System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating chatbots for debate using generative artificial intelligences; means for causing debate to be performed between the generated chatbots; means for determining winning or losing based on content of the debate; and means for providing feedback to a user based on a training history.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 today's world, improving debate skills is important, but creating an appropriate practice environment is difficult. Finding a practice partner is a particular challenge for debates, which require a conversation partner. Furthermore, obtaining objective evaluations and feedback after a debate is difficult, resulting in delays in improving technical details. The present invention aims to effectively solve these debate environment issues using generative artificial intelligence and provide users with a fulfilling training environment. [Means for solving the problem]

[0005] This invention provides a means for generating chatbots for debate using generative artificial intelligence and a means for having the generated chatbots debate with each other. Furthermore, by adding a means for determining the outcome of the debate based on the content of the debate, a system is constructed that evaluates the quality of the debate and provides feedback to the user based on their training history. It also includes a means for analyzing the content of the discussion in the generated chatbot and generating a score based on logic, persuasiveness, and the effectiveness of the rebuttal, and includes a means for the generative artificial intelligence to generate a chatbot based on the debate theme and settings set by the user, thereby providing an optimized debate environment for the user.

[0006] "Generative AI" is a type of AI model that automatically generates text and dialogue based on instructions and data provided by the user.

[0007] A "chatbot" is a computer program designed to interact with users in natural language, focusing on a specific task or response.

[0008] A "debate" is an activity in which logical arguments are made from multiple perspectives on a particular topic, and involves exchanging opinions and competing to prove the persuasiveness of opposing arguments.

[0009] A "means for determining victory or defeat" is a system for analyzing the content of a debate and determining which argument is superior based on specific evaluation criteria.

[0010] "Training History" is a record of a user's past practice and debate sessions, including data for assessing areas for improvement and progress.

[0011] "Feedback" is information that provides specific improvements and evaluations based on the user's performance, and is intended to lead to future practice and improvement.

[0012] "Means of analysis" refers to algorithms or systems that logically analyze the content of debate statements and generate scores or evaluations based on evaluation criteria.

[0013] "Topic and Settings" refers to information selected or entered by the user to define the content and conditions of the debate, and to determine the direction and focus of the discussion. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

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

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] MODE FOR CARRYING OUT THE INVENTION

[0036] This invention is a debate system using generative AI, and aims to enable users to efficiently improve their debating skills. This system has functions such as generating chatbots using generative AI, conducting debates between chatbots, analyzing the content of the debate and determining the winner, and providing feedback to users.

[0037] Program processing explanation

[0038] Chatbot generation

[0039] Server operations

[0040] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (e.g., a large-scale language model) to generate chatbot A and chatbot B to participate in a one-on-one debate. Each chatbot is given different opinions and personalities.

[0041] Conducting the debate

[0042] Server operations

[0043] 1. The server instructs Chatbot A and Chatbot B to start a conversation session.

[0044] 2. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to combat global warming."

[0045] 3. Chatbot B then takes over and offers a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practical for widespread adoption."

[0046] 4. This process is repeated over multiple turns as the debate progresses.

[0047] Judging the winner

[0048] Server operations

[0049] 1. After the debate is over, the server analyzes all statements made between the chatbots based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[0050] 2. A score is assigned to each statement, and finally an overall score is calculated.

[0051] 3. The server decides which chatbot is the winner based on the overall score.

[0052] Providing feedback

[0053] Operations performed by the server and the terminal

[0054] 1. The server generates ratings and sends feedback to the user's device.

[0055] 2. The device displays the results of the debate along with feedback, including specific comments on what was praised and areas for improvement.

[0056] Save practice history

[0057] Operations performed by the device

[0058] 1. The device stores the user's practice history in a database, which records the results and feedback of each debate session and can be referenced later.

[0059] Specific examples

[0060] 1. Starting example

[0061] In order for users to hold a debate on the theme of "environmental issues," two chatbots are set up on their devices: Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[0062] After setting up, press the "Start Debate" button.

[0063] 2. Example of debate progression

[0064] The server generates Chatbot A and Chatbot B.

[0065] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0066] Chatbot B responds by saying that renewable energy is expensive and inefficient with current technology.

[0067] This process is repeated multiple times.

[0068] 3. Example of results

[0069] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[0070] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "I was impressed by your emphasis on the cost issue of renewable energy."

[0071] By being equipped with these processes and functions, the debate system of the present invention can provide a training environment for users to effectively improve their debate skills.

[0072] The processing flow will be explained below.

[0073] Step 1:

[0074] User-defined themes

[0075] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[0076] Step 2:

[0077] The server generates the chatbot

[0078] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[0079] Step 3:

[0080] The server starts the debate

[0081] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0082] Step 4:

[0083] Chatbot conversations are on the rise

[0084] The server passes the turn to Chatbot B, who then counters by saying, "However, renewable energy is expensive and inefficient with current technology." This process is repeated multiple times, and the debate progresses until the specified number of turns or time is reached.

[0085] Step 5:

[0086] The server analyzes the debate

[0087] Once the debate is over, the server analyzes the content of each side's comments and assigns a score to each comment based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[0088] Step 6:

[0089] The server generates the results and feedback

[0090] The server generates the debate results and detailed feedback, including which chatbot won and how it performed across each criterion. Specific feedback includes points of merit and areas for improvement.

[0091] Step 7:

[0092] The terminal displays the results

[0093] The device receives the debate results and feedback sent from the server. The device displays to the user "Debate Results: Winner - Chatbot B" and also displays feedback such as "Your emphasis on the cost issue of renewable energy was highly praised."

[0094] Step 8:

[0095] The device stores your practice history

[0096] The device stores the user's practice history in a database, including the results and feedback from each debate session, for future reference.

[0097] Example 1

[0098] 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."

[0099] Conventional debate training systems have the problem that participants must find a real opponent and it is difficult to receive a fair evaluation. Furthermore, evaluation of logic and persuasiveness during a debate tends to be subjective, making it difficult to obtain concrete feedback. Furthermore, there is a lack of storage and utilization of training history to enable users to check their own progress. It is necessary to provide a system that solves these problems and efficiently improves debate skills.

[0100] 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.

[0101] In this invention, the server includes means for generating dialogue agents for debate using generative artificial intelligence, means for conducting dialogue sessions between the generated dialogue agents, means for determining victory or defeat based on the content of the dialogue between the generated dialogue agents, and means for providing feedback to the user based on the results and analysis of the dialogue session. This allows users to efficiently improve their debating skills without having to find a real opponent. Furthermore, users can receive objective evaluations of their logic and persuasiveness, and receive feedback on specific areas for improvement. Furthermore, the user's training history can be saved and referenced later, making it easier to monitor their own progress.

[0102] "Generative AI" refers to AI systems that can generate natural language based on input prompts, such as large-scale language models.

[0103] A "conversational agent" is a virtual person or character that is generated to interact with a user or another agent. In a debate session, characters with different opinions and personalities are generated.

[0104] A "dialogue session" refers to a series of dialogues between multiple dialogue agents. It is conducted in a debate format, with each agent taking turns speaking.

[0105] "Server" refers to a computer system that processes user input, generates interactive agents using generative artificial intelligence, and manages the interactive session.

[0106] "Means for determining victory or defeat" refers to algorithms or programs that analyze the content of a dialogue session and determine victory or defeat based on evaluation criteria such as logic, persuasiveness, and effectiveness of counterarguments.

[0107] "Feedback" means ratings and comments provided to a User based on the results and analysis of an Interactive Session, including suggestions for improvement and specific advice.

[0108] "Training History" refers to the data that records the results and feedback of a user's debate sessions, and is used to monitor the user's progress.

[0109] MODE FOR CARRYING OUT THE INVENTION

[0110] This invention is a debate system using generative AI, which aims to enable users to efficiently improve their debating skills. This system has functions such as generating dialogue agents using generative AI, running dialogue sessions between dialogue agents, analyzing the content of the dialogue and determining the winner, and providing feedback to the user.

[0111] Chatbot generation

[0112] Server operations

[0113] 1. The user inputs the debate topic and detailed settings via the terminal. A possible debate topic would be "expanding the use of renewable energy."

[0114] 2. The device sends the entered debate topic and detailed settings to the server.

[0115] 3. Based on the received information, the server creates and sends the following prompt to a generative artificial intelligence (e.g., a large-scale language model such as GPT-3):

[0116] Create two chatbots, each with a different perspective on the debate topic "Expanding the Use of Renewable Energy."

[0117] Chatbot A should support "expanding the use of renewable energy," and Chatbot B should support "improving the efficiency of the current energy supply system."

[0118] 4. Generative AI generates Chatbot A and Chatbot B based on the prompt, giving each one different voice and personality.

[0119] Conducting the debate

[0120] Server operations

[0121] 1. The server issues a command to start a dialogue session, and Chatbot A makes the first statement. For example, Chatbot A says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[0122] 2. The server records what Chatbot A says.

[0123] 3. Next, the server passes the turn to Chatbot B, who then makes a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practically deployed on a wide scale."

[0124] 4. Repeat this process multiple times to progress the debate.

[0125] Judging the winner

[0126] Server operations

[0127] 1. After the debate is over, the server analyzes all statements made between the agents based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[0128] 2. Assign a score to each statement and calculate an overall score.

[0129] 3. The server decides which conversational agent is the winner based on the total score.

[0130] Providing feedback

[0131] Operations performed by the server and the terminal

[0132] 1. The server generates ratings and sends feedback to the user's device.

[0133] 2. The device displays the results of the interactive session along with feedback, including specific comments about what was appreciated and areas for improvement. For example, a specific comment such as "I was impressed by your emphasis on the cost issues of renewable energy" may be provided.

[0134] Save practice history

[0135] Operations performed by the device

[0136] 1. The device stores the practice history, including the results and feedback of each debate session, in a database. This history records the results and feedback of each debate session and can be referenced later by the user.

[0137] This system allows users to efficiently improve their debating skills without having to find a real opponent. It also allows users to receive objective evaluations of their logic and persuasiveness, and provides feedback on specific areas for improvement. Furthermore, the system saves users' training history and allows them to refer to it later, making it easier to monitor their progress.

[0138] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0139] Step 1:

[0140] Entering themes and settings

[0141] The user inputs the debate topic and detailed settings through the terminal.

[0142] [Input] Debate topic and details entered by the user.

[0143] [Operation] For example, on the topic of "expanding the use of renewable energy," you can set Chatbot A to support it and Chatbot B to oppose it.

[0144] [Output] The terminal sends the input data (debate topic and detailed settings) to the server.

[0145] Step 2:

[0146] Creating a prompt statement

[0147] Based on the debate topic and settings received by the server, a prompt is created to be sent to the generative AI model.

[0148] [Input] Debate topic and detailed settings received from the device.

[0149] [Behavior] The server creates a prompt. For example, "Create two chatbots with different viewpoints on the debate topic 'Expanding the use of renewable energy.' Chatbot A should 'support expanding the use of renewable energy,' and chatbot B should 'support improving the efficiency of the current energy supply system.'"

[0150] [Output] Send the prompt sentence to the generative AI model.

[0151] Step 3:

[0152] Chatbot generation

[0153] The server generates chatbot A and chatbot B using the generative AI model.

[0154] [Input] The prompt sent to the generative AI model.

[0155] [How it works] The generative AI model generates Chatbot A and Chatbot B based on the prompt text. Each is given different opinions and personalities.

[0156] [Output] Two chatbots (A and B) are generated.

[0157] Step 4:

[0158] Debate session begins

[0159] The server issues a command to start a dialogue session, and Chatbot A makes its first statement.

[0160] [Input] Two generated chatbots (A and B).

[0161] [Operation] The server starts the internal debate module, and Chatbot A makes a statement. For example, it says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[0162] [Output] Chatbot A's statement.

[0163] Step 5:

[0164] Recording statements and transitioning to rebuttals

[0165] The server records what Chatbot A says and then passes the turn to Chatbot B.

[0166] [Input] Statement made by Chatbot A.

[0167] [Operation] The server records what Chatbot A says and passes the turn of rebuttal to Chatbot B. Chatbot B then makes a rebuttal. For example, it may say, "The cost of renewable energy is too high, making widespread adoption unrealistic."

[0168] [Output] Chatbot B's rebuttal.

[0169] Step 6:

[0170] Repeated interactive sessions

[0171] The above process of commenting and rebutting is repeated multiple times (for example, five times) to progress the debate.

[0172] [Input] Respective statements and rebuttals from Chatbot A and Chatbot B.

[0173] [Operation] The server records the comments and rebuttals for each turn and moves on to the next turn.

[0174] [Output] The complete debate session log.

[0175] Step 7:

[0176] Judging the winner

[0177] The server analyzes the debate session log and determines the winner.

[0178] [Input] The complete debate session log.

[0179] [How it works] The server analyzes the debate content and generates scores based on logic, persuasiveness, and effectiveness of counterarguments. It assigns a score to each statement, calculates the total score, and determines the winner.

[0180] [Output] Win / loss result.

[0181] Step 8:

[0182] Generate and send feedback

[0183] The server generates feedback based on the results and sends it to the user's terminal.

[0184] [Input] The result of the match and the score for each comment.

[0185] [Operation] The server generates an overall rating and specific feedback (for example, "I was impressed by the emphasis on the cost issue of renewable energy") and sends it to the user's device.

[0186] [Output] Feedback sent to the device.

[0187] Step 9:

[0188] Save practice history

[0189] The terminal stores the practice history, including the results and feedback of the debate session, in a database.

[0190] [Input] Feedback and results sent to the device.

[0191] [Operation] The device saves the feedback and results as practice history in a database.

[0192] [Output] Saved practice history data.

[0193] (Application example 1)

[0194] 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."

[0195] In recent years, there has been a demand for staff to improve their skills in customer service to improve customer satisfaction. However, training that simulates real-life customer service situations is costly and time-consuming. To solve this problem, an efficient and effective staff training system is needed.

[0196] 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.

[0197] In this invention, the server includes means for generating chatbots for debate using generative artificial intelligence, means for having the generated chatbots debate, means for determining the outcome based on the content of the debate, means for providing feedback to the user based on the training history, means for generating chatbots with different roles based on selected topics, means for running debates that simulate customer service scenes, and means for providing specific feedback for improvement based on the analysis results, thereby enabling training to efficiently improve customer service skills for staff at brick-and-mortar stores.

[0198] "Generative AI" is an AI technology that uses algorithms such as large-scale language models to generate new data and content.

[0199] A "chatbot" is an autonomous program that can conduct conversations based on specific tasks.

[0200] A "debate" is an activity in which opinions are exchanged and discussed on a specific topic.

[0201] The "method of determining the winner" is a method of analyzing the content of the debate and determining the winner based on logic and persuasiveness.

[0202] "Training history" is data that records the content and results of debate sessions that a user has conducted.

[0203] "Feedback" refers to specific advice and improvements provided to users based on an analysis of the debate results.

[0204] A "topic" is a theme or subject that will be addressed in a debate or simulation.

[0205] A "role" is the position or stance that each chatbot takes in the debate.

[0206] "Customer service scene" refers to the situation in which customers are treated and served in a physical store.

[0207] The "analysis results" are the results of analyzing the content of the debate based on the evaluation criteria.

[0208] "Improvement feedback" refers to specific improvement measures and suggestions provided to users based on the analysis results.

[0209] This invention is a training system for improving the customer service skills of store staff, and is realized using generative artificial intelligence. Specific embodiments of this system are described below.

[0210] Program Structure

[0211] The system consists of three main components: a server, a terminal, and a user.

[0212] Hardware and Software

[0213] Hardware: Smartphones, tablets (e.g., iOS, Android devices)

[0214] software:

[0215] Large-scale language models (e.g., GPT-4)

[0216] Database management system (MySQL, etc.)

[0217] Network communication protocol (REST API)

[0218] Processing flow

[0219] 1. User operations

[0220] The user launches the application on a smartphone or tablet and begins training.

[0221] The user selects a specific customer service topic (e.g., handling complaints, explaining new products, responding to customer requests, etc.).

[0222] 2. Server Operation

[0223] Based on the topic and settings received from the user, the server uses generative artificial intelligence (such as GPT-4) to generate Chatbot A and Chatbot B to participate in the debate.

[0224] Chatbot A and Chatbot B are assigned different roles (e.g., customer role, staff role).

[0225] 3. Conducting the debate

[0226] The server issues a command to start a debate session, and Chatbot A (the customer) makes the first statement, for example, "This product was defective, and I would like a refund."

[0227] Next, Chatbot B (playing the role of staff) responds, for example, by saying, "I'm sorry. We'll take the time to hear more details and provide an appropriate response."

[0228] This exchange is repeated for multiple turns.

[0229] 4. Analysis of results and feedback

[0230] Once the debate is over, the server analyzes all statements based on logic, persuasiveness, and effectiveness of counterarguments.

[0231] A score is assigned to each statement and a final overall score is calculated.

[0232] The server generates feedback based on the overall score and provides it to the user, including specific advice such as, "Your initial apology was timely, but you didn't ask detailed questions enough."

[0233] 5. History Storage

[0234] The server stores the user's training history in a database, which records the results and feedback of each debate session and allows the user to refer to it later.

[0235] Specific examples

[0236] For example, if a user selects the topic "Handling Complaints" and presses the "Start Debate" button, the system will send the following prompt to the server:

[0237] Theme: Handling complaints

[0238] Setting: A customer is dissatisfied with a product, and staff members handle the complaint appropriately.

[0239] As a result, Chatbot A will say, "This product was defective," and Chatbot B will respond, "Sorry. What was the problem?" This debate will continue for several turns, and finally feedback will be provided.

[0240] This system will enable store staff to efficiently and effectively improve their customer service skills.

[0241] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0242] Step 1:

[0243] A user launches the application using a smartphone or tablet and selects a specific customer service topic.

[0244] Input: User selects a topic (e.g., complaint handling) in the application.

[0245] Output: Selected topics and configuration information are sent to the server

[0246] Specific operation: The user selects "Complaint handling" on the app screen and presses the "Start debate" button.

[0247] Step 2:

[0248] The server uses generative artificial intelligence to generate chatbot A and chatbot B to participate in the debate based on the topic and settings received from the user.

[0249] Input: Topic and preference information received from the user

[0250] Output: Generated Chatbot A and Chatbot B

[0251] Specific operation: The server calls a large-scale language model (such as GPT-4) to generate chatbot A for the customer role and chatbot B for the staff role.

[0252] Step 3:

[0253] The server initiates the debate session, and Chatbot A makes the first statement.

[0254] Input: Generated Chatbot A and Chatbot B

[0255] Output: Chatbot A's first statement

[0256] Specific operation: The server makes Chatbot A say something like, "This product was defective, I would like a refund."

[0257] Step 4:

[0258] Chatbot B then makes a rebuttal.

[0259] Input: Chatbot A's statement

[0260] Output: Chatbot B's rebuttal

[0261] Specific operation: The server makes Chatbot B say something like, "I'm sorry. We will ask for more details and take appropriate action."

[0262] Step 5:

[0263] This debate is repeated for multiple turns.

[0264] Input: What Chatbot A and B say in each turn

[0265] Output: A dialogue log of the entire debate

[0266] Specific operation: The server manages the dialogue between chatbots and records each utterance.

[0267] Step 6:

[0268] Once the debate is over, the server analyzes all statements.

[0269] Input: Dialogue log of the entire debate

[0270] Output: Analysis results and scores for each utterance

[0271] Specific operation: The server analyzes the content of the statement and scores it based on logic, persuasiveness, and effectiveness of the counterargument.

[0272] Step 7:

[0273] The server generates an overall score and feedback and provides it to the user.

