannels and questions

[0660] The server generates appropriate responses and questions based on the analyzed speech content and emotional data. For example, if the user is nervous, the AI ​​will generate a response such as "Please relax. I'm speaking clearly."

[0661] Additionally, if the AI ​​detects that the user is confused, it will generate questions such as, "Could you please explain this in more detail?"

[0662] 5. Device: Display and playback

[0663] The user's device displays the responses and questions sent from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0664] 6. Server: Conference progress management

[0665] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0666] 7. Server: Meeting summary creation

[0667] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0668] The server summarizes the main points of the meeting and generates a summary comment, which the user's terminal displays on its screen and plays back audibly.

[0669] 8. Server: End the conference

[0670] The server confirms that the conference has ended and sends an end notice to all users, terminates the connections of all users, and saves the conference log data.

[0671] Example 1: Meeting initiation and emotion recognition

[0672] 1. When a user joins a conference, the server uses emotion analysis means to recognize the user's initial emotional state.

[0673] 2. The user begins to introduce themselves, and the system detects the user's nervousness.

[0674] 3. The server uses AI to generate a response such as "Relax, this is a good start."

[0675] 4. The user's device displays the backchannel and plays it back aloud.

[0676] Example 2: Facilitation and emotional response

[0677] 1. The server detects that the discussion is stagnating and uses sentiment analysis to recognize that a particular user is confused.

[0678] 2. The server generates a question for a specific user, such as "Could you please be more specific about this problem?"

[0679] 3. The user's device displays the question on the screen and plays it aloud.

[0680] 4. Users answer questions, getting the discussion moving again.

[0681] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[0682] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[0683] The processing flow will be explained below.

[0684] Program processing flow

[0685] Step 1:

[0686] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[0687] Step 2:

[0688] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[0689] Step 3:

[0690] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[0691] Step 4:

[0692] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[0693] Step 5:

[0694] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[0695] Step 6:

[0696] The server applies emotion analysis means based on the collected audio and video data to recognize the user's emotional state.

[0697] Step 7:

[0698] The server uses the emotional data obtained by the emotion analysis means to adjust the content of the interjections and questions. For example, if the user is nervous, the AI ​​will generate an interjection such as "Please relax, I'm speaking clearly." If the user is confused, it will generate a question such as "Could you explain this point in more detail?"

[0699] Step 8:

[0700] The server sends the generated responses and questions to each user's device.

[0701] Step 9:

[0702] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0703] Step 10:

[0704] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0705] Step 11:

[0706] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[0707] Step 12:

[0708] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0709] Step 13:

[0710] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[0711] Step 14:

[0712] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[0713] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[0714] Example 2

[0715] 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."

[0716] In online meetings, it is necessary to encourage participants to speak up, prevent discussions from stalling, and ensure the meeting progresses smoothly and actively. However, current online meeting systems have limitations in their ability to recognize stalls in speech and participants' emotional states and respond appropriately, resulting in reduced meeting efficiency. The objective of this invention is to solve these problems and provide more effective, human-like interactions.

[0717] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for monitoring the progress of the conference, a means for collecting voice data from participants in real time, a means for analyzing the voice data and converting it into text data, a means for analyzing the content of remarks based on the text data, a means for analyzing the emotions of the participants, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions. This allows the conference to proceed smoothly and enables more lively discussions by recognizing the emotional states of the participants and taking appropriate measures. The "means for monitoring the progress of the conference" refers to a technology for monitoring the progress of the conference in real time and grasping the progress of the discussion.

[0718] The "means for collecting voice data from participants in real time" is a technology for capturing and collecting the voices of participants in real time during a conference.

[0719] "Means for analyzing voice data and converting it into text data" refers to technology for analyzing collected voice data and converting it into text data.

[0720] "Means for analyzing speech content based on text data" refers to a technology that uses converted text data to analyze the content of speech and use the results to help progress the meeting.

[0721] "Means for analyzing participants' emotions" refers to technology for recognizing and analyzing participants' emotional states based on collected audio and video data.

[0722] "Means for generating appropriate responses and questions based on analysis results" refers to technology for generating responses and questions appropriate to the progress of a meeting based on the results of analyzing the content of statements and the emotional state of participants.

[0723] The "means for displaying and playing back generated responses and questions" refers to technology for visually displaying generated responses and questions and playing them back as audio. This invention is a system for encouraging participants to speak up in online meetings and promoting smooth and lively discussions. This system has the ability to recognize the user's emotional state and provide appropriate responses and questions.

[0724] The specific details of the system and the program processing are explained below.

[0725] System configuration

[0726] This system consists of the following main means:

[0727] A means of monitoring the progress of a meeting

[0728] A means of collecting real-time audio data from participants

[0729] A means of analyzing voice data and converting it into text data

[0730] A method for analyzing speech content based on text data

[0731] A means of analyzing participants' emotions

[0732] A means of generating appropriate responses and questions based on the analysis results

[0733] A means to view and play generated responses and questions

[0734] Program processing overview

[0735] 1. Server: Conference preparation and connection

[0736] The server sends the meeting invitation link to the user at the specified date and time as an email or notification, for example, using a calendar management system.

[0737] The server checks the connections of all users and confirms that everyone has joined. It uses WebSocket to monitor the connection status in real time.

[0738] 2. Server: Collecting and analyzing voice data

[0739] The server collects voice data from each user in real time, which is streamed using WebRTC technology.

[0740] The server converts the voice data into text data using a voice analysis engine, for example, a voice recognition API.

[0741] The text data is analyzed using natural language processing (NLP) techniques, which can be natural language understanding APIs.

[0742] 3. Server: Applying sentiment analysis methods

[0743] The server uses an emotion analysis engine to recognize the user's emotional state based on the collected audio and video data. For example, it uses an image analysis API for emotion analysis.

[0744] The server analyzes the recognized emotion data and uses it to generate appropriate responses and questions.

[0745] 4. Server: Generates backchannels and questions

[0746] The server uses a generative AI model to generate appropriate responses and questions based on the analyzed speech content and emotion data. The generative AI model uses a natural language generation API.

[0747] For example, if the user is nervous, a back-channel response such as "Relax, I'm speaking clearly" can be generated.

[0748] 5. Device: Display and playback

[0749] The user's terminal displays the responses and questions sent from the server on the screen.

[0750] The device plays back responses and questions aloud along with the animation of the AI ​​character, using a speech synthesis API as its speech synthesis engine.

[0751] 6. Server: Conference progress management

[0752] The server continuously monitors the progress of the conference and detects when the discussion stagnates.

[0753] When a stall is identified, generate appropriate questions for specific users, such as "Could you explain this in more detail?"

[0754] 7. Server: Meeting summary creation

[0755] The server monitors the end time of the meeting and notifies everyone when the end time is approaching.

[0756] The server summarizes the main points of the meeting and generates summary comments using a generative AI model, then sends the results to the user's device.

[0757] 8. Server: End the conference

[0758] The server confirms that the time for the conference to end has come and sends an end notice to all users.

[0759] All users are terminated and the log data of the conference is saved, which can be saved in cloud storage, for example.

[0760] Examples and prompts

[0761] Example 1: Meeting initiation and emotion recognition

[0762] The server uses emotion analysis means to recognize the initial emotional state of the user when the user joins the conference.

[0763] The user begins to introduce himself, and the server detects the user's nervousness.

[0764] The server uses AI to generate responses such as "Relax, this is a good start."

[0765] The device displays the backchannel and plays it back aloud.

[0766] Example 2: Facilitation and emotional response

[0767] The server detects that the discussion is stagnating and uses emotion analysis means to recognize that a particular user is confused.

[0768] The server generates a question for a particular user, such as "Could you please explain this problem in more detail?"

[0769] The device displays the question on the screen and plays it aloud.

[0770] Users can answer questions to get the discussion moving again.

[0771] Prompt Sentence Examples

[0772] "Meeting start prompt": "Analyze the user's current emotional state, determine if they are nervous, and generate appropriate responses."

[0773] "Prompt for when the discussion stalls": "Monitor the progress of the discussion and generate appropriate questions when the discussion stalls."

[0774] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[0775] The flow of the identification process in the second embodiment will be described with reference to Fig. 13. Step 1: Conference invitation and connection confirmation

[0776] The server sends a meeting invitation link to the user via email or notification at the specified date and time. The input is the user's email address and notification information, and the output is a notification that the invitation link will be sent. Specifically, the server manages the meeting schedule using a calendar management system and sends notifications at specific times.

[0777] The server confirms that the user clicked the link to join the conference. The input is the user's connection request, and the output is connection establishment information. WebSocket is used to monitor the connection status in real time.

[0778] Step 2: Collect and convert audio data

[0779] The server collects voice data from each user in real time. The input is the user's voice stream, and the output is the collected voice data. The voice data is streamed using WebRTC technology.

[0780] The server converts the collected voice data into text data using a voice analysis engine. The input is voice data, and the output is the converted text data. The specific process of converting voice data into text data is performed using a voice recognition API.

[0781] Step 3: Analyzing the speech

[0782] The server analyzes the text data using natural language processing (NLP) technology. The input is the converted text data, and the output is the analysis result of the speech content. Using a natural language understanding API, the intent and content of the speech are analyzed from the text data.

[0783] Step 4: Sentiment Analysis

[0784] The server uses an emotion analysis engine based on audio and video data to recognize the user's emotional state. The input is the user's audio and video data, and the output is the emotion analysis results. Specific processing to identify the emotional state is performed using image analysis APIs, etc.

[0785] Step 5: Generate responses and questions

[0786] The server uses a generative AI model to generate appropriate backchannels and questions based on the analyzed utterance content and emotional data. The input is the analysis results of the utterance content and the emotional analysis results, and the output is the generated backchannels and questions. Using the generative AI model, it is possible to generate a backchannel such as "Relax, that's a good start."

[0787] Step 6: View and play back responses and questions

[0788] The terminal displays the backchannels and questions sent from the server on the screen. The input is the backchannels and questions sent from the server, and the output is a text message displayed on the terminal.

[0789] The device plays back responses and questions aloud along with the animation of the AI ​​character. The input is voice data sent from the server, and the output is voice playback. The specific playback processing is performed using a speech synthesis API.

[0790] Step 7: Managing the meeting

[0791] The server continuously monitors the progress of the conference and detects when the discussion stagnates. The input is conference progress data, and the output is information on the detection of stagnation of the discussion.

[0792] The server generates an appropriate question for a specific user when the discussion stalls. The input is the stall detection information, and the output is the generated question. For example, it generates a question such as, "Could you please explain this point in more detail?"

[0793] Step 8: Create a meeting summary

[0794] The server monitors the end of the meeting and notifies everyone when the end is approaching. The input is the current time and the scheduled end time of the meeting, and the output is a notification that the end is imminent.

[0795] The server summarizes the main points of the meeting and generates summary comments. The input is a text log of the meeting content, and the output is summary comments. A generative AI model is used to create a meeting summary.

[0796] Step 9: End the meeting

[0797] The server confirms that the conference has ended and sends an end notification to all users. The input is the current time and the scheduled end time, and the output is the end notification.

[0798] The server terminates all user connections and saves the conference log data. The input is the participant connection information and conference log data, and the output is the saved log file. The server then processes the log to save it in cloud storage.

[0799] Through these steps, the system supports the progress of online meetings and provides an environment in which participants can be more actively involved.