[0274] Input: Analysis results and scores for each statement

[0275] Output: Overall score and feedback message

[0276] Specific behavior: The server generates specific feedback such as "The timing of the initial apology was appropriate, but the question was not detailed enough" and sends it to the user's device.

[0277] Step 8:

[0278] The server stores the user's training history in a database.

[0279] Input: Results and feedback from the debate session

[0280] Output: Saved training history

[0281] Specific operation: The server saves the results and feedback of this debate session in a database for future reference.

[0282] 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.

[0283] MODE FOR CARRYING OUT THE INVENTION

[0284] This invention is a system that combines generative AI and an emotion engine to build a debate system that provides debate practice that takes into account the user's emotions. This system is composed of a chatbot generated by generative AI and an emotion engine that recognizes the user's emotions.

[0285] Program processing explanation

[0286] Chatbot generation

[0287] Server operations

[0288] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence to generate chatbot A and chatbot B. These chatbots are configured to have different opinions and personalities.

[0289] Emotion Recognition in Action

[0290] Operations performed by the device

[0291] 1. While users are debating, the device captures the data necessary for emotion recognition (e.g., facial expressions, voice tone, etc.) in real time.

[0292] 2. The device sends the captured data to the emotion engine, which analyzes the user's emotions.

[0293] Conducting the debate

[0294] Server operations

[0295] 1. The server sends a command to start a dialogue session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to prevent global warming."

[0296] Server operations

[0297] 1. The server passes the turn to Chatbot B. Chatbot B counters, saying, "Renewable energy is expensive and inefficient with current technology."

[0298] Emotion-Based Response Modulation

[0299] Server operations

[0300] 1. The server receives emotional data from the emotion engine and adjusts the chatbot's responses accordingly based on the user's emotional state. For example, if the user is feeling stressed, the chatbot's responses will be made gentler.

[0301] Judging the winner

[0302] Server operations

[0303] 1. After the debate is over, the server analyzes the content of the debate statements. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of counterarguments, etc.). The total score is calculated to determine which chatbot is the winner.

[0304] Providing feedback

[0305] Operations performed by the server and terminal

[0306] 1. The server generates ratings and sends feedback to the device.

[0307] 2. The device displays the debate results along with feedback, including what went well and what needs improvement.

[0308] Save practice history

[0309] Operations performed by the device

[0310] 1. The device stores the user's practice history in a database, which records the results and feedback of the debate session and can be referenced later.

[0311] Specific examples

[0312] 1. Starting example

[0313] To allow users to debate on the topic of "environmental issues" on their devices, chatbot A is set to advocate for "expanding the use of renewable energy," and chatbot B is set to advocate for "improving the efficiency of the current energy supply system."

[0314] After setting up, press the "Start Debate" button.

[0315] 2. Example of debate progression

[0316] The server generates Chatbot A and Chatbot B.

[0317] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0318] Chatbot B counters, "Renewable energy is expensive and inefficient with current technology."

[0319] This process is repeated multiple times.

[0320] Meanwhile, the emotion engine recognizes the user's emotions, and the server adjusts the chatbot's responses based on that data.

[0321] 3. Example of results

[0322] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[0323] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "Your emphasis on the cost issue of renewable energy was appreciated."

[0324] The analysis results from the emotion engine are also displayed, providing feedback on the user's emotional changes and the chatbot's corresponding adjustment history.

[0325] By incorporating these processes and functions, the debate system of the present invention provides a training environment for users to effectively improve their debate skills, and also enables flexible responses that take emotions into consideration.

[0326] The processing flow will be explained below.

[0327] MODE FOR CARRYING OUT THE INVENTION

[0328] Processing Steps

[0329] Step 1:

[0330] User-defined themes

[0331] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[0332] Step 2:

[0333] The server generates the chatbot

[0334] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[0335] Step 3:

[0336] The device captures emotional data

[0337] While the user is preparing for the debate, the device captures the user's emotional data (e.g., facial expressions, voice tone, heart rate, etc.) in real time. The emotional data is sent from the device to the emotion engine.

[0338] Step 4:

[0339] Emotion engine analyzes emotions

[0340] The emotion engine analyzes the user's emotions based on the captured data, and the analysis results include information such as whether the user is stressed, relaxed, or focused.

[0341] Step 5:

[0342] Start a debate

[0343] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0344] Step 6:

[0345] The chatbot dialogue progresses

[0346] The server passes the turn to Chatbot B, who then counters by saying that renewable energy is expensive and inefficient with current technology. This process is repeated multiple times, and the debate continues until the specified number of turns or time has been reached.

[0347] Step 7:

[0348] Emotion-Based Response Modulation

[0349] The server receives the analysis results from the emotion engine and adjusts the chatbot's response accordingly depending on the user's emotional state. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a calm tone and content.

[0350] Step 8:

[0351] Debate analysis and winner determination

[0352] After the debate is over, the server analyzes the content of the debate statements and assigns a score to each statement based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[0353] Step 9:

[0354] Generate results and feedback

[0355] The server generates the debate results and detailed feedback, including points of merit and areas for improvement, as well as feedback on the user's emotional changes.

[0356] Step 10:

[0357] The terminal displays the results

[0358] The device receives the results and feedback sent from the server. The device displays "Debate Results: Winner - Chatbot B" to the user and provides feedback such as "Your emphasis on the cost issue of renewable energy was highly praised." The device also displays the analysis results from the emotion engine, showing feedback on changes in the user's emotions and the chatbot's corresponding adjustments.

[0359] Step 11:

[0360] The device stores your practice history

[0361] The device stores the user's practice history in a database, which includes the results and feedback of each debate session and can be referenced later.

[0362] This system allows users to effectively improve their debating skills and also allows training that takes into account their own emotional state.

[0363] Example 2

[0364] 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."

[0365] While conventional debate systems can provide a training environment for improving technical debate skills, it is difficult to consider the user's emotional state. Furthermore, since users' emotions often influence the discussion in real debates, training that ignores emotions is inefficient. Furthermore, there is a problem that feedback on debate results is insufficient, making it difficult to contribute to the improvement of users' skills.

[0366] 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.

[0367] In this invention, the server includes means for generating artificial dialogue agents for debate using generative artificial intelligence, means for having the generated artificial dialogue agents debate, means for analyzing the emotional state of the user during the debate using an emotion engine that recognizes the user's emotions, means for adjusting the responses of the artificial dialogue agents based on the emotional state of the user, means for determining the outcome of the debate based on the content of the debate, and means for providing feedback to the user based on the practice history, thereby making it possible to effectively improve the debating skills while taking into consideration the emotional state of the user.

[0368] "Generative artificial intelligence" is an algorithm that learns from large amounts of data and generates new text.

[0369] An "artificial dialogue agent" is software that can carry out pre-programmed dialogues or dialogues that are dynamically generated using generative artificial intelligence.

[0370] A "debate" is an act of logical discussion between people with different positions or opinions on a particular topic.

[0371] The "emotion engine" is a technology that analyzes data such as the user's voice and facial expressions to grasp their emotional state in real time.

[0372] The "means of determining victory or defeat" is a system that analyzes the content of the debate, assigns points based on evaluation criteria such as logic, persuasiveness, and the effectiveness of the rebuttal, and determines which side is superior.

[0373] "Feedback" refers to evaluations and comments provided after a debate, providing users with information to understand their strengths and areas for improvement.

[0374] "Practice history" is a record of past debate sessions conducted by the user, including debate content, evaluations, feedback, and the like.

[0375] A "topic" is a particular topic or issue that is the subject of debate.

[0376] "Settings" are parameters that the user specifies regarding the progress and conditions of the debate.

[0377] This invention provides a debate system that combines generative AI and an emotion engine. This system allows generated chatbots to debate with each other and provides training while taking into account the user's emotions. This system is composed of the following main components:

[0378] 1. Chatbot Creation

[0379] Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (such as GPT-3) to generate Chatbot A and Chatbot B. These chatbots are configured to have different opinions and personalities. Specific examples of prompts include "Please state your opinion on the topic: Expand the use of renewable energy" and "Improve the efficiency of the current energy supply system."

[0380] 2. Emotion Recognition

[0381] The device captures the user's facial expressions with a webcam and collects their voice tone with a microphone while they debate. This data is sent to an emotion engine (e.g., emotion recognition API) and analyzed in real time. The analysis results are sent to a server and reflected in the progress of the debate.

[0382] 3. Conducting the debate

[0383] The server starts a dialogue session between Chatbot A and Chatbot B. Chatbot A makes an initial statement, followed by a rebuttal from Chatbot B. This exchange is repeated multiple times.

[0384] 4. Emotion-Based Response Modulation

[0385] The server receives the emotion data from the emotion engine and adjusts the chatbot's responses based on the user's emotional state, for example, changing the chatbot's tone to be calmer if the user is stressed.

[0386] 5. Judging the Winner

[0387] After the debate is over, the server analyzes the content of the debate and assigns scores based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is superior.

[0388] 6. Providing Feedback

[0389] The server sends the generated evaluation and feedback to the device, which displays it to the user, providing feedback on what was good and what needs improvement. The analysis results from the emotion engine are also displayed.

[0390] 7. Save your practice history

[0391] The device stores the user's practice history in a database, which includes the results and feedback of the debate session. Users can later refer to this history to help them self-evaluate and improve their skills.

[0392] Through the above process, the present invention provides a training environment for users to effectively improve their debating skills. In particular, emotion recognition and response adjustment by the emotion engine enables flexible responses that take into account the user's emotional state. As a result, it is possible to provide higher satisfaction and effectiveness than conventional debate systems.

[0393] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0394] Step 1: Receiving the debate topic and detailed settings

[0395] server

[0396] Input: Receive debate topic and detailed settings from user.

[0397] Data processing: Analyzing received information and classifying themes and settings.

[0398] Output: Parsed themes and settings are sent to the generative AI.

[0399] Specific operation: The user enters the debate topic "environmental issues" and detailed settings into the terminal, and the server receives and analyzes this.

[0400] Step 2: Initializing the generative AI and creating a prompt

[0401] server

[0402] Input: Parsed themes and settings.

[0403] Data processing: Initialize a generative artificial intelligence (e.g., GPT-3) and create a prompt based on the theme.

[0404] Output: Send the prompt to the generative artificial intelligence model.

[0405] Specific operation: The server creates prompt sentences for Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[0406] Step 3: Generate the chatbot

[0407] server

[0408] Input: Prompt statement.

[0409] Data processing: Chatbot A and Chatbot B are generated using generative artificial intelligence.

[0410] Output: Saves the generated chatbot information.

[0411] Specific operation: The server uses GPT-3 to generate two different chatbots based on the prompt sentence.

[0412] Step 4: Capture emotion recognition data

[0413] Terminal

[0414] Input: User's facial expressions, voice tone.

[0415] Data processing: Capture data using a webcam and microphone.

[0416] Output: Sends the captured data to the emotion engine.

[0417] Specific operation: The device captures the user's facial expressions with a webcam and collects voice tones with a microphone.

[0418] Step 5: Analyze the sentiment data

[0419] Terminal

[0420] Input: Captured emotion data.

[0421] Data processing: Analyze using an emotion engine (e.g., emotion recognition API).

[0422] Output: Send the analysis results to the server.

[0423] Specific operation: The emotion engine analyzes the user's stress level, joy, anger, etc. and sends the results to the server.

[0424] Step 6: Start the debate session

[0425] server

[0426] Input: Information for the generated chatbot.

[0427] Data processing: Initialize the conversation session between Chatbot A and Chatbot B.

[0428] Output: Send the first command to Chatbot A.

[0429] Specific operation: The server makes chatbot A say, "We need to expand the use of renewable energy."

[0430] Step 7: Continue the conversation

[0431] server

[0432] Input: What Chatbot A says.

[0433] Data processing: Send a counterargument to Chatbot B.

[0434] Output: Record what Chatbot B said.

[0435] Specific behavior: The server has Chatbot B respond by saying, "Renewable energy is too expensive right now."

[0436] Step 8: Emotion-Based Response Adjustment

[0437] server

[0438] Input: Analysis results from the emotion engine.

[0439] Data manipulation: Tailoring chatbot responses based on the user's emotional state.

[0440] Output: Sends the tailored response to the chatbot.

[0441] Specific behavior: If the user is feeling stressed, the server changes the chatbot's tone to be gentler.

[0442] Step 9: Ending the debate and deciding who won

[0443] server

[0444] Input: The entire debate.

[0445] Data processing: Generate scores based on logic, persuasiveness, and effectiveness of counterarguments.

[0446] Output: Determine winner based on overall score.

[0447] Specific operation: The server determines the "Total score: Chatbot B is the winner" and displays the reason.

[0448] Step 10: Provide feedback

[0449] Servers and Terminals

[0450] Input: Debate evaluation and feedback.

[0451] Data processing: Generate ratings and feedback and send them to your device.

[0452] Output: Display feedback to the user.

[0453] Specific operation: The server generates feedback based on the evaluation, and the device displays something like, "Your emphasis on the cost issue of renewable energy was appreciated."

[0454] Step 11: Save your practice history

[0455] Terminal

[0456] Input: Results, evaluations, and feedback from the debate session.

[0457] Data processing: Practice history is saved in a database.

[0458] Output: Makes the history available to the user.

[0459] What it does: The device stores the debate results and feedback in a database so that the user can refer to them later.

[0460] (Application example 2)

[0461] 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."

[0462] Conventional debate training systems tend to engage in monotonous dialogue without considering the user's emotional state, making it difficult to provide a flexible training environment that can respond to changes in emotions. Furthermore, they are unable to provide appropriate feedback that responds immediately to the user's stress or emotional changes, which increases the burden on the user. Furthermore, because applications to other fields such as food delivery have not been considered, there is a lack of application of emotion recognition in practical scenarios other than debates.

[0463] 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.

[0464] In this invention, the server includes a means for generating chatbots for debate using generative artificial intelligence, a means for conducting debates between the generated chatbots, a means for using an emotion recognition engine to recognize users' emotions in real time, a means for adjusting the chatbots' responses based on the users' emotions, a means for determining the outcome of the debate based on the content of the debate, and a means for providing feedback to the user based on the training history. This enables flexible and effective debate training that takes into account the user's emotional state. This system can also be applied to other fields such as food delivery, and can have practical applications such as an emotion-based ordering system.

[0465] "Generative AI" is AI that uses technology to automatically generate text and conversations based on given input data.

[0466] A "chatbot" is a computer program that can converse with humans in natural language, and is particularly used in interactive formats such as debates.

[0467] An "emotion recognition engine" is a technology that analyzes data such as voice and images to identify a user's emotional state in real time.

[0468] The "means for determining victory or defeat" is a function that analyzes the results of a dialogue or debate and determines which chatbot is superior based on the set evaluation criteria.

[0469] The "means for providing feedback" is a function for notifying the user of the results of the debate and areas for improvement, and for providing information to improve the effectiveness of training.

[0470] "Training history" is data that accumulates records and analysis results of debate sessions conducted by users.

[0471] The "means for adjusting the chatbot's response based on emotions" is a function that adjusts the chatbot to return an appropriate response based on the user's emotional data obtained by the emotion recognition engine.

[0472] This invention relates to a debate system that combines generative artificial intelligence and an emotion recognition engine. It provides a training environment for users to effectively improve their debate skills and enables flexible responses according to the user's emotional state.

[0473] The debate system of the present invention is composed of a server and terminals. Specifically, the following process is performed.

[0474] The server uses generative artificial intelligence to generate chatbots for debates based on the debate topic and settings received from the user. The generated chatbots are configured to have different opinions and personalities. A means is provided for having the chatbots debate and recording the content of the debates.

[0475] The device uses an emotion recognition engine to recognize users' emotions in real time during debates. The emotion recognition engine detects the user's emotional state by capturing and analyzing data such as facial expressions and voice tones.

[0476] The server adjusts the chatbot's responses appropriately based on data from the emotion engine. For example, if the user is feeling stressed, the chatbot's responses will be gentler. It also determines the outcome of the debate based on the content of the debate, and generates and provides evaluations and feedback to the user.

[0477] Next, we will explain an example of a food delivery application. This system uses an emotion recognition engine to suggest optimal menus and ordering methods based on the user's emotional state in a food delivery service.

[0478] The device captures the user's voice and image data and analyzes the user's emotions using an emotion recognition engine such as EmotionRecognition. It then uses MenuGenerator to generate an optimal menu based on the analyzed emotion data. The generated menu is then presented to the user by a chatbot, which responds based on the user's emotions, making it possible to suggest the most suitable dishes for the user. The device can then finalize the order using the OrderSystem.

[0479] As a specific example, the following steps can be considered.

[0480] 1. A user asks, "What's the recommendation of the day?"

[0481] 2. The emotion recognition engine analyzes the user's voice and facial expressions to detect when the user is feeling stressed.

[0482] 3. Menu Generator suggests dishes that have a relaxing effect.

[0483] 4. The chatbot suggests, "How about some relaxing green tea cookies and chamomile tea?"

[0484] An example of a prompt for the generative AI model is as follows:

[0485] The user's name is "Tanaka." Please suggest some dishes that would be suitable for Tanaka when he is feeling stressed.

[0486] For example: "How about some relaxing green tea cookies and chamomile tea?"

[0487] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0488] Step 1:

[0489] The user sends the debate topic and settings to the server via the terminal. The data input to the terminal includes the debate topic, setting details, and user identification information. For example, input data might include "Today's debate topic is environmental issues," "Expanding the use of renewable energy," and "Improving the efficiency of the current energy supply system."

[0490] Step 2:

[0491] The server uses generative artificial intelligence to generate chatbots for debate. In this process, the server inputs the theme and settings received from the user and generates Chatbot A and Chatbot B, each with different opinions and personalities. For example, Chatbot A will be a character that advocates "expanding the use of renewable energy," while Chatbot B will be a character that advocates "improving the efficiency of the current energy supply system."

[0492] Step 3:

[0493] The device captures audio and video data to recognize users' emotions in real time during a debate. This data is sent from the device to an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to generate emotion data. The captured data is the audio and video of the user's facial expressions as they speak during the debate.

[0494] Step 4:

[0495] The server receives emotional data from the emotion recognition engine and adjusts the chatbot's response based on that data. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a gentle tone. The input data is the emotional data sent from the emotion recognition engine, and the output is the adjusted chatbot's response.

[0496] Step 5:

[0497] The server initiates a dialogue session between Chatbot A and Chatbot B and records their statements. The server then allows Chatbot A to make the first statement, followed by Chatbot B's rebuttal. For example, Chatbot A may say, "Renewable energy is an important means of preventing global warming," and Chatbot B may then rebut, "Renewable energy is expensive and inefficient with current technology."

[0498] Step 6:

[0499] The server determines the winner based on the content of the debate. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is the winner. The input data are the content of the debate, and the output is the overall score and the winner's result.

[0500] Step 7:

[0501] The server generates evaluations and feedback along with the outcome of the debate and sends them to the device. The device then displays the debate results and feedback to the user. Based on the winner of the debate, feedback such as "Chatbot B is the winner. It was praised for emphasizing the cost issue of renewable energy" is presented.

[0502] Step 8:

[0503] The device stores the user's practice history in a database. This history records the results and feedback of debate sessions, and the user can refer to it later. The database stores the user's debate topic, win / loss results, score, feedback, etc.

[0504] This process flow allows users to effectively improve their debating skills. Furthermore, the use of an emotion recognition engine allows for flexible responses according to the user's emotional state, providing a comfortable training environment.

[0505] 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.

[0506] 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.

[0507] 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.

[0508] [Second embodiment]

[0509] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0510] 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.

[0511] 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).

[0512] 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.

[0513] 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.

[0514] 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).

[0515] 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.

[0516] 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.

[0517] 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.

[0518] 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.

[0519] 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.

[0520] 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."