[0800] (Application example 2)

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

[0802] Online meetings often lack interaction between participants, leading to stagnation in discussions. In particular, ignoring participants' emotional states can lead to tension and confusion, preventing meaningful dialogue. Furthermore, manually managing meeting progress is cumbersome and reduces efficiency. Meanwhile, virtual stores lack a system for providing appropriate product recommendations and support based on users' emotional states, making it difficult to increase purchasing motivation or streamline customer support.

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

[0804] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing participants' comments, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for recognizing the user's emotional state, and means for adjusting the responses and questions based on the user's emotional state. This provides an environment in online conferences where participants can speak more actively, improving the efficiency of the conference. Furthermore, in virtual stores, interactions tailored to the user's emotional state can be tailored to increase purchasing motivation and provide efficient customer support. The "means for monitoring the progress of the conference" is a tool for monitoring the progress of the conference in real time and proposing the next action at the appropriate time.

[0805] The "means for analyzing the content of comments" is a tool that converts participants' comments into text data and uses natural language processing technology to understand the content.

[0806] The "means for generating responses and questions" is a system for automatically generating appropriate responses and follow-up questions based on the analyzed content of statements.

[0807] The "means for displaying and playing" is a system for displaying the generated responses and questions on a user interface and playing them back aloud if necessary.

[0808] The "means for recognizing emotional states" is a system for analyzing a user's voice and video data and determining the user's emotional state in real time.

[0809] The "means for adjusting responses and questions based on emotional state" refers to a system for appropriately adjusting the content and tone of generated responses and questions in accordance with the recognized emotional state of the user. The present invention is a system for recognizing the emotional state of participants or users in online meetings or virtual stores and providing appropriate responses and questions based on that state. This system includes the following means.

[0810] 1. System program generation

[0811] This system is configured using the following hardware and software:

[0812] Sentiment analysis engine: Uses Amazon Rekognition and Google Cloud Speech-to-Text to analyze audio and video data.

[0813] Natural Language Processing Engine: Uses OpenAI GPT-4 to analyze what participants and users say and generate appropriate responses and questions.

[0814] User interface: We will use Unity to build interfaces that run on smartphones and head-mounted displays (HMDs).

[0815] Data collection: Real-time audio and video streaming is performed using WebRTC.

[0816] 2. Program processing explanation

[0817] a. Data Collection and Connections

[0818] When a user accesses a virtual store or online conference, the server uses WebRTC to collect real-time audio and video data. The user's device then sends this data to the server, where analysis begins.

[0819] b. Speech and emotion analysis

[0820] The server converts the collected voice data into text data using Google Cloud Speech-to-Text, and analyzes the video data in real time using Amazon Rekognition to detect the user's emotional state.

[0821] c. Natural Language Processing and Generation

[0822] The analyzed text data undergoes natural language processing using OpenAI GPT-4, which automatically generates appropriate responses and questions based on the content of the speech. For example, if the user does not understand a question, the system generates a question such as, "Can you explain more about the features of this product?"

[0823] d. Display / playback

[0824] The generated responses and questions are displayed in a Unity-based user interface, and the voice comments are played back using a TTS engine, providing the user with both visual and auditory feedback.

[0825] 3. Specific Examples

[0826] Example 1: Prompts for when the user is confused

[0827] It seems like the user doesn't understand the information on the product page. Suggest, "Can you explain more about this product's features?"

[0828] Example 2: Prompts for when the user is anxious

[0829] It seems like the user is hesitant to make a purchase. Suggest that you can find other products that fit your needs and budget.

[0830] effect

[0831] This system will improve the efficiency of online meetings by providing an environment in which participants can speak more actively. In addition, in virtual stores, it will enable interactions that are tailored to the user's emotional state, which will increase purchasing motivation and provide more efficient customer support.

[0832] The flow of the specific processing in Application Example 2 will be explained with reference to Figure 14. Step 1:

[0833] When a user accesses a virtual store or online conference, the server uses WebRTC to collect audio and video data in real time. This data is sent from the user's device to the server. The input is audio and video data, and the output is that this data is streamed to the server. Specifically, the moment the user accesses the server, WebRTC establishes a connection and starts streaming data.

[0834] Step 2:

[0835] To analyze the collected voice data, the server converts it into text data using the Google Cloud Speech-to-Text API. The input is voice data, and the output is text data. Once the text data is obtained, the server proceeds to the next step. Specifically, the server sends the voice data to the API, receives the returned text, and saves it in storage.

[0836] Step 3:

[0837] The server analyzes the collected video data using Amazon Rekognition to recognize the user's emotional state. The input is video data and the output is the user's emotional state. Specifically, the server sends the video data to the API, receives the emotional data returned as the analysis result, and stores it.

[0838] Step 4:

[0839] The server analyzes the text data and emotion data using the OpenAI GPT-4 model and generates appropriate responses and questions based on the content of the speech. The input is text data and emotion data, and the output is the generated response or question. Specifically, the server inputs the text data and emotion data as prompts for the generative AI model and obtains the generated sentence.

[0840] Step 5:

[0841] The server sends the generated responses and questions to the user interface. The input is the text of the responses and questions, and the output is the text displayed on the user interface. Specifically, the server sends the responses and questions in text format and displays them on the user's terminal.

[0842] Step 6:

[0843] The device displays the received responses and questions on the screen and plays them aloud if necessary. The input is text data from the server, and the output is visual and audio information presented to the user. Specifically, the text data is sent to the display component, and then played back as audio using the audio playback engine.

[0844] Step 7:

[0845] The server monitors the user's responses and continues to generate additional questions and responses as needed. The input is the user's new audio and video data, and the output is newly generated responses and questions. Specifically, it receives new data in real time via WebRTC and processes it again.

[0846] 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.

[0847] 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.

[0848] 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.

[0849] [Third embodiment]

[0850] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0851] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0852] 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).

[0853] 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.

[0854] 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.

[0855] 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).

[0856] 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.

[0857] 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.

[0858] 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.

[0859] 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.

[0860] 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.

[0861] 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."

[0862] The present invention is a system for making online conferences proceed smoothly and lively. The system includes, as its main means, a means for monitoring the progress of the conference, a means for analyzing the content of participants' remarks, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[0863] The specific program processing will be explained below.

[0864] Program processing overview

[0865] 1. Server: Conference preparation and connection

[0866] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[0867] The server checks the connections of the users and starts the conference after confirming that everyone has joined.

[0868] 2. Server: Collecting and analyzing voice data

[0869] The server collects voice data from each user in real time.

[0870] The server analyzes the voice data, converts it into text data, and analyzes the content of the speech using natural language processing (NLP) technology.

[0871] 3. Server: Generates responses and questions

[0872] The server generates appropriate responses and questions based on the analyzed content of the speech.

[0873] For example, if there is silence after a statement, the AI ​​will generate a response such as, "I see, that's interesting."

[0874] Also, when the discussion stagnates, it generates questions such as, "Mr. / Ms. XX, what do you think about this?"

[0875] 4. Device: Display and playback

[0876] The user's terminal displays the responses and questions sent from the server on the screen.

[0877] The device provides visual and auditory feedback to the user, accompanied by animations of AI characters and voice output.

[0878] Example 1: Starting a meeting and responding

[0879] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[0880] 2. The user begins to introduce themselves and the system detects periods of silence.

[0881] 3. The server uses AI to generate a response such as "That's great."

[0882] 4. The user's device displays the backchannel and plays it back aloud.

[0883] Example 2: Facilitation support

[0884] 1. The server detects that the discussion is stalled.

[0885] 2. The server generates a question for a particular user, such as "What is your opinion on this issue?"

[0886] 3. The user's device displays the question on the screen and plays it aloud.

[0887] 4. Users answer questions, getting the discussion moving again.

[0888] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving the efficiency of meetings. Even when there are few participants to speak or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and discussion to proceed in a manner that is in line with the purpose.

[0889] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[0890] The processing flow will be explained below.

[0891] Program processing flow

[0892] Step 1:

[0893] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[0894] Step 2:

[0895] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[0896] Step 3:

[0897] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[0898] Step 4:

[0899] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[0900] Step 5:

[0901] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[0902] Step 6:

[0903] The server generates appropriate responses and questions based on the analysis results. For example, if the server detects silence, the AI ​​will generate a response such as "I see, that's interesting."

[0904] Step 7:

[0905] The server sends the generated responses and questions to each user's device.

[0906] Step 8:

[0907] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[0908] Step 9:

[0909] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[0910] Step 10:

[0911] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[0912] Step 11:

[0913] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[0914] Step 12:

[0915] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[0916] Step 13:

[0917] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[0918] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[0919] Example 1

[0920] 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."

[0921] Online meetings require smooth communication between participants, but in reality, progress often stalls. For example, periods of silence and stalled discussions are major factors that reduce meeting efficiency. Furthermore, if some participants do not actively speak up, there is a lack of appropriate encouragement, which can lead to a lack of depth in the overall discussion. It is necessary to resolve these issues, facilitate smooth online meetings, and provide an environment in which all participants can actively participate.

[0922] 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.

[0923] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology, means for analyzing the text data using natural language processing technology, means for generating appropriate reactions and queries using a generative AI model, and means for converting the generated text into voice data using speech synthesis technology. This makes it possible to automatically detect periods of silence and stagnation in the discussion, and to generate, display, and play back appropriate backchannels and questions. This facilitates the progress of the conference and provides a conference environment in which all participants are actively involved.

[0924] "Means for monitoring the progress of meetings" refers to technology for monitoring the overall progress of online meetings and detecting periods of silence or stagnation in discussions.

[0925] The "means for analyzing the content of participants' statements" refers to a technology that converts the content of each participant's statements into text data based on the collected voice data, and analyzes the text.

[0926] The "means of generating appropriate responses and questions based on the analysis results" refers to a technology that automatically generates responses and questions based on analyzed text data to facilitate the smooth progress of a meeting.

[0927] "Means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology" refers to technology for collecting voices during a meeting in real time and instantly converting them into text using automatic speech recognition technology.

[0928] "Means for analyzing text data using natural language processing technology" refers to technology that uses natural language processing technology to analyze converted text data and understand its content and intent.

[0929] "Means for generating appropriate reactions and queries using a generative AI model" refers to a technology that uses a generative AI model to generate natural responses and questions based on text data.

[0930] "Means for converting text generated using speech synthesis technology into speech data" refers to a technology that uses speech synthesis technology to convert generated text into speech data that provides visual and auditory feedback to the user.

[0931] The present invention provides a system for making online conferences proceed smoothly and lively. The system includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[0932] System configuration

[0933] Hardware and software used

[0934] Hardware: Servers (high-performance central processing units (CPUs), memory devices (RAM), large-capacity storage), user devices (personal computers (PCs), tablets, smartphones)

[0935] Software: natural language processing engines (e.g., general-purpose cloud natural language processing APIs or commercial natural language processing engines), automatic speech recognition engines (e.g., cloud-based speech recognition services), speech synthesis engines (e.g., cloud-based text-to-speech services), generative AI models (e.g., general-purpose chatbot models)

[0936] Program processing

[0937] 1. Server: Prepares the meeting and sends the meeting invitation link to users via email or notification at the specified date and time. The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[0938] 2. Server: When a meeting starts, the WebRTC protocol is used to collect each user's voice in real time, and the collected voice data is converted to text data using a cloud-based automatic speech recognition service.

[0939] 3. Server: The converted text data is sent to a natural language processing engine, where its content is analyzed. Based on this analyzed data, the meaning of what the participants said is understood.