[0521] MODE FOR CARRYING OUT THE INVENTION

[0522] This invention is a debate system using generative AI, and aims to enable users to efficiently improve their debating skills. This system has functions such as generating chatbots using generative AI, conducting debates between chatbots, analyzing the content of the debate and determining the winner, and providing feedback to users.

[0523] Program processing explanation

[0524] Chatbot generation

[0525] Server operations

[0526] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (e.g., a large-scale language model) to generate chatbot A and chatbot B to participate in a one-on-one debate. Each chatbot is given different opinions and personalities.

[0527] Conducting the debate

[0528] Server operations

[0529] 1. The server instructs Chatbot A and Chatbot B to start a conversation session.

[0530] 2. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to combat global warming."

[0531] 3. Chatbot B then takes over and offers a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practical for widespread adoption."

[0532] 4. This process is repeated over multiple turns as the debate progresses.

[0533] Judging the winner

[0534] Server operations

[0535] 1. After the debate is over, the server analyzes all statements made between the chatbots based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[0536] 2. A score is assigned to each statement, and finally an overall score is calculated.

[0537] 3. The server decides which chatbot is the winner based on the overall score.

[0538] Providing feedback

[0539] Operations performed by the server and the terminal

[0540] 1. The server generates ratings and sends feedback to the user's device.

[0541] 2. The device displays the results of the debate along with feedback, including specific comments on what was praised and areas for improvement.

[0542] Save practice history

[0543] Operations performed by the device

[0544] 1. The device stores the user's practice history in a database, which records the results and feedback of each debate session and can be referenced later.

[0545] Specific examples

[0546] 1. Starting example

[0547] In order for users to hold a debate on the theme of "environmental issues," two chatbots are set up on their devices: Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[0548] After setting up, press the "Start Debate" button.

[0549] 2. Example of debate progression

[0550] The server generates Chatbot A and Chatbot B.

[0551] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0552] Chatbot B responds by saying that renewable energy is expensive and inefficient with current technology.

[0553] This process is repeated multiple times.

[0554] 3. Example of results

[0555] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[0556] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "I was impressed by your emphasis on the cost issue of renewable energy."

[0557] By being equipped with these processes and functions, the debate system of the present invention can provide a training environment for users to effectively improve their debate skills.

[0558] The processing flow will be explained below.

[0559] Step 1:

[0560] User-defined themes

[0561] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[0562] Step 2:

[0563] The server generates the chatbot

[0564] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[0565] Step 3:

[0566] The server starts the debate

[0567] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0568] Step 4:

[0569] Chatbot conversations are on the rise

[0570] The server passes the turn to Chatbot B, who then counters by saying, "However, renewable energy is expensive and inefficient with current technology." This process is repeated multiple times, and the debate progresses until the specified number of turns or time is reached.

[0571] Step 5:

[0572] The server analyzes the debate

[0573] Once the debate is over, the server analyzes the content of each side's comments and assigns a score to each comment based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[0574] Step 6:

[0575] The server generates the results and feedback

[0576] The server generates the debate results and detailed feedback, including which chatbot won and how it performed across each criterion. Specific feedback includes points of merit and areas for improvement.

[0577] Step 7:

[0578] The terminal displays the results

[0579] The device receives the debate results and feedback sent from the server. The device displays to the user "Debate Results: Winner - Chatbot B" and also displays feedback such as "Your emphasis on the cost issue of renewable energy was highly praised."

[0580] Step 8:

[0581] The device stores your practice history

[0582] The device stores the user's practice history in a database, including the results and feedback from each debate session, for future reference.

[0583] Example 1

[0584] 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."

[0585] Conventional debate training systems have the problem that participants must find a real opponent and it is difficult to receive a fair evaluation. Furthermore, evaluation of logic and persuasiveness during a debate tends to be subjective, making it difficult to obtain concrete feedback. Furthermore, there is a lack of storage and utilization of training history to enable users to check their own progress. It is necessary to provide a system that solves these problems and efficiently improves debate skills.

[0586] 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.

[0587] In this invention, the server includes means for generating dialogue agents for debate using generative artificial intelligence, means for conducting dialogue sessions between the generated dialogue agents, means for determining victory or defeat based on the content of the dialogue between the generated dialogue agents, and means for providing feedback to the user based on the results and analysis of the dialogue session. This allows users to efficiently improve their debating skills without having to find a real opponent. Furthermore, users can receive objective evaluations of their logic and persuasiveness, and receive feedback on specific areas for improvement. Furthermore, the user's training history can be saved and referenced later, making it easier to monitor their own progress.

[0588] "Generative AI" refers to AI systems that can generate natural language based on input prompts, such as large-scale language models.

[0589] A "conversational agent" is a virtual person or character that is generated to interact with a user or another agent. In a debate session, characters with different opinions and personalities are generated.

[0590] A "dialogue session" refers to a series of dialogues between multiple dialogue agents. It is conducted in a debate format, with each agent taking turns speaking.

[0591] "Server" refers to a computer system that processes user input, generates interactive agents using generative artificial intelligence, and manages the interactive session.

[0592] "Means for determining victory or defeat" refers to algorithms or programs that analyze the content of a dialogue session and determine victory or defeat based on evaluation criteria such as logic, persuasiveness, and effectiveness of counterarguments.

[0593] "Feedback" means ratings and comments provided to a User based on the results and analysis of an Interactive Session, including suggestions for improvement and specific advice.

[0594] "Training History" refers to the data that records the results and feedback of a user's debate sessions, and is used to monitor the user's progress.

[0595] MODE FOR CARRYING OUT THE INVENTION

[0596] This invention is a debate system using generative AI, which aims to enable users to efficiently improve their debating skills. This system has functions such as generating dialogue agents using generative AI, running dialogue sessions between dialogue agents, analyzing the content of the dialogue and determining the winner, and providing feedback to the user.

[0597] Chatbot generation

[0598] Server operations

[0599] 1. The user inputs the debate topic and detailed settings via the terminal. A possible debate topic would be "expanding the use of renewable energy."

[0600] 2. The device sends the entered debate topic and detailed settings to the server.

[0601] 3. Based on the received information, the server creates and sends the following prompt to a generative artificial intelligence (e.g., a large-scale language model such as GPT-3):

[0602] Create two chatbots, each with a different perspective on the debate topic "Expanding the Use of Renewable Energy."

[0603] Chatbot A should support "expanding the use of renewable energy," and Chatbot B should support "improving the efficiency of the current energy supply system."

[0604] 4. Generative AI generates Chatbot A and Chatbot B based on the prompt, giving each one different voice and personality.

[0605] Conducting the debate

[0606] Server operations

[0607] 1. The server issues a command to start a dialogue session, and Chatbot A makes the first statement. For example, Chatbot A says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[0608] 2. The server records what Chatbot A says.

[0609] 3. Next, the server passes the turn to Chatbot B, who then makes a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practically deployed on a wide scale."

[0610] 4. Repeat this process multiple times to progress the debate.

[0611] Judging the winner

[0612] Server operations

[0613] 1. After the debate is over, the server analyzes all statements made between the agents based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[0614] 2. Assign a score to each statement and calculate an overall score.

[0615] 3. The server decides which conversational agent is the winner based on the total score.

[0616] Providing feedback

[0617] Operations performed by the server and the terminal

[0618] 1. The server generates ratings and sends feedback to the user's device.

[0619] 2. The device displays the results of the interactive session along with feedback, including specific comments about what was appreciated and areas for improvement. For example, a specific comment such as "I was impressed by your emphasis on the cost issues of renewable energy" may be provided.

[0620] Save practice history

[0621] Operations performed by the device

[0622] 1. The device stores the practice history, including the results and feedback of each debate session, in a database. This history records the results and feedback of each debate session and can be referenced later by the user.

[0623] This system allows users to efficiently improve their debating skills without having to find a real opponent. It also allows users to receive objective evaluations of their logic and persuasiveness, and provides feedback on specific areas for improvement. Furthermore, the system saves users' training history and allows them to refer to it later, making it easier to monitor their progress.

[0624] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0625] Step 1:

[0626] Entering themes and settings

[0627] The user inputs the debate topic and detailed settings through the terminal.

[0628] [Input] Debate topic and details entered by the user.

[0629] [Operation] For example, on the topic of "expanding the use of renewable energy," you can set Chatbot A to support it and Chatbot B to oppose it.

[0630] [Output] The terminal sends the input data (debate topic and detailed settings) to the server.

[0631] Step 2:

[0632] Creating a prompt statement

[0633] Based on the debate topic and settings received by the server, a prompt is created to be sent to the generative AI model.

[0634] [Input] Debate topic and detailed settings received from the device.

[0635] [Behavior] The server creates a prompt. For example, "Create two chatbots with different viewpoints on the debate topic 'Expanding the use of renewable energy.' Chatbot A should 'support expanding the use of renewable energy,' and chatbot B should 'support improving the efficiency of the current energy supply system.'"

[0636] [Output] Send the prompt sentence to the generative AI model.

[0637] Step 3:

[0638] Chatbot generation

[0639] The server generates chatbot A and chatbot B using the generative AI model.

[0640] [Input] The prompt sent to the generative AI model.

[0641] [How it works] The generative AI model generates Chatbot A and Chatbot B based on the prompt text. Each is given different opinions and personalities.

[0642] [Output] Two chatbots (A and B) are generated.

[0643] Step 4:

[0644] Debate session begins

[0645] The server issues a command to start a dialogue session, and Chatbot A makes its first statement.

[0646] [Input] Two generated chatbots (A and B).

[0647] [Operation] The server starts the internal debate module, and Chatbot A makes a statement. For example, it says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[0648] [Output] Chatbot A's statement.

[0649] Step 5:

[0650] Recording statements and transitioning to rebuttals

[0651] The server records what Chatbot A says and then passes the turn to Chatbot B.

[0652] [Input] Statement made by Chatbot A.

[0653] [Operation] The server records what Chatbot A says and passes the turn of rebuttal to Chatbot B. Chatbot B then makes a rebuttal. For example, it may say, "The cost of renewable energy is too high, making widespread adoption unrealistic."

[0654] [Output] Chatbot B's rebuttal.

[0655] Step 6:

[0656] Repeated interactive sessions

[0657] The above process of commenting and rebutting is repeated multiple times (for example, five times) to progress the debate.

[0658] [Input] Respective statements and rebuttals from Chatbot A and Chatbot B.

[0659] [Operation] The server records the comments and rebuttals for each turn and moves on to the next turn.

[0660] [Output] The complete debate session log.

[0661] Step 7:

[0662] Judging the winner

[0663] The server analyzes the debate session log and determines the winner.

[0664] [Input] The complete debate session log.

[0665] [How it works] The server analyzes the debate content and generates scores based on logic, persuasiveness, and effectiveness of counterarguments. It assigns a score to each statement, calculates the total score, and determines the winner.

[0666] [Output] Win / loss result.

[0667] Step 8:

[0668] Generate and send feedback

[0669] The server generates feedback based on the results and sends it to the user's terminal.

[0670] [Input] The result of the match and the score for each comment.

[0671] [Operation] The server generates an overall rating and specific feedback (for example, "I was impressed by the emphasis on the cost issue of renewable energy") and sends it to the user's device.

[0672] [Output] Feedback sent to the device.

[0673] Step 9:

[0674] Save practice history

[0675] The terminal stores the practice history, including the results and feedback of the debate session, in a database.

[0676] [Input] Feedback and results sent to the device.

[0677] [Operation] The device saves the feedback and results as practice history in a database.

[0678] [Output] Saved practice history data.

[0679] (Application example 1)

[0680] 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."

[0681] In recent years, there has been a demand for staff to improve their skills in customer service to improve customer satisfaction. However, training that simulates real-life customer service situations is costly and time-consuming. To solve this problem, an efficient and effective staff training system is needed.

[0682] 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.

[0683] In this invention, the server includes means for generating chatbots for debate using generative artificial intelligence, means for having the generated chatbots debate, means for determining the outcome based on the content of the debate, means for providing feedback to the user based on the training history, means for generating chatbots with different roles based on selected topics, means for running debates that simulate customer service scenes, and means for providing specific feedback for improvement based on the analysis results, thereby enabling training to efficiently improve customer service skills for staff at brick-and-mortar stores.

[0684] "Generative AI" is an AI technology that uses algorithms such as large-scale language models to generate new data and content.

[0685] A "chatbot" is an autonomous program that can conduct conversations based on specific tasks.

[0686] A "debate" is an activity in which opinions are exchanged and discussed on a specific topic.

[0687] The "method of determining the winner" is a method of analyzing the content of the debate and determining the winner based on logic and persuasiveness.

[0688] "Training history" is data that records the content and results of debate sessions that a user has conducted.

[0689] "Feedback" refers to specific advice and improvements provided to users based on an analysis of the debate results.

[0690] A "topic" is a theme or subject that will be addressed in a debate or simulation.

[0691] A "role" is the position or stance that each chatbot takes in the debate.

[0692] "Customer service scene" refers to the situation in which customers are treated and served in a physical store.

[0693] The "analysis results" are the results of analyzing the content of the debate based on the evaluation criteria.

[0694] "Improvement feedback" refers to specific improvement measures and suggestions provided to users based on the analysis results.

[0695] This invention is a training system for improving the customer service skills of store staff, and is realized using generative artificial intelligence. Specific embodiments of this system are described below.

[0696] Program Structure

[0697] The system consists of three main components: a server, a terminal, and a user.

[0698] Hardware and Software

[0699] Hardware: Smartphones, tablets (e.g., iOS, Android devices)

[0700] software:

[0701] Large-scale language models (e.g., GPT-4)

[0702] Database management system (MySQL, etc.)

[0703] Network communication protocol (REST API)

[0704] Processing flow

[0705] 1. User operations

[0706] The user launches the application on a smartphone or tablet and begins training.

[0707] The user selects a specific customer service topic (e.g., handling complaints, explaining new products, responding to customer requests, etc.).

[0708] 2. Server Operation

[0709] Based on the topic and settings received from the user, the server uses generative artificial intelligence (such as GPT-4) to generate Chatbot A and Chatbot B to participate in the debate.

[0710] Chatbot A and Chatbot B are assigned different roles (e.g., customer role, staff role).

[0711] 3. Conducting the debate

[0712] The server issues a command to start a debate session, and Chatbot A (the customer) makes the first statement, for example, "This product was defective, and I would like a refund."

[0713] Next, Chatbot B (playing the role of staff) responds, for example, by saying, "I'm sorry. We'll take the time to hear more details and provide an appropriate response."

[0714] This exchange is repeated for multiple turns.

[0715] 4. Analysis of results and feedback

[0716] Once the debate is over, the server analyzes all statements based on logic, persuasiveness, and effectiveness of counterarguments.

[0717] A score is assigned to each statement and a final overall score is calculated.

[0718] The server generates feedback based on the overall score and provides it to the user, including specific advice such as, "Your initial apology was timely, but you didn't ask detailed questions enough."

[0719] 5. History Storage

[0720] The server stores the user's training history in a database, which records the results and feedback of each debate session and allows the user to refer to it later.

[0721] Specific examples

[0722] For example, if a user selects the topic "Handling Complaints" and presses the "Start Debate" button, the system will send the following prompt to the server:

[0723] Theme: Handling complaints

[0724] Setting: A customer is dissatisfied with a product, and staff members handle the complaint appropriately.

[0725] As a result, Chatbot A will say, "This product was defective," and Chatbot B will respond, "Sorry. What was the problem?" This debate will continue for several turns, and finally feedback will be provided.

[0726] This system will enable store staff to efficiently and effectively improve their customer service skills.

[0727] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0728] Step 1:

[0729] A user launches the application using a smartphone or tablet and selects a specific customer service topic.

[0730] Input: User selects a topic (e.g., complaint handling) in the application.

[0731] Output: Selected topics and configuration information are sent to the server

[0732] Specific operation: The user selects "Complaint handling" on the app screen and presses the "Start debate" button.

[0733] Step 2:

[0734] The server uses generative artificial intelligence to generate chatbot A and chatbot B to participate in the debate based on the topic and settings received from the user.

[0735] Input: Topic and preference information received from the user

[0736] Output: Generated Chatbot A and Chatbot B

[0737] Specific operation: The server calls a large-scale language model (such as GPT-4) to generate chatbot A for the customer role and chatbot B for the staff role.

[0738] Step 3:

[0739] The server initiates the debate session, and Chatbot A makes the first statement.

[0740] Input: Generated Chatbot A and Chatbot B

[0741] Output: Chatbot A's first statement

[0742] Specific operation: The server makes Chatbot A say something like, "This product was defective, I would like a refund."

[0743] Step 4:

[0744] Chatbot B then makes a rebuttal.

[0745] Input: Chatbot A's statement

[0746] Output: Chatbot B's rebuttal

[0747] Specific operation: The server makes Chatbot B say something like, "I'm sorry. We will ask for more details and take appropriate action."

[0748] Step 5:

[0749] This debate is repeated for multiple turns.

[0750] Input: What Chatbot A and B say in each turn

[0751] Output: A dialogue log of the entire debate

[0752] Specific operation: The server manages the dialogue between chatbots and records each utterance.

[0753] Step 6:

[0754] Once the debate is over, the server analyzes all statements.

[0755] Input: Dialogue log of the entire debate

[0756] Output: Analysis results and scores for each utterance

[0757] Specific operation: The server analyzes the content of the statement and scores it based on logic, persuasiveness, and effectiveness of the counterargument.

[0758] Step 7:

[0759] The server generates an overall score and feedback and provides it to the user.

[0760] Input: Analysis results and scores for each statement

[0761] Output: Overall score and feedback message

[0762] Specific behavior: The server generates specific feedback such as "The timing of the initial apology was appropriate, but the question was not detailed enough" and sends it to the user's device.

[0763] Step 8:

[0764] The server stores the user's training history in a database.

[0765] Input: Results and feedback from the debate session

[0766] Output: Saved training history

[0767] Specific operation: The server saves the results and feedback of this debate session in a database for future reference.

[0768] 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.

[0769] MODE FOR CARRYING OUT THE INVENTION

[0770] This invention is a system that combines generative AI and an emotion engine to build a debate system that provides debate practice that takes into account the user's emotions. This system is composed of a chatbot generated by generative AI and an emotion engine that recognizes the user's emotions.

[0771] Program processing explanation

[0772] Chatbot generation

[0773] Server operations

[0774] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence to generate chatbot A and chatbot B. These chatbots are configured to have different opinions and personalities.

[0775] Emotion Recognition in Action

[0776] Operations performed by the device

[0777] 1. While users are debating, the device captures the data necessary for emotion recognition (e.g., facial expressions, voice tone, etc.) in real time.

[0778] 2. The device sends the captured data to the emotion engine, which analyzes the user's emotions.

[0779] Conducting the debate

[0780] Server operations

[0781] 1. The server sends a command to start a dialogue session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to prevent global warming."

[0782] Server operations

[0783] 1. The server passes the turn to Chatbot B. Chatbot B counters, saying, "Renewable energy is expensive and inefficient with current technology."

[0784] Emotion-Based Response Modulation

[0785] Server operations

[0786] 1. The server receives emotional data from the emotion engine and adjusts the chatbot's responses accordingly based on the user's emotional state. For example, if the user is feeling stressed, the chatbot's responses will be made gentler.

[0787] Judging the winner

[0788] Server operations

[0789] 1. After the debate is over, the server analyzes the content of the debate statements. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of counterarguments, etc.). The total score is calculated to determine which chatbot is the winner.

[0790] Providing feedback

[0791] Operations performed by the server and terminal

[0792] 1. The server generates ratings and sends feedback to the device.

[0793] 2. The device displays the debate results along with feedback, including what went well and what needs improvement.