[0940] 4. Server: Uses a generative AI model to generate appropriate responses and questions from the analyzed data. For example, if there is silence, it generates a response such as "I see, that's interesting," and if the discussion stagnates, it generates a question such as "What is your opinion on this issue?" An example of a prompt is, "If the user finishes speaking and there is silence, generate an appropriate response."

[0941] 5. Server: The generated text of the interjections and questions is sent to a speech synthesis engine, where it is converted into audio data. A cloud-based text-to-speech service is used to output the generated text as audio.

[0942] 6. Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio. The device uses HTML and JavaScript to display the text in the interface and the Web Audio API to play the audio.

[0943] Specific examples

[0944] Example 1: Starting a meeting and responding

[0945] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[0946] 2. The user begins to introduce themselves and the system detects periods of silence.

[0947] 3. The server uses AI to generate a response such as "That's great."

[0948] 4. The user's device displays the response and plays it aloud.

[0949] Example 2: Facilitation support

[0950] 1. The server detects that the discussion is stalled.

[0951] 2. The server generates a question for a specific user, such as "What is your opinion on this issue?"

[0952] 3. The user's device displays the question on the screen and plays it aloud.

[0953] 4. Users answer questions, getting the discussion moving again.

[0954] As described above, this invention can provide an environment in which participants can speak more actively in online meetings, thereby improving the efficiency of meetings. Even when there are few participants or the discussion stagnates, AI can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is in line with the purpose.

[0955] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0956] Step 1:

[0957] Server: Prepares the meeting and sends the meeting invitation link to the user at the specified date and time.

[0958] What happens: The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[0959] Input: A list of pre-registered user email addresses and / or notification system addresses.

[0960] Output: An email or notification containing the meeting invite link.

[0961] Step 2:

[0962] Users: Click the invitation link they receive to join the meeting.

[0963] Specific operation: The user clicks on the link in the email and enters the conference room via a web browser or dedicated app.

[0964] Input: The invitation link the user received.

[0965] Output: Connection status information to the conference platform.

[0966] Step 3:

[0967] Server: Checks user connection status in real time, confirms that everyone is participating, and starts the meeting.

[0968] What it does: The server monitors the user's online status via a WebSocket or HTTP connection.

[0969] Input: Connection log of the conference platform.

[0970] Output: Trigger to start the conference.

[0971] Step 4:

[0972] Server: Collects real-time audio streams from each user during the conference.

[0973] What happens: The server receives the audio data using the WebRTC protocol.

[0974] Input: The audio stream during the conference.

[0975] Output: Collected audio data.

[0976] Step 5:

[0977] Server: The collected voice data is converted into text data using automatic speech recognition (ASR) technology.

[0978] What it does: The server calls a cloud-based automatic speech recognition service to convert the audio data into text.

[0979] Input: Audio data.

[0980] Output: Text data.

[0981] Step 6:

[0982] Server: Analyzes the converted text data using natural language processing (NLP) technology to understand what is being said.

[0983] Specific operation: The server uses a natural language processing engine to analyze the content and intent of the text data.

[0984] Input: Text data.

[0985] Output: Analysis results (understanding of what was said).

[0986] Step 7:

[0987] Server: Uses generative AI models to generate appropriate reactions and queries based on the analysis results.

[0988] Specific operation: The server inputs a prompt sentence into the generative AI model and generates appropriate responses or questions.

[0989] Input: Analysis results and prompt statement.

[0990] Output: Generated text (backchannel or question).

[0991] Step 8:

[0992] Server: The generated text is passed to a speech synthesis engine and converted into voice data.

[0993] What it does: The server uses a cloud-based text-to-speech service to convert the generated text into audio.

[0994] Input: The generated text.

[0995] Output: Audio data.

[0996] Step 9:

[0997] Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio.

[0998] What it does: The device uses HTML and JavaScript to display text and the Web Audio API to play audio.

[0999] Input: Audio and text data sent from the server.

[1000] Output: Screen display, audio playback.

[1001] Step 10:

[1002] User: The user answers the displayed questions by voice, and the voice data is collected and analyzed by the server again.

[1003] Specific operation: The user speaks through the device's microphone, and the voice is sent to the server.

[1004] Input: The user's spoken response.

[1005] Output: New audio data is sent to the server.

[1006] (Application example 1)

[1007] 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."

[1008] Conventional online conference systems analyze participants' comments and generate appropriate responses and questions to facilitate smooth progress in the meeting. However, they have limitations in generating, displaying, and playing back responses in real time. Furthermore, when dealing with customers in virtual stores, there are cases where the conversation with the customer is interrupted or an appropriate response is not given. This can lead to a decrease in customer satisfaction and the loss of business opportunities.

[1009] 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.

[1010] In this invention, the server includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, and a means for generating appropriate responses and questions based on the analysis results. In addition, the server includes a means for monitoring interactions with customers in the virtual space and generating responses and questions at appropriate times, and a means for visually and audibly displaying and playing back the generated responses and questions through avatars. This not only facilitates the progress of online conferences, but also enables more effective customer service in virtual stores.

[1011] "Meeting proceedings" refers to the process by which participants come together to discuss and negotiate a specific topic or theme.

[1012] "Monitoring means" are methods or tools that allow a system to constantly monitor and collect data on specific conditions or events.

[1013] "Means for analyzing speech content" refers to methods and tools for collecting participants' speech and understanding and analyzing its content using technologies such as natural language processing.

[1014] "Means for generating appropriate responses or questions based on the analysis results" refers to methods or tools for automatically generating appropriate responses or questions based on the analyzed utterance content.

[1015] "Means for displaying and playing generated responses and questions" means methods and tools for visually and audibly presenting generated responses and questions to the user.

[1016] "Virtual customer interaction" refers to the act of interacting and communicating with customers in virtual reality or other digital platforms.

[1017] An "avatar" is a digital character that functions as a user's alter ego in a virtual space.

[1018] "Displaying and playing visually and audibly" means providing information to the user both visually and audibly by not only displaying the generated responses and questions on the screen but also playing them as audio.

[1019] A "system" is a collection of integrated devices and software that combines multiple components to achieve specific functions or services.

[1020] This invention is a system for facilitating customer service in online conferences and virtual stores. This system includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for monitoring interactions with customers in a virtual space, means for generating responses and questions at appropriate times, and means for visually and audibly displaying and playing back the generated responses and questions through an avatar.

[1021] Program processing overview

[1022] The server manages meetings and customer service in virtual stores in the following steps: First, the server collects voice data in real time and converts it into text data. Next, it uses natural language processing technology to analyze what is being said and generates appropriate responses and questions based on the analysis results. The generated responses and questions are sent to the user's device and displayed and played back visually and audibly through the avatar.

[1023] Hardware and software used

[1024] Hardware:

[1025] Smartphone

[1026] Head-mounted display (HMD)

[1027] server

[1028] software:

[1029] Python

[1030] Speech Recognition Library

[1031] Natural language processing library (spaCy)

[1032] Generative AI model (OpenAI GPT-3)

[1033] Program processing

[1034] The server uses a speech recognition library to collect voice data and convert it into text data. It then uses a natural language processing library to analyze the text data and generate appropriate responses and questions based on the content. This generation process uses a generative AI model such as OpenAI GPT-3.

[1035] The generated responses and questions are sent to the user's device, where they are displayed visually and audibly through an avatar. For example, if there is silence after a user makes a statement, the system generates a response such as "I see, that's interesting." If the discussion reaches a deadlock, the system generates a question such as "What do you think about this?"

[1036] Examples and prompts

[1037] Example 1: Starting a meeting and responding

[1038] The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting. When the user begins to introduce themselves, if the system detects a period of silence, the server uses AI to generate a backchannel such as "That's great." The user's device displays the backchannel and plays it back aloud.

[1039] Example 2: Facilitation support

[1040] The server detects when the discussion has stalled and generates a question for a specific user, such as "What is your opinion on this issue?" The user's device displays the question on the screen and plays it back aloud. When the user answers the question, the discussion resumes its smooth flow.

[1041] Examples of prompt statements

[1042] "The customer is experiencing periods of silence. Please generate an appropriate response."

[1043] Based on this prompt, the generative AI model generates appropriate responses and questions, which are then provided to the user through the system, making customer interactions in online meetings and virtual stores smoother and more effective.

[1044] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1045] Step 1:

[1046] The server collects voice data using a voice recognition library. Specifically, it records the user's speech in real time and obtains the voice data. The input of this step is the user's raw voice, and the output is voice data.

[1047] Step 2:

[1048] The server converts the collected voice data into text data. Specifically, it analyzes the voice data using a speech recognition library (SpeechRecognition) and converts the content into text format. The input of this step is voice data, and the output is the converted text data.

[1049] Step 3:

[1050] The server analyzes the text data using a natural language processing library (e.g., spaCy). Specifically, it analyzes the context and meaning of the text data and identifies situations such as periods of silence and stalled discussions. The input for this step is the text data, and the output is the analysis results.

[1051] Step 4:

[1052] The server generates appropriate responses and questions based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-3) to generate responses and questions corresponding to the analysis results. The input of this step is the analysis results, and the output is the generated responses and questions.

[1053] Step 5:

[1054] The server sends the generated responses and questions to the user's device. Specifically, it sends the responses and questions to the device via data communication. The input to this step is the generated responses and questions, and the output is the sent data.

[1055] Step 6:

[1056] The device displays the transmitted responses and questions through an avatar and plays them back as audio. Specifically, it uses the avatar's movements and speech synthesis technology to present the responses and questions to the user visually and audibly. The input of this step is the transmitted data, and the output is the display and audio.

[1057] Step 7:

[1058] The user responds to the responses and questions presented through the terminal. Specifically, the user makes a new statement, which is then sent back to the server. The input of this step is the user's response, and the output is new voice data.

[1059] The above is the processing steps and specific operational flow of the system that realizes this application example. This series of processes is expected to improve the user experience by enabling smooth customer service in online meetings and virtual stores.

[1060] 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.

[1061] This invention is a system for making online conferences proceed smoothly and lively, and provides more human-like interactions by combining an emotion engine that recognizes user emotions. The system mainly includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, a means for displaying and playing back the generated responses and questions, and a means for emotion analysis.

[1062] The specific program processing will be explained below.

[1063] Program processing overview

[1064] 1. Server: Conference preparation and connection

[1065] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[1066] The server checks the users' connections and ensures that everyone has joined.

[1067] 2. Server: Collecting and analyzing voice data

[1068] The server collects voice data from each user in real time.

[1069] The server converts the voice data into text data using a voice analysis engine and analyzes the content of the speech using natural language processing (NLP) technology.

[1070] 3. Server: Applying sentiment analysis methods

[1071] The server applies emotion analysis means based on the collected user voice and video data to recognize the user's emotional state.

[1072] The server adjusts responses and questions using the emotion data obtained by the emotion analysis means.

[1073] 4. Server: Generates backchannels and questions

[1074] The server generates appropriate responses and questions based on the analyzed speech content and emotional data. For example, if the user is nervous, the AI ​​will generate a response such as "Please relax. I'm speaking clearly."

[1075] Additionally, if the AI ​​detects that the user is confused, it will generate questions such as, "Could you please explain this in more detail?"

[1076] 5. Device: Display and playback

[1077] The user's device displays the responses and questions sent from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[1078] 6. Server: Conference progress management

[1079] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[1080] 7. Server: Meeting summary creation

[1081] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[1082] The server summarizes the main points of the meeting and generates a summary comment, which the user's terminal displays on its screen and plays back audibly.