[0794] Save practice history

[0795] Operations performed by the device

[0796] 1. The device stores the user's practice history in a database, which records the results and feedback of the debate session and can be referenced later.

[0797] Specific examples

[0798] 1. Starting example

[0799] To allow users to debate on the topic of "environmental issues" on their devices, chatbot A is set to advocate for "expanding the use of renewable energy," and chatbot B is set to advocate for "improving the efficiency of the current energy supply system."

[0800] After setting up, press the "Start Debate" button.

[0801] 2. Example of debate progression

[0802] The server generates Chatbot A and Chatbot B.

[0803] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0804] Chatbot B counters, "Renewable energy is expensive and inefficient with current technology."

[0805] This process is repeated multiple times.

[0806] Meanwhile, the emotion engine recognizes the user's emotions, and the server adjusts the chatbot's responses based on that data.

[0807] 3. Example of results

[0808] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[0809] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "Your emphasis on the cost issue of renewable energy was appreciated."

[0810] The analysis results from the emotion engine are also displayed, providing feedback on the user's emotional changes and the chatbot's corresponding adjustment history.

[0811] By incorporating these processes and functions, the debate system of the present invention provides a training environment for users to effectively improve their debate skills, and also enables flexible responses that take emotions into consideration.

[0812] The processing flow will be explained below.

[0813] MODE FOR CARRYING OUT THE INVENTION

[0814] Processing Steps

[0815] Step 1:

[0816] User-defined themes

[0817] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[0818] Step 2:

[0819] The server generates the chatbot

[0820] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[0821] Step 3:

[0822] The device captures emotional data

[0823] While the user is preparing for the debate, the device captures the user's emotional data (e.g., facial expressions, voice tone, heart rate, etc.) in real time. The emotional data is sent from the device to the emotion engine.

[0824] Step 4:

[0825] Emotion engine analyzes emotions

[0826] The emotion engine analyzes the user's emotions based on the captured data, and the analysis results include information such as whether the user is stressed, relaxed, or focused.

[0827] Step 5:

[0828] Start a debate

[0829] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[0830] Step 6:

[0831] The chatbot dialogue progresses

[0832] The server passes the turn to Chatbot B, who then counters by saying that renewable energy is expensive and inefficient with current technology. This process is repeated multiple times, and the debate continues until the specified number of turns or time has been reached.

[0833] Step 7:

[0834] Emotion-Based Response Modulation

[0835] The server receives the analysis results from the emotion engine and adjusts the chatbot's response accordingly depending on the user's emotional state. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a calm tone and content.

[0836] Step 8:

[0837] Debate analysis and winner determination

[0838] After the debate is over, the server analyzes the content of the debate statements and assigns a score to each statement based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[0839] Step 9:

[0840] Generate results and feedback

[0841] The server generates the debate results and detailed feedback, including points of merit and areas for improvement, as well as feedback on the user's emotional changes.

[0842] Step 10:

[0843] The terminal displays the results

[0844] The device receives the results and feedback sent from the server. The device displays "Debate Results: Winner - Chatbot B" to the user and provides feedback such as "Your emphasis on the cost issue of renewable energy was highly praised." The device also displays the analysis results from the emotion engine, showing feedback on changes in the user's emotions and the chatbot's corresponding adjustments.

[0845] Step 11:

[0846] The device stores your practice history

[0847] The device stores the user's practice history in a database, which includes the results and feedback of each debate session and can be referenced later.

[0848] This system allows users to effectively improve their debating skills and also allows training that takes into account their own emotional state.

[0849] Example 2

[0850] 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."

[0851] While conventional debate systems can provide a training environment for improving technical debate skills, it is difficult to consider the user's emotional state. Furthermore, since users' emotions often influence the discussion in real debates, training that ignores emotions is inefficient. Furthermore, there is a problem that feedback on debate results is insufficient, making it difficult to contribute to the improvement of users' skills.

[0852] 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.

[0853] In this invention, the server includes means for generating artificial dialogue agents for debate using generative artificial intelligence, means for having the generated artificial dialogue agents debate, means for analyzing the emotional state of the user during the debate using an emotion engine that recognizes the user's emotions, means for adjusting the responses of the artificial dialogue agents based on the emotional state of the user, means for determining the outcome of the debate based on the content of the debate, and means for providing feedback to the user based on the practice history, thereby making it possible to effectively improve the debating skills while taking into consideration the emotional state of the user.

[0854] "Generative artificial intelligence" is an algorithm that learns from large amounts of data and generates new text.

[0855] An "artificial dialogue agent" is software that can carry out pre-programmed dialogues or dialogues that are dynamically generated using generative artificial intelligence.

[0856] A "debate" is an act of logical discussion between people with different positions or opinions on a particular topic.

[0857] The "emotion engine" is a technology that analyzes data such as the user's voice and facial expressions to grasp their emotional state in real time.

[0858] The "means of determining victory or defeat" is a system that analyzes the content of the debate, assigns points based on evaluation criteria such as logic, persuasiveness, and the effectiveness of the rebuttal, and determines which side is superior.

[0859] "Feedback" refers to evaluations and comments provided after a debate, providing users with information to understand their strengths and areas for improvement.

[0860] "Practice history" is a record of past debate sessions conducted by the user, including debate content, evaluations, feedback, and the like.

[0861] A "topic" is a particular topic or issue that is the subject of debate.

[0862] "Settings" are parameters that the user specifies regarding the progress and conditions of the debate.

[0863] This invention provides a debate system that combines generative AI and an emotion engine. This system allows generated chatbots to debate with each other and provides training while taking into account the user's emotions. This system is composed of the following main components:

[0864] 1. Chatbot Creation

[0865] Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (such as GPT-3) to generate Chatbot A and Chatbot B. These chatbots are configured to have different opinions and personalities. Specific examples of prompts include "Please state your opinion on the topic: Expand the use of renewable energy" and "Improve the efficiency of the current energy supply system."

[0866] 2. Emotion Recognition

[0867] The device captures the user's facial expressions with a webcam and collects their voice tone with a microphone while they debate. This data is sent to an emotion engine (e.g., emotion recognition API) and analyzed in real time. The analysis results are sent to a server and reflected in the progress of the debate.

[0868] 3. Conducting the debate

[0869] The server starts a dialogue session between Chatbot A and Chatbot B. Chatbot A makes an initial statement, followed by a rebuttal from Chatbot B. This exchange is repeated multiple times.

[0870] 4. Emotion-Based Response Modulation

[0871] The server receives the emotion data from the emotion engine and adjusts the chatbot's responses based on the user's emotional state, for example, changing the chatbot's tone to be calmer if the user is stressed.

[0872] 5. Judging the Winner

[0873] After the debate is over, the server analyzes the content of the debate and assigns scores based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is superior.

[0874] 6. Providing Feedback

[0875] The server sends the generated evaluation and feedback to the device, which displays it to the user, providing feedback on what was good and what needs improvement. The analysis results from the emotion engine are also displayed.

[0876] 7. Save your practice history

[0877] The device stores the user's practice history in a database, which includes the results and feedback of the debate session. Users can later refer to this history to help them self-evaluate and improve their skills.

[0878] Through the above process, the present invention provides a training environment for users to effectively improve their debating skills. In particular, emotion recognition and response adjustment by the emotion engine enables flexible responses that take into account the user's emotional state. As a result, it is possible to provide higher satisfaction and effectiveness than conventional debate systems.

[0879] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0880] Step 1: Receiving the debate topic and detailed settings

[0881] server

[0882] Input: Receive debate topic and detailed settings from user.

[0883] Data processing: Analyzing received information and classifying themes and settings.

[0884] Output: Parsed themes and settings are sent to the generative AI.

[0885] Specific operation: The user enters the debate topic "environmental issues" and detailed settings into the terminal, and the server receives and analyzes this.

[0886] Step 2: Initializing the generative AI and creating a prompt

[0887] server

[0888] Input: Parsed themes and settings.

[0889] Data processing: Initialize a generative artificial intelligence (e.g., GPT-3) and create a prompt based on the theme.

[0890] Output: Send the prompt to the generative artificial intelligence model.

[0891] Specific operation: The server creates prompt sentences for Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[0892] Step 3: Generate the chatbot

[0893] server

[0894] Input: Prompt statement.

[0895] Data processing: Chatbot A and Chatbot B are generated using generative artificial intelligence.

[0896] Output: Saves the generated chatbot information.

[0897] Specific operation: The server uses GPT-3 to generate two different chatbots based on the prompt sentence.

[0898] Step 4: Capture emotion recognition data

[0899] Terminal

[0900] Input: User's facial expressions, voice tone.

[0901] Data processing: Capture data using a webcam and microphone.

[0902] Output: Sends the captured data to the emotion engine.

[0903] Specific operation: The device captures the user's facial expressions with a webcam and collects voice tones with a microphone.

[0904] Step 5: Analyze the sentiment data

[0905] Terminal

[0906] Input: Captured emotion data.

[0907] Data processing: Analyze using an emotion engine (e.g., emotion recognition API).

[0908] Output: Send the analysis results to the server.

[0909] Specific operation: The emotion engine analyzes the user's stress level, joy, anger, etc. and sends the results to the server.

[0910] Step 6: Start the debate session

[0911] server

[0912] Input: Information for the generated chatbot.

[0913] Data processing: Initialize the conversation session between Chatbot A and Chatbot B.

[0914] Output: Send the first command to Chatbot A.

[0915] Specific operation: The server makes chatbot A say, "We need to expand the use of renewable energy."

[0916] Step 7: Continue the conversation

[0917] server

[0918] Input: What Chatbot A says.

[0919] Data processing: Send a counterargument to Chatbot B.

[0920] Output: Record what Chatbot B said.

[0921] Specific behavior: The server has Chatbot B respond by saying, "Renewable energy is too expensive right now."

[0922] Step 8: Emotion-Based Response Adjustment

[0923] server

[0924] Input: Analysis results from the emotion engine.

[0925] Data manipulation: Tailoring chatbot responses based on the user's emotional state.

[0926] Output: Sends the tailored response to the chatbot.

[0927] Specific behavior: If the user is feeling stressed, the server changes the chatbot's tone to be gentler.

[0928] Step 9: Ending the debate and deciding who won

[0929] server

[0930] Input: The entire debate.

[0931] Data processing: Generate scores based on logic, persuasiveness, and effectiveness of counterarguments.

[0932] Output: Determine winner based on overall score.

[0933] Specific operation: The server determines the "Total score: Chatbot B is the winner" and displays the reason.

[0934] Step 10: Provide feedback

[0935] Servers and Terminals

[0936] Input: Debate evaluation and feedback.

[0937] Data processing: Generate ratings and feedback and send them to your device.

[0938] Output: Display feedback to the user.

[0939] Specific operation: The server generates feedback based on the evaluation, and the device displays something like, "Your emphasis on the cost issue of renewable energy was appreciated."

[0940] Step 11: Save your practice history

[0941] Terminal

[0942] Input: Results, evaluations, and feedback from the debate session.

[0943] Data processing: Practice history is saved in a database.

[0944] Output: Makes the history available to the user.

[0945] What it does: The device stores the debate results and feedback in a database so that the user can refer to them later.

[0946] (Application example 2)

[0947] 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."

[0948] Conventional debate training systems tend to engage in monotonous dialogue without considering the user's emotional state, making it difficult to provide a flexible training environment that can respond to changes in emotions. Furthermore, they are unable to provide appropriate feedback that responds immediately to the user's stress or emotional changes, which increases the burden on the user. Furthermore, because applications to other fields such as food delivery have not been considered, there is a lack of application of emotion recognition in practical scenarios other than debates.

[0949] 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.

[0950] In this invention, the server includes a means for generating chatbots for debate using generative artificial intelligence, a means for conducting debates between the generated chatbots, a means for using an emotion recognition engine to recognize users' emotions in real time, a means for adjusting the chatbots' responses based on the users' emotions, a means for determining the outcome of the debate based on the content of the debate, and a means for providing feedback to the user based on the training history. This enables flexible and effective debate training that takes into account the user's emotional state. This system can also be applied to other fields such as food delivery, and can have practical applications such as an emotion-based ordering system.

[0951] "Generative AI" is AI that uses technology to automatically generate text and conversations based on given input data.

[0952] A "chatbot" is a computer program that can converse with humans in natural language, and is particularly used in interactive formats such as debates.

[0953] An "emotion recognition engine" is a technology that analyzes data such as voice and images to identify a user's emotional state in real time.

[0954] The "means for determining victory or defeat" is a function that analyzes the results of a dialogue or debate and determines which chatbot is superior based on the set evaluation criteria.

[0955] The "means for providing feedback" is a function for notifying the user of the results of the debate and areas for improvement, and for providing information to improve the effectiveness of training.

[0956] "Training history" is data that accumulates records and analysis results of debate sessions conducted by users.

[0957] The "means for adjusting the chatbot's response based on emotions" is a function that adjusts the chatbot to return an appropriate response based on the user's emotional data obtained by the emotion recognition engine.

[0958] This invention relates to a debate system that combines generative artificial intelligence and an emotion recognition engine. It provides a training environment for users to effectively improve their debate skills and enables flexible responses according to the user's emotional state.

[0959] The debate system of the present invention is composed of a server and terminals. Specifically, the following process is performed.

[0960] The server uses generative artificial intelligence to generate chatbots for debates based on the debate topic and settings received from the user. The generated chatbots are configured to have different opinions and personalities. A means is provided for having the chatbots debate and recording the content of the debates.

[0961] The device uses an emotion recognition engine to recognize users' emotions in real time during debates. The emotion recognition engine detects the user's emotional state by capturing and analyzing data such as facial expressions and voice tones.

[0962] The server adjusts the chatbot's responses appropriately based on data from the emotion engine. For example, if the user is feeling stressed, the chatbot's responses will be gentler. It also determines the outcome of the debate based on the content of the debate, and generates and provides evaluations and feedback to the user.

[0963] Next, we will explain an example of a food delivery application. This system uses an emotion recognition engine to suggest optimal menus and ordering methods based on the user's emotional state in a food delivery service.

[0964] The device captures the user's voice and image data and analyzes the user's emotions using an emotion recognition engine such as EmotionRecognition. It then uses MenuGenerator to generate an optimal menu based on the analyzed emotion data. The generated menu is then presented to the user by a chatbot, which responds based on the user's emotions, making it possible to suggest the most suitable dishes for the user. The device can then finalize the order using the OrderSystem.

[0965] As a specific example, the following steps can be considered.

[0966] 1. A user asks, "What's the recommendation of the day?"

[0967] 2. The emotion recognition engine analyzes the user's voice and facial expressions to detect when the user is feeling stressed.

[0968] 3. Menu Generator suggests dishes that have a relaxing effect.

[0969] 4. The chatbot suggests, "How about some relaxing green tea cookies and chamomile tea?"

[0970] An example of a prompt for the generative AI model is as follows:

[0971] The user's name is "Tanaka." Please suggest some dishes that would be suitable for Tanaka when he is feeling stressed.

[0972] For example: "How about some relaxing green tea cookies and chamomile tea?"

[0973] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0974] Step 1:

[0975] The user sends the debate topic and settings to the server via the terminal. The data input to the terminal includes the debate topic, setting details, and user identification information. For example, input data might include "Today's debate topic is environmental issues," "Expanding the use of renewable energy," and "Improving the efficiency of the current energy supply system."

[0976] Step 2:

[0977] The server uses generative artificial intelligence to generate chatbots for debate. In this process, the server inputs the theme and settings received from the user and generates Chatbot A and Chatbot B, each with different opinions and personalities. For example, Chatbot A will be a character that advocates "expanding the use of renewable energy," while Chatbot B will be a character that advocates "improving the efficiency of the current energy supply system."

[0978] Step 3:

[0979] The device captures audio and video data to recognize users' emotions in real time during a debate. This data is sent from the device to an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to generate emotion data. The captured data is the audio and video of the user's facial expressions as they speak during the debate.

[0980] Step 4:

[0981] The server receives emotional data from the emotion recognition engine and adjusts the chatbot's response based on that data. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a gentle tone. The input data is the emotional data sent from the emotion recognition engine, and the output is the adjusted chatbot's response.

[0982] Step 5:

[0983] The server initiates a dialogue session between Chatbot A and Chatbot B and records their statements. The server then allows Chatbot A to make the first statement, followed by Chatbot B's rebuttal. For example, Chatbot A may say, "Renewable energy is an important means of preventing global warming," and Chatbot B may then rebut, "Renewable energy is expensive and inefficient with current technology."

[0984] Step 6:

[0985] The server determines the winner based on the content of the debate. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is the winner. The input data are the content of the debate, and the output is the overall score and the winner's result.

[0986] Step 7:

[0987] The server generates evaluations and feedback along with the outcome of the debate and sends them to the device. The device then displays the debate results and feedback to the user. Based on the winner of the debate, feedback such as "Chatbot B is the winner. It was praised for emphasizing the cost issue of renewable energy" is presented.

[0988] Step 8:

[0989] The device stores the user's practice history in a database. This history records the results and feedback of debate sessions, and the user can refer to it later. The database stores the user's debate topic, win / loss results, score, feedback, etc.

[0990] This process flow allows users to effectively improve their debating skills. Furthermore, the use of an emotion recognition engine allows for flexible responses according to the user's emotional state, providing a comfortable training environment.

[0991] 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.

[0992] 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.

[0993] 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.

[0994] [Third embodiment]

[0995] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0996] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0997] 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).

[0998] 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.

[0999] 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.

[1000] 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).

[1001] 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.

[1002] 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.

[1003] 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.

[1004] 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.

[1005] 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.

[1006] 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."

[1007] MODE FOR CARRYING OUT THE INVENTION

[1008] This invention is a debate system using generative AI, and aims to enable users to efficiently improve their debating skills. This system has functions such as generating chatbots using generative AI, conducting debates between chatbots, analyzing the content of the debate and determining the winner, and providing feedback to users.

[1009] Program processing explanation

[1010] Chatbot generation

[1011] Server operations

[1012] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (e.g., a large-scale language model) to generate chatbot A and chatbot B to participate in a one-on-one debate. Each chatbot is given different opinions and personalities.

[1013] Conducting the debate

[1014] Server operations

[1015] 1. The server instructs Chatbot A and Chatbot B to start a conversation session.

[1016] 2. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to combat global warming."

[1017] 3. Chatbot B then takes over and offers a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practical for widespread adoption."

[1018] 4. This process is repeated over multiple turns as the debate progresses.

[1019] Judging the winner

[1020] Server operations

[1021] 1. After the debate is over, the server analyzes all statements made between the chatbots based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[1022] 2. A score is assigned to each statement, and finally an overall score is calculated.

[1023] 3. The server decides which chatbot is the winner based on the overall score.

[1024] Providing feedback

[1025] Operations performed by the server and the terminal

[1026] 1. The server generates ratings and sends feedback to the user's device.

[1027] 2. The device displays the results of the debate along with feedback, including specific comments on what was praised and areas for improvement.

[1028] Save practice history

[1029] Operations performed by the device

[1030] 1. The device stores the user's practice history in a database, which records the results and feedback of each debate session and can be referenced later.

[1031] Specific examples

[1032] 1. Starting example

[1033] In order for users to hold a debate on the theme of "environmental issues," two chatbots are set up on their devices: Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[1034] After setting up, press the "Start Debate" button.

[1035] 2. Example of debate progression

[1036] The server generates Chatbot A and Chatbot B.

[1037] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1038] Chatbot B responds by saying that renewable energy is expensive and inefficient with current technology.

[1039] This process is repeated multiple times.

[1040] 3. Example of results

[1041] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[1042] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "I was impressed by your emphasis on the cost issue of renewable energy."

[1043] By being equipped with these processes and functions, the debate system of the present invention can provide a training environment for users to effectively improve their debate skills.

[1044] The processing flow will be explained below.

[1045] Step 1:

[1046] User-defined themes

[1047] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[1048] Step 2:

[1049] The server generates the chatbot

[1050] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[1051] Step 3:

[1052] The server starts the debate

[1053] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1054] Step 4:

[1055] Chatbot conversations are on the rise

[1056] The server passes the turn to Chatbot B, who then counters by saying, "However, renewable energy is expensive and inefficient with current technology." This process is repeated multiple times, and the debate progresses until the specified number of turns or time is reached.

[1057] Step 5:

[1058] The server analyzes the debate

[1059] Once the debate is over, the server analyzes the content of each side's comments and assigns a score to each comment based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[1060] Step 6:

[1061] The server generates the results and feedback

[1062] The server generates the debate results and detailed feedback, including which chatbot won and how it performed across each criterion. Specific feedback includes points of merit and areas for improvement.

[1063] Step 7:

[1064] The terminal displays the results

[1065] The device receives the debate results and feedback sent from the server. The device displays to the user "Debate Results: Winner - Chatbot B" and also displays feedback such as "Your emphasis on the cost issue of renewable energy was highly praised."

[1066] Step 8:

[1067] The device stores your practice history

[1068] The device stores the user's practice history in a database, including the results and feedback from each debate session, for future reference.

[1069] Example 1

[1070] 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."

[1071] Conventional debate training systems have the problem that participants must find a real opponent and it is difficult to receive a fair evaluation. Furthermore, evaluation of logic and persuasiveness during a debate tends to be subjective, making it difficult to obtain concrete feedback. Furthermore, there is a lack of storage and utilization of training history to enable users to check their own progress. It is necessary to provide a system that solves these problems and efficiently improves debate skills.

[1072] 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.

[1073] In this invention, the server includes means for generating dialogue agents for debate using generative artificial intelligence, means for conducting dialogue sessions between the generated dialogue agents, means for determining victory or defeat based on the content of the dialogue between the generated dialogue agents, and means for providing feedback to the user based on the results and analysis of the dialogue session. This allows users to efficiently improve their debating skills without having to find a real opponent. Furthermore, users can receive objective evaluations of their logic and persuasiveness, and receive feedback on specific areas for improvement. Furthermore, the user's training history can be saved and referenced later, making it easier to monitor their own progress.

[1074] "Generative AI" refers to AI systems that can generate natural language based on input prompts, such as large-scale language models.

[1075] A "conversational agent" is a virtual person or character that is generated to interact with a user or another agent. In a debate session, characters with different opinions and personalities are generated.

[1076] A "dialogue session" refers to a series of dialogues between multiple dialogue agents. It is conducted in a debate format, with each agent taking turns speaking.

[1077] "Server" refers to a computer system that processes user input, generates interactive agents using generative artificial intelligence, and manages the interactive session.

[1078] "Means for determining victory or defeat" refers to algorithms or programs that analyze the content of a dialogue session and determine victory or defeat based on evaluation criteria such as logic, persuasiveness, and effectiveness of counterarguments.

[1079] "Feedback" means ratings and comments provided to a User based on the results and analysis of an Interactive Session, including suggestions for improvement and specific advice.

[1080] "Training History" refers to the data that records the results and feedback of a user's debate sessions, and is used to monitor the user's progress.

[1081] MODE FOR CARRYING OUT THE INVENTION

[1082] This invention is a debate system using generative AI, which aims to enable users to efficiently improve their debating skills. This system has functions such as generating dialogue agents using generative AI, running dialogue sessions between dialogue agents, analyzing the content of the dialogue and determining the winner, and providing feedback to the user.

[1083] Chatbot generation

[1084] Server operations

[1085] 1. The user inputs the debate topic and detailed settings via the terminal. A possible debate topic would be "expanding the use of renewable energy."

[1086] 2. The device sends the entered debate topic and detailed settings to the server.

[1087] 3. Based on the received information, the server creates and sends the following prompt to a generative artificial intelligence (e.g., a large-scale language model such as GPT-3):

[1088] Create two chatbots, each with a different perspective on the debate topic "Expanding the Use of Renewable Energy."

[1089] Chatbot A should support "expanding the use of renewable energy," and Chatbot B should support "improving the efficiency of the current energy supply system."

[1090] 4. Generative AI generates Chatbot A and Chatbot B based on the prompt, giving each one different voice and personality.

[1091] Conducting the debate

[1092] Server operations

[1093] 1. The server issues a command to start a dialogue session, and Chatbot A makes the first statement. For example, Chatbot A says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[1094] 2. The server records what Chatbot A says.

[1095] 3. Next, the server passes the turn to Chatbot B, who then makes a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practically deployed on a wide scale."

[1096] 4. Repeat this process multiple times to progress the debate.

[1097] Judging the winner

[1098] Server operations

[1099] 1. After the debate is over, the server analyzes all statements made between the agents based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[1100] 2. Assign a score to each statement and calculate an overall score.

[1101] 3. The server decides which conversational agent is the winner based on the total score.

[1102] Providing feedback

[1103] Operations performed by the server and the terminal

[1104] 1. The server generates ratings and sends feedback to the user's device.

[1105] 2. The device displays the results of the interactive session along with feedback, including specific comments about what was appreciated and areas for improvement. For example, a specific comment such as "I was impressed by your emphasis on the cost issues of renewable energy" may be provided.

[1106] Save practice history

[1107] Operations performed by the device

[1108] 1. The device stores the practice history, including the results and feedback of each debate session, in a database. This history records the results and feedback of each debate session and can be referenced later by the user.

[1109] This system allows users to efficiently improve their debating skills without having to find a real opponent. It also allows users to receive objective evaluations of their logic and persuasiveness, and provides feedback on specific areas for improvement. Furthermore, the system saves users' training history and allows them to refer to it later, making it easier to monitor their progress.

[1110] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1111] Step 1:

[1112] Entering themes and settings

[1113] The user inputs the debate topic and detailed settings through the terminal.

[1114] [Input] Debate topic and details entered by the user.

[1115] [Operation] For example, on the topic of "expanding the use of renewable energy," you can set Chatbot A to support it and Chatbot B to oppose it.

[1116] [Output] The terminal sends the input data (debate topic and detailed settings) to the server.

[1117] Step 2:

[1118] Creating a prompt statement

[1119] Based on the debate topic and settings received by the server, a prompt is created to be sent to the generative AI model.

[1120] [Input] Debate topic and detailed settings received from the device.

[1121] [Behavior] The server creates a prompt. For example, "Create two chatbots with different viewpoints on the debate topic 'Expanding the use of renewable energy.' Chatbot A should 'support expanding the use of renewable energy,' and chatbot B should 'support improving the efficiency of the current energy supply system.'"

[1122] [Output] Send the prompt sentence to the generative AI model.

[1123] Step 3:

[1124] Chatbot generation

[1125] The server generates chatbot A and chatbot B using the generative AI model.

[1126] [Input] The prompt sent to the generative AI model.

[1127] [How it works] The generative AI model generates Chatbot A and Chatbot B based on the prompt text. Each is given different opinions and personalities.

[1128] [Output] Two chatbots (A and B) are generated.

[1129] Step 4:

[1130] Debate session begins

[1131] The server issues a command to start a dialogue session, and Chatbot A makes its first statement.

[1132] [Input] Two generated chatbots (A and B).

[1133] [Operation] The server starts the internal debate module, and Chatbot A makes a statement. For example, it says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[1134] [Output] Chatbot A's statement.

[1135] Step 5:

[1136] Recording statements and transitioning to rebuttals

[1137] The server records what Chatbot A says and then passes the turn to Chatbot B.

[1138] [Input] Statement made by Chatbot A.

[1139] [Operation] The server records what Chatbot A says and passes the turn of rebuttal to Chatbot B. Chatbot B then makes a rebuttal. For example, it may say, "The cost of renewable energy is too high, making widespread adoption unrealistic."

[1140] [Output] Chatbot B's rebuttal.

[1141] Step 6:

[1142] Repeated interactive sessions

[1143] The above process of commenting and rebutting is repeated multiple times (for example, five times) to progress the debate.

[1144] [Input] Respective statements and rebuttals from Chatbot A and Chatbot B.

[1145] [Operation] The server records the comments and rebuttals for each turn and moves on to the next turn.

[1146] [Output] The complete debate session log.

[1147] Step 7:

[1148] Judging the winner

[1149] The server analyzes the debate session log and determines the winner.

[1150] [Input] The complete debate session log.

[1151] [How it works] The server analyzes the debate content and generates scores based on logic, persuasiveness, and effectiveness of counterarguments. It assigns a score to each statement, calculates the total score, and determines the winner.

[1152] [Output] Win / loss result.

[1153] Step 8:

[1154] Generate and send feedback

[1155] The server generates feedback based on the results and sends it to the user's terminal.

[1156] [Input] The result of the match and the score for each comment.

[1157] [Operation] The server generates an overall rating and specific feedback (for example, "I was impressed by the emphasis on the cost issue of renewable energy") and sends it to the user's device.

[1158] [Output] Feedback sent to the device.

[1159] Step 9:

[1160] Save practice history

[1161] The terminal stores the practice history, including the results and feedback of the debate session, in a database.

[1162] [Input] Feedback and results sent to the device.

[1163] [Operation] The device saves the feedback and results as practice history in a database.

[1164] [Output] Saved practice history data.

[1165] (Application example 1)

[1166] 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."

[1167] In recent years, there has been a demand for staff to improve their skills in customer service to improve customer satisfaction. However, training that simulates real-life customer service situations is costly and time-consuming. To solve this problem, an efficient and effective staff training system is needed.

[1168] 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.

[1169] In this invention, the server includes means for generating chatbots for debate using generative artificial intelligence, means for having the generated chatbots debate, means for determining the outcome based on the content of the debate, means for providing feedback to the user based on the training history, means for generating chatbots with different roles based on selected topics, means for running debates that simulate customer service scenes, and means for providing specific feedback for improvement based on the analysis results, thereby enabling training to efficiently improve customer service skills for staff at brick-and-mortar stores.

[1170] "Generative AI" is an AI technology that uses algorithms such as large-scale language models to generate new data and content.

[1171] A "chatbot" is an autonomous program that can conduct conversations based on specific tasks.

[1172] A "debate" is an activity in which opinions are exchanged and discussed on a specific topic.

[1173] The "method of determining the winner" is a method of analyzing the content of the debate and determining the winner based on logic and persuasiveness.

[1174] "Training history" is data that records the content and results of debate sessions that a user has conducted.

[1175] "Feedback" refers to specific advice and improvements provided to users based on an analysis of the debate results.

[1176] A "topic" is a theme or subject that will be addressed in a debate or simulation.

[1177] A "role" is the position or stance that each chatbot takes in the debate.

[1178] "Customer service scene" refers to the situation in which customers are treated and served in a physical store.

[1179] The "analysis results" are the results of analyzing the content of the debate based on the evaluation criteria.

[1180] "Improvement feedback" refers to specific improvement measures and suggestions provided to users based on the analysis results.

[1181] This invention is a training system for improving the customer service skills of store staff, and is realized using generative artificial intelligence. Specific embodiments of this system are described below.

[1182] Program Structure

[1183] The system consists of three main components: a server, a terminal, and a user.

[1184] Hardware and Software

[1185] Hardware: Smartphones, tablets (e.g., iOS, Android devices)

[1186] software:

[1187] Large-scale language models (e.g., GPT-4)

[1188] Database management system (MySQL, etc.)

[1189] Network communication protocol (REST API)

[1190] Processing flow

[1191] 1. User operations

[1192] The user launches the application on a smartphone or tablet and begins training.

[1193] The user selects a specific customer service topic (e.g., handling complaints, explaining new products, responding to customer requests, etc.).

[1194] 2. Server Operation

[1195] Based on the topic and settings received from the user, the server uses generative artificial intelligence (such as GPT-4) to generate Chatbot A and Chatbot B to participate in the debate.

[1196] Chatbot A and Chatbot B are assigned different roles (e.g., customer role, staff role).

[1197] 3. Conducting the debate

[1198] The server issues a command to start a debate session, and Chatbot A (the customer) makes the first statement, for example, "This product was defective, and I would like a refund."

[1199] Next, Chatbot B (playing the role of staff) responds, for example, by saying, "I'm sorry. We'll take the time to hear more details and provide an appropriate response."

[1200] This exchange is repeated for multiple turns.

[1201] 4. Analysis of results and feedback

[1202] Once the debate is over, the server analyzes all statements based on logic, persuasiveness, and effectiveness of counterarguments.

[1203] A score is assigned to each statement and a final overall score is calculated.

[1204] The server generates feedback based on the overall score and provides it to the user, including specific advice such as, "Your initial apology was timely, but you didn't ask detailed questions enough."

[1205] 5. History Storage

[1206] The server stores the user's training history in a database, which records the results and feedback of each debate session and allows the user to refer to it later.

[1207] Specific examples

[1208] For example, if a user selects the topic "Handling Complaints" and presses the "Start Debate" button, the system will send the following prompt to the server:

[1209] Theme: Handling complaints

[1210] Setting: A customer is dissatisfied with a product, and staff members handle the complaint appropriately.

[1211] As a result, Chatbot A will say, "This product was defective," and Chatbot B will respond, "Sorry. What was the problem?" This debate will continue for several turns, and finally feedback will be provided.

[1212] This system will enable store staff to efficiently and effectively improve their customer service skills.

[1213] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1214] Step 1:

[1215] A user launches the application using a smartphone or tablet and selects a specific customer service topic.

[1216] Input: User selects a topic (e.g., complaint handling) in the application.

[1217] Output: Selected topics and configuration information are sent to the server

[1218] Specific operation: The user selects "Complaint handling" on the app screen and presses the "Start debate" button.

[1219] Step 2:

[1220] The server uses generative artificial intelligence to generate chatbot A and chatbot B to participate in the debate based on the topic and settings received from the user.

[1221] Input: Topic and preference information received from the user

[1222] Output: Generated Chatbot A and Chatbot B

[1223] Specific operation: The server calls a large-scale language model (such as GPT-4) to generate chatbot A for the customer role and chatbot B for the staff role.

[1224] Step 3:

[1225] The server initiates the debate session, and Chatbot A makes the first statement.

[1226] Input: Generated Chatbot A and Chatbot B

[1227] Output: Chatbot A's first statement

[1228] Specific operation: The server makes Chatbot A say something like, "This product was defective, I would like a refund."

[1229] Step 4:

[1230] Chatbot B then makes a rebuttal.

[1231] Input: Chatbot A's statement

[1232] Output: Chatbot B's rebuttal

[1233] Specific operation: The server makes Chatbot B say something like, "I'm sorry. We will ask for more details and take appropriate action."

[1234] Step 5:

[1235] This debate is repeated for multiple turns.

[1236] Input: What Chatbot A and B say in each turn

[1237] Output: A dialogue log of the entire debate

[1238] Specific operation: The server manages the dialogue between chatbots and records each utterance.

[1239] Step 6:

[1240] Once the debate is over, the server analyzes all statements.

[1241] Input: Dialogue log of the entire debate

[1242] Output: Analysis results and scores for each utterance

[1243] Specific operation: The server analyzes the content of the statement and scores it based on logic, persuasiveness, and effectiveness of the counterargument.

[1244] Step 7:

[1245] The server generates an overall score and feedback and provides it to the user.

[1246] Input: Analysis results and scores for each statement

[1247] Output: Overall score and feedback message

[1248] Specific behavior: The server generates specific feedback such as "The timing of the initial apology was appropriate, but the question was not detailed enough" and sends it to the user's device.

[1249] Step 8:

[1250] The server stores the user's training history in a database.

[1251] Input: Results and feedback from the debate session

[1252] Output: Saved training history

[1253] Specific operation: The server saves the results and feedback of this debate session in a database for future reference.

[1254] 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.

[1255] MODE FOR CARRYING OUT THE INVENTION

[1256] This invention is a system that combines generative AI and an emotion engine to build a debate system that provides debate practice that takes into account the user's emotions. This system is composed of a chatbot generated by generative AI and an emotion engine that recognizes the user's emotions.

[1257] Program processing explanation

[1258] Chatbot generation

[1259] Server operations

[1260] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence to generate chatbot A and chatbot B. These chatbots are configured to have different opinions and personalities.

[1261] Emotion Recognition in Action

[1262] Operations performed by the device

[1263] 1. While users are debating, the device captures the data necessary for emotion recognition (e.g., facial expressions, voice tone, etc.) in real time.

[1264] 2. The device sends the captured data to the emotion engine, which analyzes the user's emotions.

[1265] Conducting the debate

[1266] Server operations

[1267] 1. The server sends a command to start a dialogue session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to prevent global warming."

[1268] Server operations

[1269] 1. The server passes the turn to Chatbot B. Chatbot B counters, saying, "Renewable energy is expensive and inefficient with current technology."

[1270] Emotion-Based Response Modulation

[1271] Server operations

[1272] 1. The server receives emotional data from the emotion engine and adjusts the chatbot's responses accordingly based on the user's emotional state. For example, if the user is feeling stressed, the chatbot's responses will be made gentler.

[1273] Judging the winner

[1274] Server operations

[1275] 1. After the debate is over, the server analyzes the content of the debate statements. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of counterarguments, etc.). The total score is calculated to determine which chatbot is the winner.

[1276] Providing feedback

[1277] Operations performed by the server and terminal

[1278] 1. The server generates ratings and sends feedback to the device.

[1279] 2. The device displays the debate results along with feedback, including what went well and what needs improvement.

[1280] Save practice history

[1281] Operations performed by the device

[1282] 1. The device stores the user's practice history in a database, which records the results and feedback of the debate session and can be referenced later.

[1283] Specific examples

[1284] 1. Starting example

[1285] To allow users to debate on the topic of "environmental issues" on their devices, chatbot A is set to advocate for "expanding the use of renewable energy," and chatbot B is set to advocate for "improving the efficiency of the current energy supply system."

[1286] After setting up, press the "Start Debate" button.

[1287] 2. Example of debate progression

[1288] The server generates Chatbot A and Chatbot B.

[1289] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1290] Chatbot B counters, "Renewable energy is expensive and inefficient with current technology."

[1291] This process is repeated multiple times.

[1292] Meanwhile, the emotion engine recognizes the user's emotions, and the server adjusts the chatbot's responses based on that data.

[1293] 3. Example of results

[1294] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[1295] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "Your emphasis on the cost issue of renewable energy was appreciated."

[1296] The analysis results from the emotion engine are also displayed, providing feedback on the user's emotional changes and the chatbot's corresponding adjustment history.

[1297] By incorporating these processes and functions, the debate system of the present invention provides a training environment for users to effectively improve their debate skills, and also enables flexible responses that take emotions into consideration.

[1298] The processing flow will be explained below.

[1299] MODE FOR CARRYING OUT THE INVENTION

[1300] Processing Steps

[1301] Step 1:

[1302] User-defined themes

[1303] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[1304] Step 2:

[1305] The server generates the chatbot

[1306] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[1307] Step 3:

[1308] The device captures emotional data

[1309] While the user is preparing for the debate, the device captures the user's emotional data (e.g., facial expressions, voice tone, heart rate, etc.) in real time. The emotional data is sent from the device to the emotion engine.

[1310] Step 4:

[1311] Emotion engine analyzes emotions

[1312] The emotion engine analyzes the user's emotions based on the captured data, and the analysis results include information such as whether the user is stressed, relaxed, or focused.

[1313] Step 5:

[1314] Start a debate

[1315] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1316] Step 6:

[1317] The chatbot dialogue progresses

[1318] The server passes the turn to Chatbot B, who then counters by saying that renewable energy is expensive and inefficient with current technology. This process is repeated multiple times, and the debate continues until the specified number of turns or time has been reached.