[1083] 8. Server: End the conference

[1084] The server confirms that the conference has ended and sends an end notice to all users, terminates the connections of all users, and saves the conference log data.

[1085] Example 1: Meeting initiation and emotion recognition

[1086] 1. When a user joins a conference, the server uses emotion analysis means to recognize the user's initial emotional state.

[1087] 2. The user begins to introduce themselves, and the system detects the user's nervousness.

[1088] 3. The server uses AI to generate a response such as "Relax, this is a good start."

[1089] 4. The user's device displays the backchannel and plays it back aloud.

[1090] Example 2: Facilitation and emotional response

[1091] 1. The server detects that the discussion is stagnating and uses sentiment analysis to recognize that a particular user is confused.

[1092] 2. The server generates a question for a specific user, such as "Could you please be more specific about this problem?"

[1093] 3. The user's device displays the question on the screen and plays it aloud.

[1094] 4. Users answer questions, getting the discussion moving again.

[1095] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[1096] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[1097] The processing flow will be explained below.

[1098] Program processing flow

[1099] Step 1:

[1100] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[1101] Step 2:

[1102] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[1103] Step 3:

[1104] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[1105] Step 4:

[1106] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[1107] Step 5:

[1108] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[1109] Step 6:

[1110] The server applies emotion analysis means based on the collected audio and video data to recognize the user's emotional state.

[1111] Step 7:

[1112] The server uses the emotional data obtained by the emotion analysis means to adjust the content of the interjections and questions. For example, if the user is nervous, the AI ​​will generate an interjection such as "Please relax, I'm speaking clearly." If the user is confused, it will generate a question such as "Could you explain this point in more detail?"

[1113] Step 8:

[1114] The server sends the generated responses and questions to each user's device.

[1115] Step 9:

[1116] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[1117] Step 10:

[1118] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[1119] Step 11:

[1120] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[1121] Step 12:

[1122] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[1123] Step 13:

[1124] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[1125] Step 14:

[1126] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[1127] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[1128] Example 2

[1129] 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."

[1130] In online meetings, it is necessary to encourage participants to speak up, prevent discussions from stalling, and ensure the meeting progresses smoothly and actively. However, current online meeting systems have limitations in their ability to recognize stalls in speech and participants' emotional states and respond appropriately, resulting in reduced meeting efficiency. The objective of this invention is to solve these problems and provide more effective, human-like interactions.

[1131] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for monitoring the progress of the conference, a means for collecting voice data from participants in real time, a means for analyzing the voice data and converting it into text data, a means for analyzing the content of remarks based on the text data, a means for analyzing the emotions of the participants, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions. This allows the conference to proceed smoothly and enables more lively discussions by recognizing the emotional states of the participants and taking appropriate measures. The "means for monitoring the progress of the conference" refers to a technology for monitoring the progress of the conference in real time and grasping the progress of the discussion.

[1132] The "means for collecting voice data from participants in real time" is a technology for capturing and collecting the voices of participants in real time during a conference.

[1133] "Means for analyzing voice data and converting it into text data" refers to technology for analyzing collected voice data and converting it into text data.

[1134] "Means for analyzing speech content based on text data" refers to a technology that uses converted text data to analyze the content of speech and use the results to help progress the meeting.

[1135] "Means for analyzing participants' emotions" refers to technology for recognizing and analyzing participants' emotional states based on collected audio and video data.

[1136] "Means for generating appropriate responses and questions based on analysis results" refers to technology for generating responses and questions appropriate to the progress of a meeting based on the results of analyzing the content of statements and the emotional state of participants.

[1137] The "means for displaying and playing back generated responses and questions" refers to technology for visually displaying generated responses and questions and playing them back as audio. This invention is a system for encouraging participants to speak up in online meetings and promoting smooth and lively discussions. This system has the ability to recognize the user's emotional state and provide appropriate responses and questions.

[1138] The specific details of the system and the program processing are explained below.

[1139] System configuration

[1140] This system consists of the following main means:

[1141] A means of monitoring the progress of a meeting

[1142] A means of collecting real-time audio data from participants

[1143] A means of analyzing voice data and converting it into text data

[1144] A method for analyzing speech content based on text data

[1145] A means of analyzing participants' emotions

[1146] A means of generating appropriate responses and questions based on the analysis results

[1147] A means to view and play generated responses and questions

[1148] Program processing overview

[1149] 1. Server: Conference preparation and connection

[1150] The server sends the meeting invitation link to the user at the specified date and time as an email or notification, for example, using a calendar management system.

[1151] The server checks the connections of all users and confirms that everyone has joined. It uses WebSocket to monitor the connection status in real time.

[1152] 2. Server: Collecting and analyzing voice data

[1153] The server collects voice data from each user in real time, which is streamed using WebRTC technology.

[1154] The server converts the voice data into text data using a voice analysis engine, for example, a voice recognition API.

[1155] The text data is analyzed using natural language processing (NLP) techniques, which can be natural language understanding APIs.

[1156] 3. Server: Applying sentiment analysis methods

[1157] The server uses an emotion analysis engine to recognize the user's emotional state based on the collected audio and video data. For example, it uses an image analysis API for emotion analysis.

[1158] The server analyzes the recognized emotion data and uses it to generate appropriate responses and questions.

[1159] 4. Server: Generates backchannels and questions

[1160] The server uses a generative AI model to generate appropriate responses and questions based on the analyzed speech content and emotion data. The generative AI model uses a natural language generation API.

[1161] For example, if the user is nervous, a back-channel response such as "Relax, I'm speaking clearly" can be generated.

[1162] 5. Device: Display and playback

[1163] The user's terminal displays the responses and questions sent from the server on the screen.

[1164] The device plays back responses and questions aloud along with the animation of the AI ​​character, using a speech synthesis API as its speech synthesis engine.

[1165] 6. Server: Conference progress management

[1166] The server continuously monitors the progress of the conference and detects when the discussion stagnates.

[1167] When a stall is identified, generate appropriate questions for specific users, such as "Could you explain this in more detail?"

[1168] 7. Server: Meeting summary creation

[1169] The server monitors the end time of the meeting and notifies everyone when the end time is approaching.

[1170] The server summarizes the main points of the meeting and generates summary comments using a generative AI model, then sends the results to the user's device.

[1171] 8. Server: End the conference

[1172] The server confirms that the time for the conference to end has come and sends an end notice to all users.

[1173] All users are terminated and the log data of the conference is saved, which can be saved in cloud storage, for example.

[1174] Examples and prompts

[1175] Example 1: Meeting initiation and emotion recognition

[1176] The server uses emotion analysis means to recognize the initial emotional state of the user when the user joins the conference.

[1177] The user begins to introduce himself, and the server detects the user's nervousness.

[1178] The server uses AI to generate responses such as "Relax, this is a good start."

[1179] The device displays the backchannel and plays it back aloud.

[1180] Example 2: Facilitation and emotional response

[1181] The server detects that the discussion is stagnating and uses emotion analysis means to recognize that a particular user is confused.

[1182] The server generates a question for a particular user, such as "Could you please explain this problem in more detail?"

[1183] The device displays the question on the screen and plays it aloud.

[1184] Users can answer questions to get the discussion moving again.

[1185] Prompt Sentence Examples

[1186] "Meeting start prompt": "Analyze the user's current emotional state, determine if they are nervous, and generate appropriate responses."

[1187] "Prompt for when the discussion stalls": "Monitor the progress of the discussion and generate appropriate questions when the discussion stalls."

[1188] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[1189] The flow of the identification process in the second embodiment will be described with reference to Fig. 13. Step 1: Conference invitation and connection confirmation

[1190] The server sends a meeting invitation link to the user via email or notification at the specified date and time. The input is the user's email address and notification information, and the output is a notification that the invitation link will be sent. Specifically, the server manages the meeting schedule using a calendar management system and sends notifications at specific times.

[1191] The server confirms that the user clicked the link to join the conference. The input is the user's connection request, and the output is connection establishment information. WebSocket is used to monitor the connection status in real time.

[1192] Step 2: Collect and convert audio data

[1193] The server collects voice data from each user in real time. The input is the user's voice stream, and the output is the collected voice data. The voice data is streamed using WebRTC technology.

[1194] The server converts the collected voice data into text data using a voice analysis engine. The input is voice data, and the output is the converted text data. The specific process of converting voice data into text data is performed using a voice recognition API.

[1195] Step 3: Analyzing the speech

[1196] The server analyzes the text data using natural language processing (NLP) technology. The input is the converted text data, and the output is the analysis result of the speech content. Using a natural language understanding API, the intent and content of the speech are analyzed from the text data.

[1197] Step 4: Sentiment Analysis

[1198] The server uses an emotion analysis engine based on audio and video data to recognize the user's emotional state. The input is the user's audio and video data, and the output is the emotion analysis results. Specific processing to identify the emotional state is performed using image analysis APIs, etc.

[1199] Step 5: Generate responses and questions

[1200] The server uses a generative AI model to generate appropriate backchannels and questions based on the analyzed utterance content and emotional data. The input is the analysis results of the utterance content and the emotional analysis results, and the output is the generated backchannels and questions. Using the generative AI model, it is possible to generate a backchannel such as "Relax, that's a good start."

[1201] Step 6: View and play back responses and questions

[1202] The terminal displays the backchannels and questions sent from the server on the screen. The input is the backchannels and questions sent from the server, and the output is a text message displayed on the terminal.

[1203] The device plays back responses and questions aloud along with the animation of the AI ​​character. The input is voice data sent from the server, and the output is voice playback. The specific playback processing is performed using a speech synthesis API.

[1204] Step 7: Managing the meeting

[1205] The server continuously monitors the progress of the conference and detects when the discussion stagnates. The input is conference progress data, and the output is information on the detection of stagnation of the discussion.

[1206] The server generates an appropriate question for a specific user when the discussion stalls. The input is the stall detection information, and the output is the generated question. For example, it generates a question such as, "Could you please explain this point in more detail?"

[1207] Step 8: Create a meeting summary

[1208] The server monitors the end of the meeting and notifies everyone when the end is approaching. The input is the current time and the scheduled end time of the meeting, and the output is a notification that the end is imminent.

[1209] The server summarizes the main points of the meeting and generates summary comments. The input is a text log of the meeting content, and the output is summary comments. A generative AI model is used to create a meeting summary.

[1210] Step 9: End the meeting

[1211] The server confirms that the conference has ended and sends an end notification to all users. The input is the current time and the scheduled end time, and the output is the end notification.

[1212] The server terminates all user connections and saves the conference log data. The input is the participant connection information and conference log data, and the output is the saved log file. The server then processes the log to save it in cloud storage.

[1213] Through these steps, the system supports the progress of online meetings and provides an environment in which participants can be more actively involved.

[1214] (Application example 2)

[1215] 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."

[1216] Online meetings often lack interaction between participants, leading to stagnation in discussions. In particular, ignoring participants' emotional states can lead to tension and confusion, preventing meaningful dialogue. Furthermore, manually managing meeting progress is cumbersome and reduces efficiency. Meanwhile, virtual stores lack a system for providing appropriate product recommendations and support based on users' emotional states, making it difficult to increase purchasing motivation or streamline customer support.

[1217] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1218] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing participants' comments, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for recognizing the user's emotional state, and means for adjusting the responses and questions based on the user's emotional state. This provides an environment in online conferences where participants can speak more actively, improving the efficiency of the conference. Furthermore, in virtual stores, interactions tailored to the user's emotional state can be tailored to increase purchasing motivation and provide efficient customer support. The "means for monitoring the progress of the conference" is a tool for monitoring the progress of the conference in real time and proposing the next action at the appropriate time.

[1219] The "means for analyzing the content of comments" is a tool that converts participants' comments into text data and uses natural language processing technology to understand the content.

[1220] The "means for generating responses and questions" is a system for automatically generating appropriate responses and follow-up questions based on the analyzed content of statements.

[1221] The "means for displaying and playing" is a system for displaying the generated responses and questions on a user interface and playing them back aloud if necessary.

[1222] The "means for recognizing emotional states" is a system for analyzing a user's voice and video data and determining the user's emotional state in real time.

[1223] The "means for adjusting responses and questions based on emotional state" refers to a system for appropriately adjusting the content and tone of generated responses and questions in accordance with the recognized emotional state of the user. The present invention is a system for recognizing the emotional state of participants or users in online meetings or virtual stores and providing appropriate responses and questions based on that state. This system includes the following means.

[1224] 1. System program generation

[1225] This system is configured using the following hardware and software:

[1226] Sentiment analysis engine: Uses Amazon Rekognition and Google Cloud Speech-to-Text to analyze audio and video data.

[1227] Natural Language Processing Engine: Uses OpenAI GPT-4 to analyze what participants and users say and generate appropriate responses and questions.

[1228] User interface: We will use Unity to build interfaces that run on smartphones and head-mounted displays (HMDs).

[1229] Data collection: Real-time audio and video streaming is performed using WebRTC.

[1230] 2. Program processing explanation

[1231] a. Data Collection and Connections

[1232] When a user accesses a virtual store or online conference, the server uses WebRTC to collect real-time audio and video data. The user's device then sends this data to the server, where analysis begins.

[1233] b. Speech and emotion analysis

[1234] The server converts the collected voice data into text data using Google Cloud Speech-to-Text, and analyzes the video data in real time using Amazon Rekognition to detect the user's emotional state.

[1235] c. Natural Language Processing and Generation

[1236] The analyzed text data undergoes natural language processing using OpenAI GPT-4, which automatically generates appropriate responses and questions based on the content of the speech. For example, if the user does not understand a question, the system generates a question such as, "Can you explain more about the features of this product?"

[1237] d. Display / playback

[1238] The generated responses and questions are displayed in a Unity-based user interface, and the voice comments are played back using a TTS engine, providing the user with both visual and auditory feedback.

[1239] 3. Specific Examples

[1240] Example 1: Prompts for when the user is confused

[1241] It seems like the user doesn't understand the information on the product page. Suggest, "Can you explain more about this product's features?"

[1242] Example 2: Prompts for when the user is anxious

[1243] It seems like the user is hesitant to make a purchase. Suggest that you can find other products that fit your needs and budget.

[1244] effect

[1245] This system will improve the efficiency of online meetings by providing an environment in which participants can speak more actively. In addition, in virtual stores, it will enable interactions that are tailored to the user's emotional state, which will increase purchasing motivation and provide more efficient customer support.

[1246] The flow of the specific processing in Application Example 2 will be explained with reference to Figure 14. Step 1:

[1247] When a user accesses a virtual store or online conference, the server uses WebRTC to collect audio and video data in real time. This data is sent from the user's device to the server. The input is audio and video data, and the output is that this data is streamed to the server. Specifically, the moment the user accesses the server, WebRTC establishes a connection and starts streaming data.

[1248] Step 2:

[1249] To analyze the collected voice data, the server converts it into text data using the Google Cloud Speech-to-Text API. The input is voice data, and the output is text data. Once the text data is obtained, the server proceeds to the next step. Specifically, the server sends the voice data to the API, receives the returned text, and saves it in storage.

[1250] Step 3:

[1251] The server analyzes the collected video data using Amazon Rekognition to recognize the user's emotional state. The input is video data and the output is the user's emotional state. Specifically, the server sends the video data to the API, receives the emotional data returned as the analysis result, and stores it.

[1252] Step 4:

[1253] The server analyzes the text data and emotion data using the OpenAI GPT-4 model and generates appropriate responses and questions based on the content of the speech. The input is text data and emotion data, and the output is the generated response or question. Specifically, the server inputs the text data and emotion data as prompts for the generative AI model and obtains the generated sentence.

[1254] Step 5:

[1255] The server sends the generated responses and questions to the user interface. The input is the text of the responses and questions, and the output is the text displayed on the user interface. Specifically, the server sends the responses and questions in text format and displays them on the user's terminal.

[1256] Step 6:

[1257] The device displays the received responses and questions on the screen and plays them aloud if necessary. The input is text data from the server, and the output is visual and audio information presented to the user. Specifically, the text data is sent to the display component, and then played back as audio using the audio playback engine.

[1258] Step 7:

[1259] The server monitors the user's responses and continues to generate additional questions and responses as needed. The input is the user's new audio and video data, and the output is newly generated responses and questions. Specifically, it receives new data in real time via WebRTC and processes it again.

[1260] 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.

[1261] 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.

[1262] 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.

[1263] [Fourth embodiment]

[1264] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1265] 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.

[1266] 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).

[1267] 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.

[1268] 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.

[1269] 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).

[1270] 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.

[1271] 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.

[1272] 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.

[1273] 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.

[1274] 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.

[1275] 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.

[1276] 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."

[1277] The present invention is a system for making online conferences proceed smoothly and lively. The system includes, as its main means, a means for monitoring the progress of the conference, a means for analyzing the content of participants' remarks, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[1278] The specific program processing will be explained below.

[1279] Program processing overview

[1280] 1. Server: Conference preparation and connection

[1281] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[1282] The server checks the connections of the users and starts the conference after confirming that everyone has joined.

[1283] 2. Server: Collecting and analyzing voice data

[1284] The server collects voice data from each user in real time.

[1285] The server analyzes the voice data, converts it into text data, and analyzes the content of the speech using natural language processing (NLP) technology.

[1286] 3. Server: Generates responses and questions

[1287] The server generates appropriate responses and questions based on the analyzed content of the speech.

[1288] For example, if there is silence after a statement, the AI ​​will generate a response such as, "I see, that's interesting."

[1289] Also, when the discussion stagnates, it generates questions such as, "Mr. / Ms. XX, what do you think about this?"

[1290] 4. Device: Display and playback

[1291] The user's terminal displays the responses and questions sent from the server on the screen.

[1292] The device provides visual and auditory feedback to the user, accompanied by animations of AI characters and voice output.

[1293] Example 1: Starting a meeting and responding

[1294] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[1295] 2. The user begins to introduce themselves and the system detects periods of silence.

[1296] 3. The server uses AI to generate a response such as "That's great."

[1297] 4. The user's device displays the backchannel and plays it back aloud.

[1298] Example 2: Facilitation support

[1299] 1. The server detects that the discussion is stalled.

[1300] 2. The server generates a question for a particular user, such as "What is your opinion on this issue?"

[1301] 3. The user's device displays the question on the screen and plays it aloud.

[1302] 4. Users answer questions, getting the discussion moving again.

[1303] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving the efficiency of meetings. Even when there are few participants to speak or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and discussion to proceed in a manner that is in line with the purpose.

[1304] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[1305] The processing flow will be explained below.

[1306] Program processing flow

[1307] Step 1:

[1308] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[1309] Step 2:

[1310] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[1311] Step 3:

[1312] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[1313] Step 4:

[1314] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[1315] Step 5:

[1316] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[1317] Step 6:

[1318] The server generates appropriate responses and questions based on the analysis results. For example, if the server detects silence, the AI ​​will generate a response such as "I see, that's interesting."

[1319] Step 7:

[1320] The server sends the generated responses and questions to each user's device.

[1321] Step 8:

[1322] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[1323] Step 9:

[1324] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[1325] Step 10:

[1326] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[1327] Step 11:

[1328] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[1329] Step 12:

[1330] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[1331] Step 13:

[1332] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[1333] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[1334] Example 1

[1335] 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."

[1336] Online meetings require smooth communication between participants, but in reality, progress often stalls. For example, periods of silence and stalled discussions are major factors that reduce meeting efficiency. Furthermore, if some participants do not actively speak up, there is a lack of appropriate encouragement, which can lead to a lack of depth in the overall discussion. It is necessary to resolve these issues, facilitate smooth online meetings, and provide an environment in which all participants can actively participate.

[1337] 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.

[1338] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology, means for analyzing the text data using natural language processing technology, means for generating appropriate reactions and queries using a generative AI model, and means for converting the generated text into voice data using speech synthesis technology. This makes it possible to automatically detect periods of silence and stagnation in the discussion, and to generate, display, and play back appropriate backchannels and questions. This facilitates the progress of the conference and provides a conference environment in which all participants are actively involved.

[1339] "Means for monitoring the progress of meetings" refers to technology for monitoring the overall progress of online meetings and detecting periods of silence or stagnation in discussions.

[1340] The "means for analyzing the content of participants' statements" refers to a technology that converts the content of each participant's statements into text data based on the collected voice data, and analyzes the text.

[1341] "Means for generating appropriate responses and questions based on analysis results" refers to technology that automatically generates responses and questions based on analyzed text data to facilitate the progress of meetings.

[1342] "Means for collecting user voice data in real time and converting it into text data using automatic speech recognition technology" refers to technology for collecting voices during a meeting in real time and instantly converting them into text using automatic speech recognition technology.

[1343] "Means for analyzing text data using natural language processing technology" refers to technology that uses natural language processing technology to analyze converted text data and understand its content and intent.

[1344] "Means for generating appropriate reactions and queries using a generative AI model" refers to a technology that uses a generative AI model to generate natural responses and questions based on text data.

[1345] "Means for converting text generated using speech synthesis technology into speech data" refers to a technology that uses speech synthesis technology to convert generated text into speech data that provides visual and auditory feedback to the user.

[1346] The present invention provides a system for making online conferences proceed smoothly and lively. The system includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions.

[1347] System configuration

[1348] Hardware and software used

[1349] Hardware: Servers (high-performance central processing units (CPUs), memory devices (RAM), large-capacity storage), user devices (personal computers (PCs), tablets, smartphones)

[1350] Software: natural language processing engines (e.g., general-purpose cloud natural language processing APIs or commercial natural language processing engines), automatic speech recognition engines (e.g., cloud-based speech recognition services), speech synthesis engines (e.g., cloud-based text-to-speech services), generative AI models (e.g., general-purpose chatbot models)

[1351] Program processing

[1352] 1. Server: Prepares the meeting and sends the meeting invitation link to users via email or notification at the specified date and time. The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[1353] 2. Server: When a meeting starts, the WebRTC protocol is used to collect each user's voice in real time, and the collected voice data is converted to text data using a cloud-based automatic speech recognition service.

[1354] 3. Server: The converted text data is sent to a natural language processing engine, where its content is analyzed. Based on this analyzed data, the meaning of what the participants said is understood.

[1355] 4. Server: Uses a generative AI model to generate appropriate responses and questions from the analyzed data. For example, if there is silence, it generates a response such as "I see, that's interesting," and if the discussion stagnates, it generates a question such as "What is your opinion on this issue?" An example of a prompt is, "If the user finishes speaking and there is silence, generate an appropriate response."

[1356] 5. Server: The generated text of the interjections and questions is sent to a speech synthesis engine, where it is converted into audio data. A cloud-based text-to-speech service is used to output the generated text as audio.