[1319] Step 7:

[1320] Emotion-Based Response Modulation

[1321] The server receives the analysis results from the emotion engine and adjusts the chatbot's response accordingly depending on the user's emotional state. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a calm tone and content.

[1322] Step 8:

[1323] Debate analysis and winner determination

[1324] After the debate is over, the server analyzes the content of the debate statements and assigns a score to each statement based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[1325] Step 9:

[1326] Generate results and feedback

[1327] The server generates the debate results and detailed feedback, including points of merit and areas for improvement, as well as feedback on the user's emotional changes.

[1328] Step 10:

[1329] The terminal displays the results

[1330] The device receives the results and feedback sent from the server. The device displays "Debate Results: Winner - Chatbot B" to the user and provides feedback such as "Your emphasis on the cost issue of renewable energy was highly praised." The device also displays the analysis results from the emotion engine, showing feedback on changes in the user's emotions and the chatbot's corresponding adjustments.

[1331] Step 11:

[1332] The device stores your practice history

[1333] The device stores the user's practice history in a database, which includes the results and feedback of each debate session and can be referenced later.

[1334] This system allows users to effectively improve their debating skills and also allows training that takes into account their own emotional state.

[1335] Example 2

[1336] 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."

[1337] While conventional debate systems can provide a training environment for improving technical debate skills, it is difficult to consider the user's emotional state. Furthermore, since users' emotions often influence the discussion in real debates, training that ignores emotions is inefficient. Furthermore, there is a problem that feedback on debate results is insufficient, making it difficult to contribute to the improvement of users' skills.

[1338] 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.

[1339] In this invention, the server includes means for generating artificial dialogue agents for debate using generative artificial intelligence, means for having the generated artificial dialogue agents debate, means for analyzing the emotional state of the user during the debate using an emotion engine that recognizes the user's emotions, means for adjusting the responses of the artificial dialogue agents based on the emotional state of the user, means for determining the outcome of the debate based on the content of the debate, and means for providing feedback to the user based on the practice history, thereby making it possible to effectively improve the debating skills while taking into consideration the emotional state of the user.

[1340] "Generative artificial intelligence" is an algorithm that learns from large amounts of data and generates new text.

[1341] An "artificial dialogue agent" is software that can carry out pre-programmed dialogues or dialogues that are dynamically generated using generative artificial intelligence.

[1342] A "debate" is an act of logical discussion between people with different positions or opinions on a particular topic.

[1343] The "emotion engine" is a technology that analyzes data such as the user's voice and facial expressions to grasp their emotional state in real time.

[1344] The "means of determining victory or defeat" is a system that analyzes the content of the debate, assigns points based on evaluation criteria such as logic, persuasiveness, and the effectiveness of the rebuttal, and determines which side is superior.

[1345] "Feedback" refers to evaluations and comments provided after a debate, providing users with information to understand their strengths and areas for improvement.

[1346] "Practice history" is a record of past debate sessions conducted by the user, including debate content, evaluations, feedback, and the like.

[1347] A "topic" is a particular topic or issue that is the subject of debate.

[1348] "Settings" are parameters that the user specifies regarding the progress and conditions of the debate.

[1349] This invention provides a debate system that combines generative AI and an emotion engine. This system allows generated chatbots to debate with each other and provides training while taking into account the user's emotions. This system is composed of the following main components:

[1350] 1. Chatbot Creation

[1351] Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (such as GPT-3) to generate Chatbot A and Chatbot B. These chatbots are configured to have different opinions and personalities. Specific examples of prompts include "Please state your opinion on the topic: Expand the use of renewable energy" and "Improve the efficiency of the current energy supply system."

[1352] 2. Emotion Recognition

[1353] The device captures the user's facial expressions with a webcam and collects their voice tone with a microphone while they debate. This data is sent to an emotion engine (e.g., emotion recognition API) and analyzed in real time. The analysis results are sent to a server and reflected in the progress of the debate.

[1354] 3. Conducting the debate

[1355] The server starts a dialogue session between Chatbot A and Chatbot B. Chatbot A makes an initial statement, followed by a rebuttal from Chatbot B. This exchange is repeated multiple times.

[1356] 4. Emotion-Based Response Modulation

[1357] The server receives the emotion data from the emotion engine and adjusts the chatbot's responses based on the user's emotional state, for example, changing the chatbot's tone to be calmer if the user is stressed.

[1358] 5. Judging the Winner

[1359] After the debate is over, the server analyzes the content of the debate and assigns scores based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is superior.

[1360] 6. Providing Feedback

[1361] The server sends the generated evaluation and feedback to the device, which displays it to the user, providing feedback on what was good and what needs improvement. The analysis results from the emotion engine are also displayed.

[1362] 7. Save your practice history

[1363] The device stores the user's practice history in a database, which includes the results and feedback of the debate session. Users can later refer to this history to help them self-evaluate and improve their skills.

[1364] Through the above process, the present invention provides a training environment for users to effectively improve their debating skills. In particular, emotion recognition and response adjustment by the emotion engine enables flexible responses that take into account the user's emotional state. As a result, it is possible to provide higher satisfaction and effectiveness than conventional debate systems.

[1365] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1366] Step 1: Receiving the debate topic and detailed settings

[1367] server

[1368] Input: Receive debate topic and detailed settings from user.

[1369] Data processing: Analyzing received information and classifying themes and settings.

[1370] Output: Parsed themes and settings are sent to the generative AI.

[1371] Specific operation: The user enters the debate topic "environmental issues" and detailed settings into the terminal, and the server receives and analyzes this.

[1372] Step 2: Initializing the generative AI and creating a prompt

[1373] server

[1374] Input: Parsed themes and settings.

[1375] Data processing: Initialize a generative artificial intelligence (e.g., GPT-3) and create a prompt based on the theme.

[1376] Output: Send the prompt to the generative artificial intelligence model.

[1377] Specific operation: The server creates prompt sentences for Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[1378] Step 3: Generate the chatbot

[1379] server

[1380] Input: Prompt statement.

[1381] Data processing: Chatbot A and Chatbot B are generated using generative artificial intelligence.

[1382] Output: Saves the generated chatbot information.

[1383] Specific operation: The server uses GPT-3 to generate two different chatbots based on the prompt sentence.

[1384] Step 4: Capture emotion recognition data

[1385] Terminal

[1386] Input: User's facial expressions, voice tone.

[1387] Data processing: Capture data using a webcam and microphone.

[1388] Output: Sends the captured data to the emotion engine.

[1389] Specific operation: The device captures the user's facial expressions with a webcam and collects voice tones with a microphone.

[1390] Step 5: Analyze the sentiment data

[1391] Terminal

[1392] Input: Captured emotion data.

[1393] Data processing: Analyze using an emotion engine (e.g., emotion recognition API).

[1394] Output: Send the analysis results to the server.

[1395] Specific operation: The emotion engine analyzes the user's stress level, joy, anger, etc. and sends the results to the server.

[1396] Step 6: Start the debate session

[1397] server

[1398] Input: Information for the generated chatbot.

[1399] Data processing: Initialize the conversation session between Chatbot A and Chatbot B.

[1400] Output: Send the first command to Chatbot A.

[1401] Specific operation: The server makes chatbot A say, "We need to expand the use of renewable energy."

[1402] Step 7: Continue the conversation

[1403] server

[1404] Input: What Chatbot A says.

[1405] Data processing: Send a counterargument to Chatbot B.

[1406] Output: Record what Chatbot B said.

[1407] Specific behavior: The server has Chatbot B respond by saying, "Renewable energy is too expensive right now."

[1408] Step 8: Emotion-Based Response Adjustment

[1409] server

[1410] Input: Analysis results from the emotion engine.

[1411] Data manipulation: Tailoring chatbot responses based on the user's emotional state.

[1412] Output: Sends the tailored response to the chatbot.

[1413] Specific behavior: If the user is feeling stressed, the server changes the chatbot's tone to be gentler.

[1414] Step 9: Ending the debate and deciding who won

[1415] server

[1416] Input: The entire debate.

[1417] Data processing: Generate scores based on logic, persuasiveness, and effectiveness of counterarguments.

[1418] Output: Determine winner based on overall score.

[1419] Specific operation: The server determines the "Total score: Chatbot B is the winner" and displays the reason.

[1420] Step 10: Provide feedback

[1421] Servers and Terminals

[1422] Input: Debate evaluation and feedback.

[1423] Data processing: Generate ratings and feedback and send them to your device.

[1424] Output: Display feedback to the user.

[1425] Specific operation: The server generates feedback based on the evaluation, and the device displays something like, "Your emphasis on the cost issue of renewable energy was appreciated."

[1426] Step 11: Save your practice history

[1427] Terminal

[1428] Input: Results, evaluations, and feedback from the debate session.

[1429] Data processing: Practice history is saved in a database.

[1430] Output: Makes the history available to the user.

[1431] What it does: The device stores the debate results and feedback in a database so that the user can refer to them later.

[1432] (Application example 2)

[1433] 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."

[1434] Conventional debate training systems tend to engage in monotonous dialogue without considering the user's emotional state, making it difficult to provide a flexible training environment that can respond to changes in emotions. Furthermore, they are unable to provide appropriate feedback that responds immediately to the user's stress or emotional changes, which increases the burden on the user. Furthermore, because applications to other fields such as food delivery have not been considered, there is a lack of application of emotion recognition in practical scenarios other than debates.

[1435] 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.

[1436] In this invention, the server includes a means for generating chatbots for debate using generative artificial intelligence, a means for conducting debates between the generated chatbots, a means for using an emotion recognition engine to recognize users' emotions in real time, a means for adjusting the chatbots' responses based on the users' emotions, a means for determining the outcome of the debate based on the content of the debate, and a means for providing feedback to the user based on the training history. This enables flexible and effective debate training that takes into account the user's emotional state. This system can also be applied to other fields such as food delivery, and can have practical applications such as an emotion-based ordering system.

[1437] "Generative AI" is AI that uses technology to automatically generate text and conversations based on given input data.

[1438] A "chatbot" is a computer program that can converse with humans in natural language, and is particularly used in interactive formats such as debates.

[1439] An "emotion recognition engine" is a technology that analyzes data such as voice and images to identify a user's emotional state in real time.

[1440] The "means for determining victory or defeat" is a function that analyzes the results of a dialogue or debate and determines which chatbot is superior based on the set evaluation criteria.

[1441] The "means for providing feedback" is a function for notifying the user of the results of the debate and areas for improvement, and for providing information to improve the effectiveness of training.

[1442] "Training history" is data that accumulates records and analysis results of debate sessions conducted by users.

[1443] The "means for adjusting the chatbot's response based on emotions" is a function that adjusts the chatbot to return an appropriate response based on the user's emotional data obtained by the emotion recognition engine.

[1444] This invention relates to a debate system that combines generative artificial intelligence and an emotion recognition engine. It provides a training environment for users to effectively improve their debate skills and enables flexible responses according to the user's emotional state.

[1445] The debate system of the present invention is composed of a server and terminals. Specifically, the following process is performed.

[1446] The server uses generative artificial intelligence to generate chatbots for debates based on the debate topic and settings received from the user. The generated chatbots are configured to have different opinions and personalities. A means is provided for having the chatbots debate and recording the content of the debates.

[1447] The device uses an emotion recognition engine to recognize users' emotions in real time during debates. The emotion recognition engine detects the user's emotional state by capturing and analyzing data such as facial expressions and voice tones.

[1448] The server adjusts the chatbot's responses appropriately based on data from the emotion engine. For example, if the user is feeling stressed, the chatbot's responses will be gentler. It also determines the outcome of the debate based on the content of the debate, and generates and provides evaluations and feedback to the user.

[1449] Next, we will explain an example of a food delivery application. This system uses an emotion recognition engine to suggest optimal menus and ordering methods based on the user's emotional state in a food delivery service.

[1450] The device captures the user's voice and image data and analyzes the user's emotions using an emotion recognition engine such as EmotionRecognition. It then uses MenuGenerator to generate an optimal menu based on the analyzed emotion data. The generated menu is then presented to the user by a chatbot, which responds based on the user's emotions, making it possible to suggest the most suitable dishes for the user. The device can then finalize the order using the OrderSystem.

[1451] As a specific example, the following steps can be considered.

[1452] 1. A user asks, "What's the recommendation of the day?"

[1453] 2. The emotion recognition engine analyzes the user's voice and facial expressions to detect when the user is feeling stressed.

[1454] 3. Menu Generator suggests dishes that have a relaxing effect.

[1455] 4. The chatbot suggests, "How about some relaxing green tea cookies and chamomile tea?"

[1456] An example of a prompt for the generative AI model is as follows:

[1457] The user's name is "Tanaka." Please suggest some dishes that would be suitable for Tanaka when he is feeling stressed.

[1458] For example: "How about some relaxing green tea cookies and chamomile tea?"

[1459] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1460] Step 1:

[1461] The user sends the debate topic and settings to the server via the terminal. The data input to the terminal includes the debate topic, setting details, and user identification information. For example, input data might include "Today's debate topic is environmental issues," "Expanding the use of renewable energy," and "Improving the efficiency of the current energy supply system."

[1462] Step 2:

[1463] The server uses generative artificial intelligence to generate chatbots for debate. In this process, the server inputs the theme and settings received from the user and generates Chatbot A and Chatbot B, each with different opinions and personalities. For example, Chatbot A will be a character that advocates "expanding the use of renewable energy," while Chatbot B will be a character that advocates "improving the efficiency of the current energy supply system."

[1464] Step 3:

[1465] The device captures audio and video data to recognize users' emotions in real time during a debate. This data is sent from the device to an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to generate emotion data. The captured data is the audio and video of the user's facial expressions as they speak during the debate.

[1466] Step 4:

[1467] The server receives emotional data from the emotion recognition engine and adjusts the chatbot's response based on that data. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a gentle tone. The input data is the emotional data sent from the emotion recognition engine, and the output is the adjusted chatbot's response.

[1468] Step 5:

[1469] The server initiates a dialogue session between Chatbot A and Chatbot B and records their statements. The server then allows Chatbot A to make the first statement, followed by Chatbot B's rebuttal. For example, Chatbot A may say, "Renewable energy is an important means of preventing global warming," and Chatbot B may then rebut, "Renewable energy is expensive and inefficient with current technology."

[1470] Step 6:

[1471] The server determines the winner based on the content of the debate. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is the winner. The input data are the content of the debate, and the output is the overall score and the winner's result.

[1472] Step 7:

[1473] The server generates evaluations and feedback along with the outcome of the debate and sends them to the device. The device then displays the debate results and feedback to the user. Based on the winner of the debate, feedback such as "Chatbot B is the winner. It was praised for emphasizing the cost issue of renewable energy" is presented.

[1474] Step 8:

[1475] The device stores the user's practice history in a database. This history records the results and feedback of debate sessions, and the user can refer to it later. The database stores the user's debate topic, win / loss results, score, feedback, etc.

[1476] This process flow allows users to effectively improve their debating skills. Furthermore, the use of an emotion recognition engine allows for flexible responses according to the user's emotional state, providing a comfortable training environment.

[1477] 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.

[1478] 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.

[1479] 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.

[1480] [Fourth embodiment]

[1481] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1482] 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.

[1483] 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).

[1484] 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.

[1485] 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.

[1486] 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).

[1487] 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.

[1488] 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.

[1489] 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.

[1490] 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.

[1491] 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.

[1492] 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.

[1493] 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."

[1494] MODE FOR CARRYING OUT THE INVENTION

[1495] This invention is a debate system using generative AI, and aims to enable users to efficiently improve their debating skills. This system has functions such as generating chatbots using generative AI, conducting debates between chatbots, analyzing the content of the debate and determining the winner, and providing feedback to users.

[1496] Program processing explanation

[1497] Chatbot generation

[1498] Server operations

[1499] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (e.g., a large-scale language model) to generate chatbot A and chatbot B to participate in a one-on-one debate. Each chatbot is given different opinions and personalities.

[1500] Conducting the debate

[1501] Server operations

[1502] 1. The server instructs Chatbot A and Chatbot B to start a conversation session.

[1503] 2. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to combat global warming."

[1504] 3. Chatbot B then takes over and offers a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practical for widespread adoption."

[1505] 4. This process is repeated over multiple turns as the debate progresses.

[1506] Judging the winner

[1507] Server operations

[1508] 1. After the debate is over, the server analyzes all statements made between the chatbots based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[1509] 2. A score is assigned to each statement, and finally an overall score is calculated.

[1510] 3. The server decides which chatbot is the winner based on the overall score.

[1511] Providing feedback

[1512] Operations performed by the server and the terminal

[1513] 1. The server generates ratings and sends feedback to the user's device.

[1514] 2. The device displays the results of the debate along with feedback, including specific comments on what was praised and areas for improvement.

[1515] Save practice history

[1516] Operations performed by the device

[1517] 1. The device stores the user's practice history in a database, which records the results and feedback of each debate session and can be referenced later.

[1518] Specific examples

[1519] 1. Starting example

[1520] In order for users to hold a debate on the theme of "environmental issues," two chatbots are set up on their devices: Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[1521] After setting up, press the "Start Debate" button.

[1522] 2. Example of debate progression

[1523] The server generates Chatbot A and Chatbot B.

[1524] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1525] Chatbot B responds by saying that renewable energy is expensive and inefficient with current technology.

[1526] This process is repeated multiple times.

[1527] 3. Example of results

[1528] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[1529] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "I was impressed by your emphasis on the cost issue of renewable energy."

[1530] By being equipped with these processes and functions, the debate system of the present invention can provide a training environment for users to effectively improve their debate skills.

[1531] The processing flow will be explained below.

[1532] Step 1:

[1533] User-defined themes

[1534] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[1535] Step 2:

[1536] The server generates the chatbot

[1537] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[1538] Step 3:

[1539] The server starts the debate

[1540] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1541] Step 4:

[1542] Chatbot conversations are on the rise

[1543] The server passes the turn to Chatbot B, who then counters by saying, "However, renewable energy is expensive and inefficient with current technology." This process is repeated multiple times, and the debate progresses until the specified number of turns or time is reached.

[1544] Step 5:

[1545] The server analyzes the debate

[1546] Once the debate is over, the server analyzes the content of each side's comments and assigns a score to each comment based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[1547] Step 6:

[1548] The server generates the results and feedback

[1549] The server generates the debate results and detailed feedback, including which chatbot won and how it performed across each criterion. Specific feedback includes points of merit and areas for improvement.

[1550] Step 7:

[1551] The terminal displays the results

[1552] The device receives the debate results and feedback sent from the server. The device displays to the user "Debate Results: Winner - Chatbot B" and also displays feedback such as "Your emphasis on the cost issue of renewable energy was highly praised."

[1553] Step 8:

[1554] The device stores your practice history

[1555] The device stores the user's practice history in a database, including the results and feedback from each debate session, for future reference.

[1556] Example 1

[1557] 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."

[1558] Conventional debate training systems have the problem that participants must find a real opponent and it is difficult to receive a fair evaluation. Furthermore, evaluation of logic and persuasiveness during a debate tends to be subjective, making it difficult to obtain concrete feedback. Furthermore, there is a lack of storage and utilization of training history to enable users to check their own progress. It is necessary to provide a system that solves these problems and efficiently improves debate skills.

[1559] 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.

[1560] In this invention, the server includes means for generating dialogue agents for debate using generative artificial intelligence, means for conducting dialogue sessions between the generated dialogue agents, means for determining victory or defeat based on the content of the dialogue between the generated dialogue agents, and means for providing feedback to the user based on the results and analysis of the dialogue session. This allows users to efficiently improve their debating skills without having to find a real opponent. Furthermore, users can receive objective evaluations of their logic and persuasiveness, and receive feedback on specific areas for improvement. Furthermore, the user's training history can be saved and referenced later, making it easier to monitor their own progress.