[1357] 6. Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio. The device uses HTML and JavaScript to display the text in the interface and the Web Audio API to play the audio.

[1358] Specific examples

[1359] Example 1: Starting a meeting and responding

[1360] 1. The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting.

[1361] 2. The user begins to introduce themselves and the system detects periods of silence.

[1362] 3. The server uses AI to generate a response such as "That's great."

[1363] 4. The user's device displays the response and plays it aloud.

[1364] Example 2: Facilitation support

[1365] 1. The server detects that the discussion is stalled.

[1366] 2. The server generates a question for a specific user, such as "What is your opinion on this issue?"

[1367] 3. The user's device displays the question on the screen and plays it aloud.

[1368] 4. Users answer questions, getting the discussion moving again.

[1369] As described above, this invention can provide an environment in which participants can speak more actively in online meetings, thereby improving the efficiency of meetings. Even when there are few participants or the discussion stagnates, AI can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is in line with the purpose.

[1370] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1371] Step 1:

[1372] Server: Prepares the meeting and sends the meeting invitation link to the user at the specified date and time.

[1373] What happens: The server sends emails using the SMTP protocol or sends notifications using a real-time notification system.

[1374] Input: A list of pre-registered user email addresses and / or notification system addresses.

[1375] Output: An email or notification containing the meeting invite link.

[1376] Step 2:

[1377] Users: Click the invitation link they receive to join the meeting.

[1378] Specific operation: The user clicks on the link in the email and enters the conference room via a web browser or dedicated app.

[1379] Input: The invitation link the user received.

[1380] Output: Connection status information to the conference platform.

[1381] Step 3:

[1382] Server: Checks user connection status in real time, confirms that everyone is participating, and starts the meeting.

[1383] What it does: The server monitors the user's online status via a WebSocket or HTTP connection.

[1384] Input: Connection log of the conference platform.

[1385] Output: Trigger to start the conference.

[1386] Step 4:

[1387] Server: Collects real-time audio streams from each user during the conference.

[1388] What happens: The server receives the audio data using the WebRTC protocol.

[1389] Input: The audio stream during the conference.

[1390] Output: Collected audio data.

[1391] Step 5:

[1392] Server: The collected voice data is converted into text data using automatic speech recognition (ASR) technology.

[1393] What it does: The server calls a cloud-based automatic speech recognition service to convert the audio data into text.

[1394] Input: Audio data.

[1395] Output: Text data.

[1396] Step 6:

[1397] Server: Analyzes the converted text data using natural language processing (NLP) technology to understand what is being said.

[1398] Specific operation: The server uses a natural language processing engine to analyze the content and intent of the text data.

[1399] Input: Text data.

[1400] Output: Analysis results (understanding of what was said).

[1401] Step 7:

[1402] Server: Uses generative AI models to generate appropriate reactions and queries based on the analysis results.

[1403] Specific operation: The server inputs a prompt sentence into the generative AI model and generates appropriate responses or questions.

[1404] Input: Analysis results and prompt statement.

[1405] Output: Generated text (backchannel or question).

[1406] Step 8:

[1407] Server: The generated text is passed to a speech synthesis engine and converted into voice data.

[1408] What it does: The server uses a cloud-based text-to-speech service to convert the generated text into audio.

[1409] Input: The generated text.

[1410] Output: Audio data.

[1411] Step 9:

[1412] Device: The user's device displays the responses and questions sent from the server on the screen and plays them back as audio.

[1413] What it does: The device uses HTML and JavaScript to display text and the Web Audio API to play audio.

[1414] Input: Audio and text data sent from the server.

[1415] Output: Screen display, audio playback.

[1416] Step 10:

[1417] User: The user answers the displayed questions by voice, and the voice data is collected and analyzed by the server again.

[1418] Specific operation: The user speaks through the device's microphone, and the voice is sent to the server.

[1419] Input: The user's spoken response.

[1420] Output: New audio data is sent to the server.

[1421] (Application example 1)

[1422] 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."

[1423] Conventional online conference systems analyze participants' comments and generate appropriate responses and questions to facilitate smooth progress in the meeting. However, they have limitations in generating, displaying, and playing back responses in real time. Furthermore, when dealing with customers in virtual stores, there are cases where the conversation with the customer is interrupted or an appropriate response is not given. This can lead to a decrease in customer satisfaction and the loss of business opportunities.

[1424] 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.

[1425] In this invention, the server includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, and a means for generating appropriate responses and questions based on the analysis results. In addition, the server includes a means for monitoring interactions with customers in the virtual space and generating responses and questions at appropriate times, and a means for visually and audibly displaying and playing back the generated responses and questions through avatars. This not only facilitates the progress of online conferences, but also enables more effective customer service in virtual stores.

[1426] "Meeting proceedings" refers to the process by which participants come together to discuss and negotiate a specific topic or theme.

[1427] "Monitoring means" are methods or tools that allow a system to constantly monitor and collect data on specific conditions or events.

[1428] "Means for analyzing speech content" refers to methods and tools for collecting participants' speech and understanding and analyzing its content using technologies such as natural language processing.

[1429] "Means for generating appropriate responses or questions based on the analysis results" refers to methods or tools for automatically generating appropriate responses or questions based on the analyzed utterance content.

[1430] "Means for displaying and playing generated responses and questions" means methods and tools for visually and audibly presenting generated responses and questions to the user.

[1431] "Virtual customer interaction" refers to the act of interacting and communicating with customers in virtual reality or other digital platforms.

[1432] An "avatar" is a digital character that functions as a user's alter ego in a virtual space.

[1433] "Displaying and playing visually and audibly" means providing information to the user both visually and audibly by not only displaying the generated responses and questions on the screen but also playing them as audio.

[1434] A "system" is a collection of integrated devices and software that combines multiple components to achieve specific functions or services.

[1435] This invention is a system for facilitating customer service in online conferences and virtual stores. This system includes means for monitoring the progress of the conference, means for analyzing the content of participants' remarks, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for monitoring interactions with customers in a virtual space, means for generating responses and questions at appropriate times, and means for visually and audibly displaying and playing back the generated responses and questions through an avatar.

[1436] Program processing overview

[1437] The server manages meetings and customer service in virtual stores in the following steps: First, the server collects voice data in real time and converts it into text data. Next, it uses natural language processing technology to analyze what is being said and generates appropriate responses and questions based on the analysis results. The generated responses and questions are sent to the user's device and displayed and played back visually and audibly through the avatar.

[1438] Hardware and software used

[1439] Hardware:

[1440] Smartphone

[1441] Head-mounted display (HMD)

[1442] server

[1443] software:

[1444] Python

[1445] Speech Recognition Library

[1446] Natural language processing library (spaCy)

[1447] Generative AI model (OpenAI GPT-3)

[1448] Program processing

[1449] The server uses a speech recognition library to collect voice data and convert it into text data. It then uses a natural language processing library to analyze the text data and generate appropriate responses and questions based on the content. This generation process uses a generative AI model such as OpenAI GPT-3.

[1450] The generated responses and questions are sent to the user's device, where they are displayed visually and audibly through an avatar. For example, if there is silence after a user makes a statement, the system generates a response such as "I see, that's interesting." If the discussion reaches a deadlock, the system generates a question such as "What do you think about this?"

[1451] Examples and prompts

[1452] Example 1: Starting a meeting and responding

[1453] The server sends a meeting invitation to the user at a scheduled time, and the user joins the meeting. When the user begins to introduce themselves, if the system detects a period of silence, the server uses AI to generate a backchannel such as "That's great." The user's device displays the backchannel and plays it back aloud.

[1454] Example 2: Facilitation support

[1455] The server detects when the discussion has stalled and generates a question for a specific user, such as "What is your opinion on this issue?" The user's device displays the question on the screen and plays it back aloud. When the user answers the question, the discussion resumes its smooth flow.

[1456] Examples of prompt statements

[1457] "The customer is experiencing periods of silence. Please generate an appropriate response."

[1458] Based on this prompt, the generative AI model generates appropriate responses and questions, which are then provided to the user through the system, making customer interactions in online meetings and virtual stores smoother and more effective.

[1459] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1460] Step 1:

[1461] The server collects voice data using a voice recognition library. Specifically, it records the user's speech in real time and obtains the voice data. The input of this step is the user's raw voice, and the output is voice data.

[1462] Step 2:

[1463] The server converts the collected voice data into text data. Specifically, it analyzes the voice data using a speech recognition library (SpeechRecognition) and converts the content into text format. The input of this step is voice data, and the output is the converted text data.

[1464] Step 3:

[1465] The server analyzes the text data using a natural language processing library (e.g., spaCy). Specifically, it analyzes the context and meaning of the text data and identifies situations such as periods of silence and stalled discussions. The input for this step is the text data, and the output is the analysis results.

[1466] Step 4:

[1467] The server generates appropriate responses and questions based on the analysis results. Specifically, it uses a generative AI model (e.g., OpenAI GPT-3) to generate responses and questions corresponding to the analysis results. The input of this step is the analysis results, and the output is the generated responses and questions.

[1468] Step 5:

[1469] The server sends the generated responses and questions to the user's device. Specifically, it sends the responses and questions to the device via data communication. The input to this step is the generated responses and questions, and the output is the sent data.

[1470] Step 6:

[1471] The device displays the transmitted responses and questions through an avatar and plays them back as audio. Specifically, it uses the avatar's movements and speech synthesis technology to present the responses and questions to the user visually and audibly. The input of this step is the transmitted data, and the output is the display and audio.

[1472] Step 7:

[1473] The user responds to the responses and questions presented through the terminal. Specifically, the user makes a new statement, which is then sent back to the server. The input of this step is the user's response, and the output is new voice data.

[1474] The above is the processing steps and specific operational flow of the system that realizes this application example. This series of processes is expected to improve the user experience by enabling smooth customer service in online meetings and virtual stores.

[1475] 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.

[1476] This invention is a system for making online conferences proceed smoothly and lively, and provides more human-like interactions by combining an emotion engine that recognizes user emotions. The system mainly includes a means for monitoring the progress of the conference, a means for analyzing the content of participants' comments, a means for generating appropriate responses and questions based on the analysis results, a means for displaying and playing back the generated responses and questions, and a means for emotion analysis.

[1477] The specific program processing will be explained below.

[1478] Program processing overview

[1479] 1. Server: Conference preparation and connection

[1480] The server will send a meeting invitation link to the user via email or notification at the specified date and time.

[1481] The server checks the users' connections and ensures that everyone has joined.

[1482] 2. Server: Collecting and analyzing voice data

[1483] The server collects voice data from each user in real time.

[1484] The server converts the voice data into text data using a voice analysis engine and analyzes the content of the speech using natural language processing (NLP) technology.

[1485] 3. Server: Applying sentiment analysis methods

[1486] The server applies emotion analysis means based on the collected user voice and video data to recognize the user's emotional state.

[1487] The server adjusts responses and questions using the emotion data obtained by the emotion analysis means.

[1488] 4. Server: Generates backchannels and questions

[1489] The server generates appropriate responses and questions based on the analyzed speech content and emotional data. For example, if the user is nervous, the AI ​​will generate a response such as "Please relax. I'm speaking clearly."

[1490] Additionally, if the AI ​​detects that the user is confused, it will generate questions such as, "Could you please explain this in more detail?"

[1491] 5. Device: Display and playback

[1492] The user's device displays the responses and questions sent from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[1493] 6. Server: Conference progress management

[1494] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[1495] 7. Server: Meeting summary creation

[1496] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[1497] The server summarizes the main points of the meeting and generates a summary comment, which the user's terminal displays on its screen and plays back audibly.