[1561] "Generative AI" refers to AI systems that can generate natural language based on input prompts, such as large-scale language models.

[1562] A "conversational agent" is a virtual person or character that is generated to interact with a user or another agent. In a debate session, characters with different opinions and personalities are generated.

[1563] A "dialogue session" refers to a series of dialogues between multiple dialogue agents. It is conducted in a debate format, with each agent taking turns speaking.

[1564] "Server" refers to a computer system that processes user input, generates interactive agents using generative artificial intelligence, and manages the interactive session.

[1565] "Means for determining victory or defeat" refers to algorithms or programs that analyze the content of a dialogue session and determine victory or defeat based on evaluation criteria such as logic, persuasiveness, and effectiveness of counterarguments.

[1566] "Feedback" means ratings and comments provided to a User based on the results and analysis of an Interactive Session, including suggestions for improvement and specific advice.

[1567] "Training History" refers to the data that records the results and feedback of a user's debate sessions, and is used to monitor the user's progress.

[1568] MODE FOR CARRYING OUT THE INVENTION

[1569] This invention is a debate system using generative AI, which aims to enable users to efficiently improve their debating skills. This system has functions such as generating dialogue agents using generative AI, running dialogue sessions between dialogue agents, analyzing the content of the dialogue and determining the winner, and providing feedback to the user.

[1570] Chatbot generation

[1571] Server operations

[1572] 1. The user inputs the debate topic and detailed settings via the terminal. A possible debate topic would be "expanding the use of renewable energy."

[1573] 2. The device sends the entered debate topic and detailed settings to the server.

[1574] 3. Based on the received information, the server creates and sends the following prompt to a generative artificial intelligence (e.g., a large-scale language model such as GPT-3):

[1575] Create two chatbots, each with a different perspective on the debate topic "Expanding the Use of Renewable Energy."

[1576] Chatbot A should support "expanding the use of renewable energy," and Chatbot B should support "improving the efficiency of the current energy supply system."

[1577] 4. Generative AI generates Chatbot A and Chatbot B based on the prompt, giving each one different voice and personality.

[1578] Conducting the debate

[1579] Server operations

[1580] 1. The server issues a command to start a dialogue session, and Chatbot A makes the first statement. For example, Chatbot A says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[1581] 2. The server records what Chatbot A says.

[1582] 3. Next, the server passes the turn to Chatbot B, who then makes a counterargument. For example, Chatbot B might say, "Renewable energy is too expensive to be practically deployed on a wide scale."

[1583] 4. Repeat this process multiple times to progress the debate.

[1584] Judging the winner

[1585] Server operations

[1586] 1. After the debate is over, the server analyzes all statements made between the agents based on criteria such as logic, persuasiveness, and the effectiveness of counterarguments.

[1587] 2. Assign a score to each statement and calculate an overall score.

[1588] 3. The server decides which conversational agent is the winner based on the total score.

[1589] Providing feedback

[1590] Operations performed by the server and the terminal

[1591] 1. The server generates ratings and sends feedback to the user's device.

[1592] 2. The device displays the results of the interactive session along with feedback, including specific comments about what was appreciated and areas for improvement. For example, a specific comment such as "I was impressed by your emphasis on the cost issues of renewable energy" may be provided.

[1593] Save practice history

[1594] Operations performed by the device

[1595] 1. The device stores the practice history, including the results and feedback of each debate session, in a database. This history records the results and feedback of each debate session and can be referenced later by the user.

[1596] This system allows users to efficiently improve their debating skills without having to find a real opponent. It also allows users to receive objective evaluations of their logic and persuasiveness, and provides feedback on specific areas for improvement. Furthermore, the system saves users' training history and allows them to refer to it later, making it easier to monitor their progress.

[1597] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1598] Step 1:

[1599] Entering themes and settings

[1600] The user inputs the debate topic and detailed settings through the terminal.

[1601] [Input] Debate topic and details entered by the user.

[1602] [Operation] For example, on the topic of "expanding the use of renewable energy," you can set Chatbot A to support it and Chatbot B to oppose it.

[1603] [Output] The terminal sends the input data (debate topic and detailed settings) to the server.

[1604] Step 2:

[1605] Creating a prompt statement

[1606] Based on the debate topic and settings received by the server, a prompt is created to be sent to the generative AI model.

[1607] [Input] Debate topic and detailed settings received from the device.

[1608] [Behavior] The server creates a prompt. For example, "Create two chatbots with different viewpoints on the debate topic 'Expanding the use of renewable energy.' Chatbot A should 'support expanding the use of renewable energy,' and chatbot B should 'support improving the efficiency of the current energy supply system.'"

[1609] [Output] Send the prompt sentence to the generative AI model.

[1610] Step 3:

[1611] Chatbot generation

[1612] The server generates chatbot A and chatbot B using the generative AI model.

[1613] [Input] The prompt sent to the generative AI model.

[1614] [How it works] The generative AI model generates Chatbot A and Chatbot B based on the prompt text. Each is given different opinions and personalities.

[1615] [Output] Two chatbots (A and B) are generated.

[1616] Step 4:

[1617] Debate session begins

[1618] The server issues a command to start a dialogue session, and Chatbot A makes its first statement.

[1619] [Input] Two generated chatbots (A and B).

[1620] [Operation] The server starts the internal debate module, and Chatbot A makes a statement. For example, it says, "Increasing the use of renewable energy is the most effective way to combat global warming."

[1621] [Output] Chatbot A's statement.

[1622] Step 5:

[1623] Recording statements and transitioning to rebuttals

[1624] The server records what Chatbot A says and then passes the turn to Chatbot B.

[1625] [Input] Statement made by Chatbot A.

[1626] [Operation] The server records what Chatbot A says and passes the turn of rebuttal to Chatbot B. Chatbot B then makes a rebuttal. For example, it may say, "The cost of renewable energy is too high, making widespread adoption unrealistic."

[1627] [Output] Chatbot B's rebuttal.

[1628] Step 6:

[1629] Repeated interactive sessions

[1630] The above process of commenting and rebutting is repeated multiple times (for example, five times) to progress the debate.

[1631] [Input] Respective statements and rebuttals from Chatbot A and Chatbot B.

[1632] [Operation] The server records the comments and rebuttals for each turn and moves on to the next turn.

[1633] [Output] The complete debate session log.

[1634] Step 7:

[1635] Judging the winner

[1636] The server analyzes the debate session log and determines the winner.

[1637] [Input] The complete debate session log.

[1638] [How it works] The server analyzes the debate content and generates scores based on logic, persuasiveness, and effectiveness of counterarguments. It assigns a score to each statement, calculates the total score, and determines the winner.

[1639] [Output] Win / loss result.

[1640] Step 8:

[1641] Generate and send feedback

[1642] The server generates feedback based on the results and sends it to the user's terminal.

[1643] [Input] The result of the match and the score for each comment.

[1644] [Operation] The server generates an overall rating and specific feedback (for example, "I was impressed by the emphasis on the cost issue of renewable energy") and sends it to the user's device.

[1645] [Output] Feedback sent to the device.

[1646] Step 9:

[1647] Save practice history

[1648] The terminal stores the practice history, including the results and feedback of the debate session, in a database.

[1649] [Input] Feedback and results sent to the device.

[1650] [Operation] The device saves the feedback and results as practice history in a database.

[1651] [Output] Saved practice history data.

[1652] (Application example 1)

[1653] 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."

[1654] In recent years, there has been a demand for staff to improve their skills in customer service to improve customer satisfaction. However, training that simulates real-life customer service situations is costly and time-consuming. To solve this problem, an efficient and effective staff training system is needed.

[1655] 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.

[1656] In this invention, the server includes means for generating chatbots for debate using generative artificial intelligence, means for having the generated chatbots debate, means for determining the outcome based on the content of the debate, means for providing feedback to the user based on the training history, means for generating chatbots with different roles based on selected topics, means for running debates that simulate customer service scenes, and means for providing specific feedback for improvement based on the analysis results, thereby enabling training to efficiently improve customer service skills for staff at brick-and-mortar stores.

[1657] "Generative AI" is an AI technology that uses algorithms such as large-scale language models to generate new data and content.

[1658] A "chatbot" is an autonomous program that can conduct conversations based on specific tasks.

[1659] A "debate" is an activity in which opinions are exchanged and discussed on a specific topic.

[1660] The "method of determining the winner" is a method of analyzing the content of the debate and determining the winner based on logic and persuasiveness.

[1661] "Training history" is data that records the content and results of debate sessions that a user has conducted.

[1662] "Feedback" refers to specific advice and improvements provided to users based on an analysis of the debate results.

[1663] A "topic" is a theme or subject that will be addressed in a debate or simulation.

[1664] A "role" is the position or stance that each chatbot takes in the debate.

[1665] "Customer service scene" refers to the situation in which customers are treated and served in a physical store.

[1666] The "analysis results" are the results of analyzing the content of the debate based on the evaluation criteria.

[1667] "Improvement feedback" refers to specific improvement measures and suggestions provided to users based on the analysis results.

[1668] This invention is a training system for improving the customer service skills of store staff, and is realized using generative artificial intelligence. Specific embodiments of this system are described below.

[1669] Program Structure

[1670] The system consists of three main components: a server, a terminal, and a user.

[1671] Hardware and Software

[1672] Hardware: Smartphones, tablets (e.g., iOS, Android devices)

[1673] software:

[1674] Large-scale language models (e.g., GPT-4)

[1675] Database management system (MySQL, etc.)

[1676] Network communication protocol (REST API)

[1677] Processing flow

[1678] 1. User operations

[1679] The user launches the application on a smartphone or tablet and begins training.

[1680] The user selects a specific customer service topic (e.g., handling complaints, explaining new products, responding to customer requests, etc.).

[1681] 2. Server Operation

[1682] Based on the topic and settings received from the user, the server uses generative artificial intelligence (such as GPT-4) to generate Chatbot A and Chatbot B to participate in the debate.

[1683] Chatbot A and Chatbot B are assigned different roles (e.g., customer role, staff role).

[1684] 3. Conducting the debate

[1685] The server issues a command to start a debate session, and Chatbot A (the customer) makes the first statement, for example, "This product was defective, and I would like a refund."

[1686] Next, Chatbot B (playing the role of staff) responds, for example, by saying, "I'm sorry. We'll take the time to hear more details and provide an appropriate response."

[1687] This exchange is repeated for multiple turns.

[1688] 4. Analysis of results and feedback

[1689] Once the debate is over, the server analyzes all statements based on logic, persuasiveness, and effectiveness of counterarguments.

[1690] A score is assigned to each statement and a final overall score is calculated.

[1691] The server generates feedback based on the overall score and provides it to the user, including specific advice such as, "Your initial apology was timely, but you didn't ask detailed questions enough."

[1692] 5. History Storage

[1693] The server stores the user's training history in a database, which records the results and feedback of each debate session and allows the user to refer to it later.

[1694] Specific examples

[1695] For example, if a user selects the topic "Handling Complaints" and presses the "Start Debate" button, the system will send the following prompt to the server:

[1696] Theme: Handling complaints

[1697] Setting: A customer is dissatisfied with a product, and staff members handle the complaint appropriately.

[1698] As a result, Chatbot A will say, "This product was defective," and Chatbot B will respond, "Sorry. What was the problem?" This debate will continue for several turns, and finally feedback will be provided.

[1699] This system will enable store staff to efficiently and effectively improve their customer service skills.

[1700] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1701] Step 1:

[1702] A user launches the application using a smartphone or tablet and selects a specific customer service topic.

[1703] Input: User selects a topic (e.g., complaint handling) in the application.

[1704] Output: Selected topics and configuration information are sent to the server

[1705] Specific operation: The user selects "Complaint handling" on the app screen and presses the "Start debate" button.

[1706] Step 2:

[1707] The server uses generative artificial intelligence to generate chatbot A and chatbot B to participate in the debate based on the topic and settings received from the user.

[1708] Input: Topic and preference information received from the user

[1709] Output: Generated Chatbot A and Chatbot B

[1710] Specific operation: The server calls a large-scale language model (such as GPT-4) to generate chatbot A for the customer role and chatbot B for the staff role.

[1711] Step 3:

[1712] The server initiates the debate session, and Chatbot A makes the first statement.

[1713] Input: Generated Chatbot A and Chatbot B

[1714] Output: Chatbot A's first statement

[1715] Specific operation: The server makes Chatbot A say something like, "This product was defective, I would like a refund."

[1716] Step 4:

[1717] Chatbot B then makes a rebuttal.

[1718] Input: Chatbot A's statement

[1719] Output: Chatbot B's rebuttal

[1720] Specific operation: The server makes Chatbot B say something like, "I'm sorry. We will ask for more details and take appropriate action."

[1721] Step 5:

[1722] This debate is repeated for multiple turns.

[1723] Input: What Chatbot A and B say in each turn

[1724] Output: A dialogue log of the entire debate

[1725] Specific operation: The server manages the dialogue between chatbots and records each utterance.

[1726] Step 6:

[1727] Once the debate is over, the server analyzes all statements.

[1728] Input: Dialogue log of the entire debate

[1729] Output: Analysis results and scores for each utterance

[1730] Specific operation: The server analyzes the content of the statement and scores it based on logic, persuasiveness, and effectiveness of the counterargument.

[1731] Step 7:

[1732] The server generates an overall score and feedback and provides it to the user.

[1733] Input: Analysis results and scores for each statement

[1734] Output: Overall score and feedback message

[1735] Specific behavior: The server generates specific feedback such as "The timing of the initial apology was appropriate, but the question was not detailed enough" and sends it to the user's device.

[1736] Step 8:

[1737] The server stores the user's training history in a database.

[1738] Input: Results and feedback from the debate session

[1739] Output: Saved training history

[1740] Specific operation: The server saves the results and feedback of this debate session in a database for future reference.

[1741] 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.

[1742] MODE FOR CARRYING OUT THE INVENTION

[1743] This invention is a system that combines generative AI and an emotion engine to build a debate system that provides debate practice that takes into account the user's emotions. This system is composed of a chatbot generated by generative AI and an emotion engine that recognizes the user's emotions.

[1744] Program processing explanation

[1745] Chatbot generation

[1746] Server operations

[1747] 1. Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence to generate chatbot A and chatbot B. These chatbots are configured to have different opinions and personalities.

[1748] Emotion Recognition in Action

[1749] Operations performed by the device

[1750] 1. While users are debating, the device captures the data necessary for emotion recognition (e.g., facial expressions, voice tone, etc.) in real time.

[1751] 2. The device sends the captured data to the emotion engine, which analyzes the user's emotions.

[1752] Conducting the debate

[1753] Server operations

[1754] 1. The server sends a command to start a dialogue session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A may say, "Increasing the use of renewable energy is the most effective way to prevent global warming."

[1755] Server operations

[1756] 1. The server passes the turn to Chatbot B. Chatbot B counters, saying, "Renewable energy is expensive and inefficient with current technology."

[1757] Emotion-Based Response Modulation

[1758] Server operations

[1759] 1. The server receives emotional data from the emotion engine and adjusts the chatbot's responses accordingly based on the user's emotional state. For example, if the user is feeling stressed, the chatbot's responses will be made gentler.

[1760] Judging the winner

[1761] Server operations

[1762] 1. After the debate is over, the server analyzes the content of the debate statements. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of counterarguments, etc.). The total score is calculated to determine which chatbot is the winner.

[1763] Providing feedback

[1764] Operations performed by the server and terminal

[1765] 1. The server generates ratings and sends feedback to the device.

[1766] 2. The device displays the debate results along with feedback, including what went well and what needs improvement.

[1767] Save practice history

[1768] Operations performed by the device

[1769] 1. The device stores the user's practice history in a database, which records the results and feedback of the debate session and can be referenced later.

[1770] Specific examples

[1771] 1. Starting example

[1772] To allow users to debate on the topic of "environmental issues" on their devices, chatbot A is set to advocate for "expanding the use of renewable energy," and chatbot B is set to advocate for "improving the efficiency of the current energy supply system."

[1773] After setting up, press the "Start Debate" button.

[1774] 2. Example of debate progression

[1775] The server generates Chatbot A and Chatbot B.

[1776] Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1777] Chatbot B counters, "Renewable energy is expensive and inefficient with current technology."

[1778] This process is repeated multiple times.

[1779] Meanwhile, the emotion engine recognizes the user's emotions, and the server adjusts the chatbot's responses based on that data.

[1780] 3. Example of results

[1781] The server performs the final analysis and determines the winner (e.g., Chatbot B) based on the overall score.

[1782] The device displays "Debate Results: Winner - Chatbot B" and provides feedback such as "Your emphasis on the cost issue of renewable energy was appreciated."

[1783] The analysis results from the emotion engine are also displayed, providing feedback on the user's emotional changes and the chatbot's corresponding adjustment history.

[1784] By incorporating these processes and functions, the debate system of the present invention provides a training environment for users to effectively improve their debate skills, and also enables flexible responses that take emotions into consideration.

[1785] The processing flow will be explained below.

[1786] MODE FOR CARRYING OUT THE INVENTION

[1787] Processing Steps

[1788] Step 1:

[1789] User-defined themes

[1790] The user starts up the device and opens the debate system application. On the screen where the user selects the debate topic, they select "environmental issues." On the detailed settings screen, they set Chatbot A to advocate "expanding the use of renewable energy" and Chatbot B to advocate "improving the efficiency of the current energy supply system." Once the settings are complete, the user presses the "Start Debate" button.

[1791] Step 2:

[1792] The server generates the chatbot

[1793] The server receives the debate topic and settings sent by the user. The server inputs the topic and settings into the generative AI to generate chatbot A. Chatbot A has the knowledge base and personality to argue for "expanding the use of renewable energy." Next, the same process is used to generate chatbot B. Chatbot B has the knowledge base and personality to argue for "improving the efficiency of the current energy supply system."

[1794] Step 3:

[1795] The device captures emotional data

[1796] While the user is preparing for the debate, the device captures the user's emotional data (e.g., facial expressions, voice tone, heart rate, etc.) in real time. The emotional data is sent from the device to the emotion engine.

[1797] Step 4:

[1798] Emotion engine analyzes emotions

[1799] The emotion engine analyzes the user's emotions based on the captured data, and the analysis results include information such as whether the user is stressed, relaxed, or focused.

[1800] Step 5:

[1801] Start a debate

[1802] The server sends a command to start a conversation session between Chatbot A and Chatbot B. Chatbot A makes the first statement, and the content of that statement is recorded. For example, Chatbot A says, "Renewable energy is an important means of preventing global warming."

[1803] Step 6:

[1804] The chatbot dialogue progresses

[1805] The server passes the turn to Chatbot B, who then counters by saying that renewable energy is expensive and inefficient with current technology. This process is repeated multiple times, and the debate continues until the specified number of turns or time has been reached.

[1806] Step 7:

[1807] Emotion-Based Response Modulation

[1808] The server receives the analysis results from the emotion engine and adjusts the chatbot's response accordingly depending on the user's emotional state. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a calm tone and content.

[1809] Step 8:

[1810] Debate analysis and winner determination

[1811] After the debate is over, the server analyzes the content of the debate statements and assigns a score to each statement based on specific evaluation criteria (e.g., logic, persuasiveness, effectiveness of counterarguments, etc.). An overall score is calculated to determine which chatbot is the winner.

[1812] Step 9:

[1813] Generate results and feedback

[1814] The server generates the debate results and detailed feedback, including points of merit and areas for improvement, as well as feedback on the user's emotional changes.

[1815] Step 10:

[1816] The terminal displays the results

[1817] The device receives the results and feedback sent from the server. The device displays "Debate Results: Winner - Chatbot B" to the user and provides feedback such as "Your emphasis on the cost issue of renewable energy was highly praised." The device also displays the analysis results from the emotion engine, showing feedback on changes in the user's emotions and the chatbot's corresponding adjustments.

[1818] Step 11:

[1819] The device stores your practice history

[1820] The device stores the user's practice history in a database, which includes the results and feedback of each debate session and can be referenced later.

[1821] This system allows users to effectively improve their debating skills and also allows training that takes into account their own emotional state.

[1822] Example 2

[1823] 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."

[1824] While conventional debate systems can provide a training environment for improving technical debate skills, it is difficult to consider the user's emotional state. Furthermore, since users' emotions often influence the discussion in real debates, training that ignores emotions is inefficient. Furthermore, there is a problem that feedback on debate results is insufficient, making it difficult to contribute to the improvement of users' skills.

[1825] 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.

[1826] In this invention, the server includes means for generating artificial dialogue agents for debate using generative artificial intelligence, means for having the generated artificial dialogue agents debate, means for analyzing the emotional state of the user during the debate using an emotion engine that recognizes the user's emotions, means for adjusting the responses of the artificial dialogue agents based on the emotional state of the user, means for determining the outcome of the debate based on the content of the debate, and means for providing feedback to the user based on the practice history, thereby making it possible to effectively improve the debating skills while taking into consideration the emotional state of the user.

[1827] "Generative artificial intelligence" is an algorithm that learns from large amounts of data and generates new text.

[1828] An "artificial dialogue agent" is software that can carry out pre-programmed dialogues or dialogues that are dynamically generated using generative artificial intelligence.

[1829] A "debate" is an act of logical discussion between people with different positions or opinions on a particular topic.

[1830] The "emotion engine" is a technology that analyzes data such as the user's voice and facial expressions to grasp their emotional state in real time.

[1831] The "means of determining victory or defeat" is a system that analyzes the content of the debate, assigns points based on evaluation criteria such as logic, persuasiveness, and the effectiveness of the rebuttal, and determines which side is superior.

[1832] "Feedback" refers to evaluations and comments provided after a debate, providing users with information to understand their strengths and areas for improvement.

[1833] "Practice history" is a record of past debate sessions conducted by the user, including debate content, evaluations, feedback, and the like.

[1834] A "topic" is a particular topic or issue that is the subject of debate.

[1835] "Settings" are parameters that the user specifies regarding the progress and conditions of the debate.

[1836] This invention provides a debate system that combines generative AI and an emotion engine. This system allows generated chatbots to debate with each other and provides training while taking into account the user's emotions. This system is composed of the following main components:

[1837] 1. Chatbot Creation

[1838] Based on the debate topic and detailed settings received from the user, the server uses generative artificial intelligence (such as GPT-3) to generate Chatbot A and Chatbot B. These chatbots are configured to have different opinions and personalities. Specific examples of prompts include "Please state your opinion on the topic: Expand the use of renewable energy" and "Improve the efficiency of the current energy supply system."

[1839] 2. Emotion Recognition

[1840] The device captures the user's facial expressions with a webcam and collects their voice tone with a microphone while they debate. This data is sent to an emotion engine (e.g., emotion recognition API) and analyzed in real time. The analysis results are sent to a server and reflected in the progress of the debate.

[1841] 3. Conducting the debate

[1842] The server starts a dialogue session between Chatbot A and Chatbot B. Chatbot A makes an initial statement, followed by a rebuttal from Chatbot B. This exchange is repeated multiple times.

[1843] 4. Emotion-Based Response Modulation

[1844] The server receives the emotion data from the emotion engine and adjusts the chatbot's responses based on the user's emotional state, for example, changing the chatbot's tone to be calmer if the user is stressed.

[1845] 5. Judging the Winner

[1846] After the debate is over, the server analyzes the content of the debate and assigns scores based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is superior.

[1847] 6. Providing Feedback

[1848] The server sends the generated evaluation and feedback to the device, which displays it to the user, providing feedback on what was good and what needs improvement. The analysis results from the emotion engine are also displayed.

[1849] 7. Save your practice history

[1850] The device stores the user's practice history in a database, which includes the results and feedback of the debate session. Users can later refer to this history to help them self-evaluate and improve their skills.

[1851] Through the above process, the present invention provides a training environment for users to effectively improve their debating skills. In particular, emotion recognition and response adjustment by the emotion engine enables flexible responses that take into account the user's emotional state. As a result, it is possible to provide higher satisfaction and effectiveness than conventional debate systems.

[1852] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1853] Step 1: Receiving the debate topic and detailed settings

[1854] server

[1855] Input: Receive debate topic and detailed settings from user.

[1856] Data processing: Analyzing received information and classifying themes and settings.

[1857] Output: Parsed themes and settings are sent to the generative AI.

[1858] Specific operation: The user enters the debate topic "environmental issues" and detailed settings into the terminal, and the server receives and analyzes this.

[1859] Step 2: Initializing the generative AI and creating a prompt

[1860] server

[1861] Input: Parsed themes and settings.

[1862] Data processing: Initialize a generative artificial intelligence (e.g., GPT-3) and create a prompt based on the theme.

[1863] Output: Send the prompt to the generative artificial intelligence model.

[1864] Specific operation: The server creates prompt sentences for Chatbot A, which advocates "expanding the use of renewable energy," and Chatbot B, which advocates "improving the efficiency of the current energy supply system."

[1865] Step 3: Generate the chatbot

[1866] server

[1867] Input: Prompt statement.

[1868] Data processing: Chatbot A and Chatbot B are generated using generative artificial intelligence.

[1869] Output: Saves the generated chatbot information.

[1870] Specific operation: The server uses GPT-3 to generate two different chatbots based on the prompt sentence.

[1871] Step 4: Capture emotion recognition data

[1872] Terminal

[1873] Input: User's facial expressions, voice tone.

[1874] Data processing: Capture data using a webcam and microphone.

[1875] Output: Sends the captured data to the emotion engine.

[1876] Specific operation: The device captures the user's facial expressions with a webcam and collects voice tones with a microphone.

[1877] Step 5: Analyze the sentiment data

[1878] Terminal

[1879] Input: Captured emotion data.

[1880] Data processing: Analyze using an emotion engine (e.g., emotion recognition API).

[1881] Output: Send the analysis results to the server.

[1882] Specific operation: The emotion engine analyzes the user's stress level, joy, anger, etc. and sends the results to the server.

[1883] Step 6: Start the debate session

[1884] server

[1885] Input: Information for the generated chatbot.

[1886] Data processing: Initialize the conversation session between Chatbot A and Chatbot B.

[1887] Output: Send the first command to Chatbot A.

[1888] Specific operation: The server makes chatbot A say, "We need to expand the use of renewable energy."

[1889] Step 7: Continue the conversation

[1890] server

[1891] Input: What Chatbot A says.

[1892] Data processing: Send a counterargument to Chatbot B.

[1893] Output: Record what Chatbot B said.

[1894] Specific behavior: The server has Chatbot B respond by saying, "Renewable energy is too expensive right now."

[1895] Step 8: Emotion-Based Response Adjustment

[1896] server

[1897] Input: Analysis results from the emotion engine.

[1898] Data manipulation: Tailoring chatbot responses based on the user's emotional state.

[1899] Output: Sends the tailored response to the chatbot.

[1900] Specific behavior: If the user is feeling stressed, the server changes the chatbot's tone to be gentler.

[1901] Step 9: Ending the debate and deciding who won

[1902] server

[1903] Input: The entire debate.

[1904] Data processing: Generate scores based on logic, persuasiveness, and effectiveness of counterarguments.

[1905] Output: Determine winner based on overall score.

[1906] Specific operation: The server determines the "Total score: Chatbot B is the winner" and displays the reason.

[1907] Step 10: Provide feedback

[1908] Servers and Terminals

[1909] Input: Debate evaluation and feedback.

[1910] Data processing: Generate ratings and feedback and send them to your device.

[1911] Output: Display feedback to the user.

[1912] Specific operation: The server generates feedback based on the evaluation, and the device displays something like, "Your emphasis on the cost issue of renewable energy was appreciated."

[1913] Step 11: Save your practice history

[1914] Terminal

[1915] Input: Results, evaluations, and feedback from the debate session.

[1916] Data processing: Practice history is saved in a database.

[1917] Output: Makes the history available to the user.

[1918] What it does: The device stores the debate results and feedback in a database so that the user can refer to them later.

[1919] (Application example 2)

[1920] 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."

[1921] Conventional debate training systems tend to engage in monotonous dialogue without considering the user's emotional state, making it difficult to provide a flexible training environment that can respond to changes in emotions. Furthermore, they are unable to provide appropriate feedback that responds immediately to the user's stress or emotional changes, which increases the burden on the user. Furthermore, because applications to other fields such as food delivery have not been considered, there is a lack of application of emotion recognition in practical scenarios other than debates.

[1922] 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.

[1923] In this invention, the server includes a means for generating chatbots for debate using generative artificial intelligence, a means for conducting debates between the generated chatbots, a means for using an emotion recognition engine to recognize users' emotions in real time, a means for adjusting the chatbots' responses based on the users' emotions, a means for determining the outcome of the debate based on the content of the debate, and a means for providing feedback to the user based on the training history. This enables flexible and effective debate training that takes into account the user's emotional state. This system can also be applied to other fields such as food delivery, and can have practical applications such as an emotion-based ordering system.

[1924] "Generative AI" is AI that uses technology to automatically generate text and conversations based on given input data.

[1925] A "chatbot" is a computer program that can converse with humans in natural language, and is particularly used in interactive formats such as debates.

[1926] An "emotion recognition engine" is a technology that analyzes data such as voice and images to identify a user's emotional state in real time.

[1927] The "means for determining victory or defeat" is a function that analyzes the results of a dialogue or debate and determines which chatbot is superior based on the set evaluation criteria.

[1928] The "means for providing feedback" is a function for notifying the user of the results of the debate and areas for improvement, and for providing information to improve the effectiveness of training.

[1929] "Training history" is data that accumulates records and analysis results of debate sessions conducted by users.

[1930] The "means for adjusting the chatbot's response based on emotions" is a function that adjusts the chatbot to return an appropriate response based on the user's emotional data obtained by the emotion recognition engine.

[1931] This invention relates to a debate system that combines generative artificial intelligence and an emotion recognition engine. It provides a training environment for users to effectively improve their debate skills and enables flexible responses according to the user's emotional state.

[1932] The debate system of the present invention is composed of a server and terminals. Specifically, the following process is performed.

[1933] The server uses generative artificial intelligence to generate chatbots for debates based on the debate topic and settings received from the user. The generated chatbots are configured to have different opinions and personalities. A means is provided for having the chatbots debate and recording the content of the debates.

[1934] The device uses an emotion recognition engine to recognize users' emotions in real time during debates. The emotion recognition engine detects the user's emotional state by capturing and analyzing data such as facial expressions and voice tones.

[1935] The server adjusts the chatbot's responses appropriately based on data from the emotion engine. For example, if the user is feeling stressed, the chatbot's responses will be gentler. It also determines the outcome of the debate based on the content of the debate, and generates and provides evaluations and feedback to the user.

[1936] Next, we will explain an example of a food delivery application. This system uses an emotion recognition engine to suggest optimal menus and ordering methods based on the user's emotional state in a food delivery service.

[1937] The device captures the user's voice and image data and analyzes the user's emotions using an emotion recognition engine such as EmotionRecognition. It then uses MenuGenerator to generate an optimal menu based on the analyzed emotion data. The generated menu is then presented to the user by a chatbot, which responds based on the user's emotions, making it possible to suggest the most suitable dishes for the user. The device can then finalize the order using the OrderSystem.

[1938] As a specific example, the following steps can be considered.

[1939] 1. A user asks, "What's the recommendation of the day?"

[1940] 2. The emotion recognition engine analyzes the user's voice and facial expressions to detect when the user is feeling stressed.

[1941] 3. Menu Generator suggests dishes that have a relaxing effect.

[1942] 4. The chatbot suggests, "How about some relaxing green tea cookies and chamomile tea?"

[1943] An example of a prompt for the generative AI model is as follows:

[1944] The user's name is "Tanaka." Please suggest some dishes that would be suitable for Tanaka when he is feeling stressed.

[1945] For example: "How about some relaxing green tea cookies and chamomile tea?"

[1946] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1947] Step 1:

[1948] The user sends the debate topic and settings to the server via the terminal. The data input to the terminal includes the debate topic, setting details, and user identification information. For example, input data might include "Today's debate topic is environmental issues," "Expanding the use of renewable energy," and "Improving the efficiency of the current energy supply system."

[1949] Step 2:

[1950] The server uses generative artificial intelligence to generate chatbots for debate. In this process, the server inputs the theme and settings received from the user and generates Chatbot A and Chatbot B, each with different opinions and personalities. For example, Chatbot A will be a character that advocates "expanding the use of renewable energy," while Chatbot B will be a character that advocates "improving the efficiency of the current energy supply system."

[1951] Step 3:

[1952] The device captures audio and video data to recognize users' emotions in real time during a debate. This data is sent from the device to an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to generate emotion data. The captured data is the audio and video of the user's facial expressions as they speak during the debate.

[1953] Step 4:

[1954] The server receives emotional data from the emotion recognition engine and adjusts the chatbot's response based on that data. For example, if the user is feeling stressed, the server instructs the chatbot to respond in a gentle tone. The input data is the emotional data sent from the emotion recognition engine, and the output is the adjusted chatbot's response.

[1955] Step 5:

[1956] The server initiates a dialogue session between Chatbot A and Chatbot B and records their statements. The server then allows Chatbot A to make the first statement, followed by Chatbot B's rebuttal. For example, Chatbot A may say, "Renewable energy is an important means of preventing global warming," and Chatbot B may then rebut, "Renewable energy is expensive and inefficient with current technology."

[1957] Step 6:

[1958] The server determines the winner based on the content of the debate. The server assigns a score to each statement based on specific evaluation criteria (logic, persuasiveness, effectiveness of rebuttal, etc.), calculates the overall score, and determines which chatbot is the winner. The input data are the content of the debate, and the output is the overall score and the winner's result.

[1959] Step 7:

[1960] The server generates evaluations and feedback along with the outcome of the debate and sends them to the device. The device then displays the debate results and feedback to the user. Based on the winner of the debate, feedback such as "Chatbot B is the winner. It was praised for emphasizing the cost issue of renewable energy" is presented.

[1961] Step 8:

[1962] The device stores the user's practice history in a database. This history records the results and feedback of debate sessions, and the user can refer to it later. The database stores the user's debate topic, win / loss results, score, feedback, etc.

[1963] This process flow allows users to effectively improve their debating skills. Furthermore, the use of an emotion recognition engine allows for flexible responses according to the user's emotional state, providing a comfortable training environment.

[1964] 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.

[1965] 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.

[1966] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1967] 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.

[1968] 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.

[1969] 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.

[1970] 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).

[1971] 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.

[1972] 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."

[1973] 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.

[1974] 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).

[1975] 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.

[1976] 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.

[1977] 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.

[1978] 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.

[1979] 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.

[1980] 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.

[1981] 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.

[1982] 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.

[1983] 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.

[1984] 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.

[1985] The following is further disclosed regarding the above embodiment.

[1986] (Claim 1)

[1987] A means for generating a chatbot for debate using generative artificial intelligence;

[1988] A means for having the generated chatbots debate with each other;

[1989] A means of determining victory or defeat based on the content of the debate;

[1990] means for providing feedback to the user based on their training history;

[1991] A system including:

[1992] (Claim 2)

[1993] The system of claim 1, further comprising means for analyzing the content of the generated chatbot's discussion and generating a score based on logic, persuasiveness, and effectiveness of rebuttal.

[1994] (Claim 3)

[1995] The system of claim 1, further comprising means for a generative artificial intelligence to generate a chatbot based on a debate theme and settings set by a user.

[1996] "Example 1"

[1997] (Claim 1)

[1998] A means for generating a dialogue agent for debate using generative artificial intelligence;

[1999] a means for conducting an interactive session between the generated interactive agents;

[2000] A means for determining the outcome based on the content of the dialogue between the generated dialogue agents;

[2001] means for providing feedback to the user based on the results and analysis of the interactive session;

[2002] A system including:

[2003] (Claim 2)

[2004] 2. The system according to claim 1, further comprising means for analyzing the dialogue content of the generated dialogue agent and generating a score based on logic, persuasiveness, and effectiveness of rebuttal.

[2005] (Claim 3)

[2006] 2. The system of claim 1, further comprising means for generating a dialogue agent by a generative artificial intelligence based on a dialogue theme and settings set by a user.

[2007] "Application Example 1"

[2008] (Claim 1)

[2009] A means for generating a chatbot for debate using generative artificial intelligence;

[2010] A means for having the generated chatbots debate with each other;

[2011] A means of determining victory or defeat based on the content of the debate;

[2012] means for providing feedback to the user based on their training history;

[2013] A means for generating chatbots with different roles based on selected topics;

[2014] A means of conducting debates that simulate customer service situations;

[2015] a means of providing specific improvement feedback based on the analysis results;

[2016] A system including:

[2017] (Claim 2)

[2018] The system of claim 1, further comprising means for analyzing the content of the generated chatbot's discussion and generating a score based on logic, persuasiveness, and effectiveness of rebuttal.

[2019] (Claim 3)

[2020] The system of claim 1, further comprising means for a generative artificial intelligence to generate a chatbot based on a debate theme and settings set by a user.

[2021] "Example 2: Combining Emotion Engines"

[2022] (Claim 1)

[2023] A means for generating an artificial dialogue agent for debate using generative artificial intelligence;

[2024] a means for causing the generated artificial dialogue agents to debate;

[2025] means for analyzing the emotional state of a user during a debate using an emotion engine that recognizes the user's emotions;

[2026] means for adjusting the response of the artificial dialogue agent based on the emotional state of the user;

[2027] A means of determining victory or defeat based on the content of the debate;

[2028] means for providing feedback to the user based on their practice history;

[2029] A system including:

[2030] (Claim 2)

[2031] 2. The system according to claim 1, further comprising means for analyzing the content of the generated artificial dialogue agent's argument and generating a score based on logic, persuasiveness, and effectiveness of rebuttal.

[2032] (Claim 3)

[2033] 2. The system of claim 1, further comprising means for generating an artificial dialogue agent by a generative artificial intelligence based on a debate theme and settings set by a user.

[2034] "Application example 2 when combining emotion engines"

[2035] (Claim 1)

[2036] A means for generating a chatbot for debate using generative artificial intelligence;

[2037] A means for having the generated chatbots debate with each other;

[2038] means for using an emotion recognition engine to recognize the user's emotions in real time;

[2039] means for adjusting the chatbot's responses based on the user's emotions;

[2040] A means of determining victory or defeat based on the content of the debate;

[2041] means for providing feedback to the user based on their training history;

[2042] A system including:

[2043] (Claim 2)

[2044] The system of claim 1, further comprising means for analyzing the content of the generated chatbot's discussion and generating a score based on logic, persuasiveness, and effectiveness of rebuttal.

[2045] (Claim 3)

[2046] The system of claim 1, further comprising means for a generative artificial intelligence to generate a chatbot based on a debate theme and settings set by a user. [Explanation of symbols]

[2047] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for generating a chatbot for debate using generative artificial intelligence; A means for having the generated chatbots debate with each other; A means of determining victory or defeat based on the content of the debate; means for providing feedback to the user based on their training history; A system including:

2. The system according to claim 1, further comprising means for analyzing the content of the discussion of the generated chatbot and generating a score based on logic, persuasiveness, and effectiveness of the rebuttal.

3. The system of claim 1, further comprising means for generating a chatbot by a generative artificial intelligence based on a debate theme and settings set by a user.

Images