[1498] 8. Server: End the conference

[1499] The server confirms that the conference has ended and sends an end notice to all users, terminates the connections of all users, and saves the conference log data.

[1500] Example 1: Meeting initiation and emotion recognition

[1501] 1. When a user joins a conference, the server uses emotion analysis means to recognize the user's initial emotional state.

[1502] 2. The user begins to introduce themselves, and the system detects the user's nervousness.

[1503] 3. The server uses AI to generate a response such as "Relax, this is a good start."

[1504] 4. The user's device displays the backchannel and plays it back aloud.

[1505] Example 2: Facilitation and emotional response

[1506] 1. The server detects that the discussion is stagnating and uses sentiment analysis to recognize that a particular user is confused.

[1507] 2. The server generates a question for a specific user, such as "Could you please be more specific about this problem?"

[1508] 3. The user's device displays the question on the screen and plays it aloud.

[1509] 4. Users answer questions, getting the discussion moving again.

[1510] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[1511] The above is an embodiment of the present invention. According to this embodiment, the system can solve many problems in online meetings and realize more effective and lively meetings.

[1512] The processing flow will be explained below.

[1513] Program processing flow

[1514] Step 1:

[1515] The server will send a meeting invitation link to each user via email or notification at the specified date and time, and the server will confirm that each user has received it.

[1516] Step 2:

[1517] A user joins an online conference by clicking on a conference invitation link from the server, and the user's terminal streams the user's audio and video to the server.

[1518] Step 3:

[1519] The server checks which users are connected to the conference and verifies that everyone has joined. After this verification, the server starts the conference.

[1520] Step 4:

[1521] The server collects voice data from each user in real time and converts the voice data into text data using a voice analysis engine.

[1522] Step 5:

[1523] The server analyzes the text data using natural language processing (NLP) technology, which allows it to understand the content and tone of what is being said and grasp the progress of the meeting.

[1524] Step 6:

[1525] The server applies emotion analysis means based on the collected audio and video data to recognize the user's emotional state.

[1526] Step 7:

[1527] The server uses the emotional data obtained by the emotion analysis means to adjust the content of the interjections and questions. For example, if the user is nervous, the AI ​​will generate an interjection such as "Please relax, I'm speaking clearly." If the user is confused, it will generate a question such as "Could you explain this point in more detail?"

[1528] Step 8:

[1529] The server sends the generated responses and questions to each user's device.

[1530] Step 9:

[1531] The user's device displays the responses and questions received from the server on the screen, plays animations of the AI ​​character, and plays the responses and questions aloud.

[1532] Step 10:

[1533] The server continuously monitors the progress of the conference, detects when the discussion stagnates, and generates appropriate questions for specific users to restart the discussion.

[1534] Step 11:

[1535] The server sends the generated questions to each user's device, which then displays the questions on the screen and plays them back aloud, allowing users to respond to the questions and revitalize the discussion.

[1536] Step 12:

[1537] The server monitors the end time of the meeting in real time and notifies everyone when the end time approaches.

[1538] Step 13:

[1539] The server summarizes the main points of the meeting and generates a summary comment, which the user's device displays on its screen and plays back audibly.

[1540] Step 14:

[1541] The server confirms that the conference has ended and sends an end notice to all users. The server then terminates the connections of all users and saves the conference log data.

[1542] The above is the specific processing flow of the program based on the claims. This system is expected to significantly improve the efficiency and vitality of online meetings.

[1543] Example 2

[1544] 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."

[1545] In online meetings, it is necessary to encourage participants to speak up, prevent discussions from stalling, and ensure the meeting progresses smoothly and actively. However, current online meeting systems have limitations in their ability to recognize stalls in speech and participants' emotional states and respond appropriately, resulting in reduced meeting efficiency. The objective of this invention is to solve these problems and provide more effective, human-like interactions.

[1546] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for monitoring the progress of the conference, a means for collecting voice data from participants in real time, a means for analyzing the voice data and converting it into text data, a means for analyzing the content of remarks based on the text data, a means for analyzing the emotions of the participants, a means for generating appropriate responses and questions based on the analysis results, and a means for displaying and playing back the generated responses and questions. This allows the conference to proceed smoothly and enables more lively discussions by recognizing the emotional states of the participants and taking appropriate measures. The "means for monitoring the progress of the conference" refers to a technology for monitoring the progress of the conference in real time and grasping the progress of the discussion.

[1547] The "means for collecting voice data from participants in real time" is a technology for capturing and collecting the voices of participants in real time during a conference.

[1548] "Means for analyzing voice data and converting it into text data" refers to technology for analyzing collected voice data and converting it into text data.

[1549] "Means for analyzing speech content based on text data" refers to a technology that uses converted text data to analyze the content of speech and use the results to help progress the meeting.

[1550] "Means for analyzing participants' emotions" refers to technology for recognizing and analyzing participants' emotional states based on collected audio and video data.

[1551] "Means for generating appropriate responses and questions based on analysis results" refers to technology for generating responses and questions appropriate to the progress of a meeting based on the results of analyzing the content of statements and the emotional state of participants.

[1552] The "means for displaying and playing back generated responses and questions" refers to technology for visually displaying generated responses and questions and playing them back as audio. This invention is a system for encouraging participants to speak up in online meetings and promoting smooth and lively discussions. This system has the ability to recognize the user's emotional state and provide appropriate responses and questions.

[1553] The specific details of the system and the program processing are explained below.

[1554] System configuration

[1555] This system consists of the following main means:

[1556] A means of monitoring the progress of a meeting

[1557] A means of collecting real-time audio data from participants

[1558] A means of analyzing voice data and converting it into text data

[1559] A method for analyzing speech content based on text data

[1560] A means of analyzing participants' emotions

[1561] A means of generating appropriate responses and questions based on the analysis results

[1562] A means to view and play generated responses and questions

[1563] Program processing overview

[1564] 1. Server: Conference preparation and connection

[1565] The server sends the meeting invitation link to the user at the specified date and time as an email or notification, for example, using a calendar management system.

[1566] The server checks the connections of all users and confirms that everyone has joined. It uses WebSocket to monitor the connection status in real time.

[1567] 2. Server: Collecting and analyzing voice data

[1568] The server collects voice data from each user in real time, which is streamed using WebRTC technology.

[1569] The server converts the voice data into text data using a voice analysis engine, for example, a voice recognition API.

[1570] The text data is analyzed using natural language processing (NLP) techniques, which can be natural language understanding APIs.

[1571] 3. Server: Applying sentiment analysis methods

[1572] The server uses an emotion analysis engine to recognize the user's emotional state based on the collected audio and video data. For example, it uses an image analysis API for emotion analysis.

[1573] The server analyzes the recognized emotion data and uses it to generate appropriate responses and questions.

[1574] 4. Server: Generates backchannels and questions

[1575] The server uses a generative AI model to generate appropriate responses and questions based on the analyzed speech content and emotion data. The generative AI model uses a natural language generation API.

[1576] For example, if the user is nervous, a back-channel response such as "Relax, I'm speaking clearly" can be generated.

[1577] 5. Device: Display and playback

[1578] The user's terminal displays the responses and questions sent from the server on the screen.

[1579] The device plays back responses and questions aloud along with the animation of the AI ​​character, using a speech synthesis API as its speech synthesis engine.

[1580] 6. Server: Conference progress management

[1581] The server continuously monitors the progress of the conference and detects when the discussion stagnates.

[1582] When a stall is identified, generate appropriate questions for specific users, such as "Could you explain this in more detail?"

[1583] 7. Server: Meeting summary creation

[1584] The server monitors the end time of the meeting and notifies everyone when the end time is approaching.

[1585] The server summarizes the main points of the meeting and generates summary comments using a generative AI model, then sends the results to the user's device.

[1586] 8. Server: End the conference

[1587] The server confirms that the time for the conference to end has come and sends an end notice to all users.

[1588] All users are terminated and the log data of the conference is saved, which can be saved in cloud storage, for example.

[1589] Examples and prompts

[1590] Example 1: Meeting initiation and emotion recognition

[1591] The server uses emotion analysis means to recognize the initial emotional state of the user when the user joins the conference.

[1592] The user begins to introduce himself, and the server detects the user's nervousness.

[1593] The server uses AI to generate responses such as "Relax, this is a good start."

[1594] The device displays the backchannel and plays it back aloud.

[1595] Example 2: Facilitation and emotional response

[1596] The server detects that the discussion is stagnating and uses emotion analysis means to recognize that a particular user is confused.

[1597] The server generates a question for a particular user, such as "Could you please explain this problem in more detail?"

[1598] The device displays the question on the screen and plays it aloud.

[1599] Users can answer questions to get the discussion moving again.

[1600] Prompt Sentence Examples

[1601] "Meeting start prompt": "Analyze the user's current emotional state, determine if they are nervous, and generate appropriate responses."

[1602] "Prompt for when the discussion stalls": "Monitor the progress of the discussion and generate appropriate questions when the discussion stalls."

[1603] The system of the present invention provides an environment in which participants can speak more actively in online meetings, improving meeting efficiency. Even when there are few participants speaking or the discussion stagnates, the AI ​​can provide appropriate responses and questions in real time, allowing the meeting to proceed smoothly and the discussion to proceed in a manner that is consistent with the purpose. In addition, emotion analysis means enables interactions that are tailored to the user's emotional state, providing a more human-like meeting experience.

[1604] The flow of the identification process in the second embodiment will be described with reference to Fig. 13. Step 1: Conference invitation and connection confirmation

[1605] The server sends a meeting invitation link to the user via email or notification at the specified date and time. The input is the user's email address and notification information, and the output is a notification that the invitation link will be sent. Specifically, the server manages the meeting schedule using a calendar management system and sends notifications at specific times.

[1606] The server confirms that the user clicked the link to join the conference. The input is the user's connection request, and the output is connection establishment information. WebSocket is used to monitor the connection status in real time.

[1607] Step 2: Collect and convert audio data

[1608] The server collects voice data from each user in real time. The input is the user's voice stream, and the output is the collected voice data. The voice data is streamed using WebRTC technology.

[1609] The server converts the collected voice data into text data using a voice analysis engine. The input is voice data, and the output is the converted text data. The specific process of converting voice data into text data is performed using a voice recognition API.

[1610] Step 3: Analyzing the speech

[1611] The server analyzes the text data using natural language processing (NLP) technology. The input is the converted text data, and the output is the analysis result of the speech content. Using a natural language understanding API, the intent and content of the speech are analyzed from the text data.

[1612] Step 4: Sentiment Analysis

[1613] The server uses an emotion analysis engine based on audio and video data to recognize the user's emotional state. The input is the user's audio and video data, and the output is the emotion analysis results. Specific processing to identify the emotional state is performed using image analysis APIs, etc.

[1614] Step 5: Generate responses and questions

[1615] The server uses a generative AI model to generate appropriate backchannels and questions based on the analyzed utterance content and emotional data. The input is the analysis results of the utterance content and the emotional analysis results, and the output is the generated backchannels and questions. Using the generative AI model, it is possible to generate a backchannel such as "Relax, that's a good start."

[1616] Step 6: View and play back responses and questions

[1617] The terminal displays the backchannels and questions sent from the server on the screen. The input is the backchannels and questions sent from the server, and the output is a text message displayed on the terminal.

[1618] The device plays back responses and questions aloud along with the animation of the AI ​​character. The input is voice data sent from the server, and the output is voice playback. The specific playback processing is performed using a speech synthesis API.

[1619] Step 7: Managing the meeting

[1620] The server continuously monitors the progress of the conference and detects when the discussion stagnates. The input is conference progress data, and the output is information on the detection of stagnation of the discussion.

[1621] The server generates an appropriate question for a specific user when the discussion stalls. The input is the stall detection information, and the output is the generated question. For example, it generates a question such as, "Could you please explain this point in more detail?"

[1622] Step 8: Create a meeting summary

[1623] The server monitors the end of the meeting and notifies everyone when the end is approaching. The input is the current time and the scheduled end time of the meeting, and the output is a notification that the end is imminent.

[1624] The server summarizes the main points of the meeting and generates summary comments. The input is a text log of the meeting content, and the output is summary comments. A generative AI model is used to create a meeting summary.

[1625] Step 9: End the meeting

[1626] The server confirms that the conference has ended and sends an end notification to all users. The input is the current time and the scheduled end time, and the output is the end notification.

[1627] The server terminates all user connections and saves the conference log data. The input is the participant connection information and conference log data, and the output is the saved log file. The server then processes the log to save it in cloud storage.

[1628] Through these steps, the system supports the progress of online meetings and provides an environment in which participants can be more actively involved.

[1629] (Application example 2)

[1630] 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."

[1631] Online meetings often lack interaction between participants, leading to stagnation in discussions. In particular, ignoring participants' emotional states can lead to tension and confusion, preventing meaningful dialogue. Furthermore, manually managing meeting progress is cumbersome and reduces efficiency. Meanwhile, virtual stores lack a system for providing appropriate product recommendations and support based on users' emotional states, making it difficult to increase purchasing motivation or streamline customer support.

[1632] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1633] In this invention, the server includes means for monitoring the progress of the conference, means for analyzing participants' comments, means for generating appropriate responses and questions based on the analysis results, means for displaying and playing back the generated responses and questions, means for recognizing the user's emotional state, and means for adjusting the responses and questions based on the user's emotional state. This provides an environment in online conferences where participants can speak more actively, improving the efficiency of the conference. Furthermore, in virtual stores, interactions tailored to the user's emotional state can be tailored to increase purchasing motivation and provide efficient customer support. The "means for monitoring the progress of the conference" is a tool for monitoring the progress of the conference in real time and proposing the next action at the appropriate time.

[1634] The "means for analyzing the content of comments" is a tool that converts participants' comments into text data and uses natural language processing technology to understand the content.

[1635] The "means for generating responses and questions" is a system for automatically generating appropriate responses and follow-up questions based on the analyzed content of statements.

[1636] The "means for displaying and playing" is a system for displaying the generated responses and questions on a user interface and playing them back aloud if necessary.

[1637] The "means for recognizing emotional states" is a system for analyzing a user's voice and video data and determining the user's emotional state in real time.

[1638] The "means for adjusting responses and questions based on emotional state" refers to a system for appropriately adjusting the content and tone of generated responses and questions in accordance with the recognized emotional state of the user. The present invention is a system for recognizing the emotional state of participants or users in online meetings or virtual stores and providing appropriate responses and questions based on that state. This system includes the following means.

[1639] 1. System program generation

[1640] This system is configured using the following hardware and software:

[1641] Sentiment analysis engine: Uses Amazon Rekognition and Google Cloud Speech-to-Text to analyze audio and video data.

[1642] Natural Language Processing Engine: Uses OpenAI GPT-4 to analyze what participants and users say and generate appropriate responses and questions.

[1643] User interface: We will use Unity to build interfaces that run on smartphones and head-mounted displays (HMDs).

[1644] Data collection: Real-time audio and video streaming is performed using WebRTC.

[1645] 2. Program processing explanation

[1646] a. Data Collection and Connections

[1647] When a user accesses a virtual store or online conference, the server uses WebRTC to collect real-time audio and video data. The user's device then sends this data to the server, where analysis begins.

[1648] b. Speech and emotion analysis

[1649] The server converts the collected voice data into text data using Google Cloud Speech-to-Text, and analyzes the video data in real time using Amazon Rekognition to detect the user's emotional state.

[1650] c. Natural Language Processing and Generation

[1651] The analyzed text data undergoes natural language processing using OpenAI GPT-4, which automatically generates appropriate responses and questions based on the content of the speech. For example, if the user does not understand a question, the system generates a question such as, "Can you explain more about the features of this product?"

[1652] d. Display / playback

[1653] The generated responses and questions are displayed in a Unity-based user interface, and the voice comments are played back using a TTS engine, providing the user with both visual and auditory feedback.

[1654] 3. Specific Examples

[1655] Example 1: Prompts for when the user is confused

[1656] It seems like the user doesn't understand the information on the product page. Suggest, "Can you explain more about this product's features?"

[1657] Example 2: Prompts for when the user is anxious

[1658] It seems like the user is hesitant to make a purchase. Suggest that you can find other products that fit your needs and budget.

[1659] effect

[1660] This system will improve the efficiency of online meetings by providing an environment in which participants can speak more actively. In addition, in virtual stores, it will enable interactions that are tailored to the user's emotional state, which will increase purchasing motivation and provide more efficient customer support.

[1661] The flow of the specific processing in Application Example 2 will be explained with reference to Figure 14. Step 1:

[1662] When a user accesses a virtual store or online conference, the server uses WebRTC to collect audio and video data in real time. This data is sent from the user's device to the server. The input is audio and video data, and the output is that this data is streamed to the server. Specifically, the moment the user accesses the server, WebRTC establishes a connection and starts streaming data.

[1663] Step 2:

[1664] To analyze the collected voice data, the server converts it into text data using the Google Cloud Speech-to-Text API. The input is voice data, and the output is text data. Once the text data is obtained, the server proceeds to the next step. Specifically, the server sends the voice data to the API, receives the returned text, and saves it in storage.

[1665] Step 3:

[1666] The server analyzes the collected video data using Amazon Rekognition to recognize the user's emotional state. The input is video data and the output is the user's emotional state. Specifically, the server sends the video data to the API, receives the emotional data returned as the analysis result, and stores it.

[1667] Step 4:

[1668] The server analyzes the text data and emotion data using the OpenAI GPT-4 model and generates appropriate responses and questions based on the content of the speech. The input is text data and emotion data, and the output is the generated response or question. Specifically, the server inputs the text data and emotion data as prompts for the generative AI model and obtains the generated sentence.

[1669] Step 5:

[1670] The server sends the generated responses and questions to the user interface. The input is the text of the responses and questions, and the output is the text displayed on the user interface. Specifically, the server sends the responses and questions in text format and displays them on the user's terminal.

[1671] Step 6:

[1672] The device displays the received responses and questions on the screen and plays them aloud if necessary. The input is text data from the server, and the output is visual and audio information presented to the user. Specifically, the text data is sent to the display component, and then played back as audio using the audio playback engine.

[1673] Step 7:

[1674] The server monitors the user's responses and continues to generate additional questions and responses as needed. The input is the user's new audio and video data, and the output is newly generated responses and questions. Specifically, it receives new data in real time via WebRTC and processes it again.

[1675] 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.

[1676] 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.

[1677] 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 robot 414.

[1678] 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.

[1679] 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.

[1680] 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.

[1681] 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).

[1682] 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.

[1683] 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."

[1684] 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.

[1685] 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).

[1686] 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.

[1687] 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.

[1688] 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.

[1689] 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.

[1690] 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.

[1691] 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.

[1692] 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.

[1693] 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.

[1694] 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.

[1695] 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.

[1696] The following is further disclosed regarding the above embodiment.

[1697] (Claim 1)

[1698] a means of monitoring the progress of the meeting;

[1699] A means of analyzing the content of participants' comments;

[1700] A means for generating appropriate responses and questions based on the analysis results;

[1701] A means to view and play generated responses and questions;

[1702] A system including:

[1703] (Claim 2)

[1704] The system of claim 1, wherein questions are generated to support the progress of the meeting.

[1705] (Claim 3)

[1706] 2. The system according to claim 1, wherein the system encourages participants to speak at appropriate times based on their speaking tendencies.

[1707] "Example 1"

[1708] (Claim 1)

[1709] a means of monitoring the progress of the meeting;

[1710] A means of analyzing the content of participants' comments;

[1711] A means for generating appropriate responses and questions based on the analysis results;

[1712] A means to view and play generated responses and questions;

[1713] A means for collecting user voice data in real time and converting it into text data using automatic voice recognition technology;

[1714] means for analyzing text data using natural language processing techniques;

[1715] A means to generate appropriate reactions and queries using generative AI models; and

[1716] means for converting the generated text into speech data using speech synthesis technology;

[1717] A system including:

[1718] (Claim 2)

[1719] The system of claim 1, wherein questions are generated to support the progress of the meeting.

[1720] (Claim 3)

[1721] 2. The system according to claim 1, wherein the system encourages participants to speak at appropriate times based on their speaking tendencies.

[1722] "Application Example 1"

[1723] (Claim 1)

[1724] a means of monitoring the progress of the meeting;

[1725] A means of analyzing the content of participants' comments;

[1726] A means for generating appropriate responses and questions based on the analysis results;

[1727] A means to view and play generated responses and questions;

[1728] Monitor customer interactions in virtual space,

[1729] a means of generating responses and questions at the right time;

[1730] A means to visually and audibly display and play generated responses and questions through an avatar

[1731] A system including:

[1732] (Claim 2)

[1733] The system of claim 1, wherein questions are generated to support the progress of the meeting.

[1734] (Claim 3)

[1735] 2. The system according to claim 1, wherein the system encourages participants to speak at appropriate times based on their speaking tendencies.

[1736] "Example 2 when combining emotion engines" (Claim 1)

[1737] a means of monitoring the progress of the meeting;

[1738] a means for collecting real-time audio data from participants;

[1739] A means for analyzing the voice data and converting it into text data;

[1740] A means for analyzing the content of statements based on text data;

[1741] A means of analyzing participants' emotions;

[1742] A means for generating appropriate responses and questions based on the analysis results;

[1743] A means to view and play generated responses and questions;

[1744] A system including:

[1745] (Claim 2)

[1746] The system of claim 1, wherein questions are generated to support the progress of the meeting.

[1747] (Claim 3)

[1748] The system according to claim 1, wherein the system encourages participants to speak based on their emotional data.

[1749] "Application example 2 when combining emotion engines"

[1750] (Claim 1)

[1751] a means of monitoring the progress of the meeting;

[1752] A means of analyzing the content of participants' comments;

[1753] A means for generating appropriate responses and questions based on the analysis results;

[1754] A means to view and play generated responses and questions;

[1755] means for recognizing the emotional state of a user;

[1756] means for tailoring responses and questions based on the user's emotional state;

[1757] A system including:

[1758] (Claim 2)

[1759] The system of claim 1, wherein questions are generated to support the progress of the meeting.

[1760] (Claim 3)

[1761] The system according to claim 1, wherein the system prompts participants to speak at appropriate times based on their speaking tendencies and emotional states. [Explanation of symbols]

[1762] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means of monitoring the progress of the meeting; A means of analyzing the content of participants' comments; A means for generating appropriate responses and questions based on the analysis results; A means to display and play back the generated responses and questions; A system including:

2. The system of claim 1 , wherein questions are generated to support the progress of a meeting.

3. The system according to claim 1, wherein the system encourages participants to speak at an appropriate timing based on their tendency to speak.

Images