system

Abstract

Provide a system. 【Solution means】means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording the user's speech in real time, means for analyzing the user's speech and generating an appropriate response, means for analyzing and summarizing the recorded data of the rebate, means for displaying the summarized minutes, means for generating presentation materials based on the minutes, means for brushing up the presentation materials with the user, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot 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 an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance that responds to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In conventional rebate classes, discussions are only held among students in the classroom, so there is a problem that opinions and perspectives are limited. Also, it was difficult for teachers to appropriately provide diverse viewpoints and external opinions. In addition, there was also a problem that it took time and effort for students to create minutes or prepare presentation materials after the rebate. To solve such problems, a system that has an external perspective and can provide diverse opinions is needed.

Means for Solving the Problems

[0005] The present invention is a system that includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, and means for refining the presentation materials with the user. This system introduces an external perspective into rebates, broadens students' horizons by providing diverse opinions, and enables them to efficiently handle complex tasks. Furthermore, this system includes a function for extracting and analyzing keywords when analyzing user statements, and helps to convey information more clearly and visually by providing the generated presentation materials in a format that includes charts and graphs.

[0006] A "rebate topic" refers to the subject matter that is discussed during a rebate.

[0007] "Means of collecting information" refers to a system that automatically collects relevant data and literature from the internet and databases.

[0008] "Methods of summarization" refer to the function of extracting the key points of collected information and putting them into a concise summary.

[0009] "Means of display" refers to displays or screens used to visually provide users with collected information or summaries.

[0010] "Means for recognizing and recording user speech in real time" refers to a function that acquires and saves user voice input as text data.

[0011] "Means for analyzing user statements and generating appropriate responses" refers to a function that analyzes the content of a statement and automatically generates an appropriate reply corresponding to that content.

[0012] "Means for analyzing recorded data" refers to a function that extracts and organizes important information from recorded audio data.

[0013] "Meeting minutes" refers to a document that records the details of a bribe and summarizes its key points and conclusions.

[0014] "Means for generating presentation materials" refers to the function of creating visual slides and presentation materials based on meeting minutes.

[0015] "Refining" refers to the process of further improving the generated presentation materials and making adjustments to communicate information more effectively.

[0016] "Keyword extraction methods" refer to functions that identify and extract important words and phrases from user statements.

[0017] "Formats including charts and graphs" refers to presentation formats that include charts and graphs to visually represent data and information. [Brief explanation of the drawing]

[0018] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7]It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Modes for Carrying Out the Invention

[0019] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

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

[0021] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0022] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0023] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0024] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0026] [First Embodiment]

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

[0028] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0029] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0031] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0035] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0038] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0039] This invention relates to a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. This embodiment details how the system achieves these functions.

[0040] The server first receives a rebate topic entered by the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and its internal database. The collected information is summarized and stored in the server's database. Subsequently, the terminal (tablet or robot) displays this summarized information and provides it to the user (student).

[0041] When the rebate program begins, the server records each user's statements as text data in real time using speech recognition technology. The recorded statements are analyzed by a keyword extraction algorithm, and the server generates an appropriate response. For example, if a user (student) says, "Renewable energy is expensive," the server analyzes this statement and, based on the keyword "cost," generates a response such as, "While the initial investment in solar power generation is high, it has cost-saving effects in the long term."

[0042] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate. This summary is compiled into meeting minutes and displayed on the terminal. The meeting minutes can be reviewed and edited by the user (teacher).

[0043] Next, the server generates a presentation template based on the meeting minutes data. This template is provided in a format that includes charts and graphs and is displayed on the device (tablet or robot). The user (student) reviews the content of the material together with the rebate robot using this template and makes improvements.

[0044] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive." In response to this statement, the server generates an appropriate response, and the terminal (rebate robot) replies, "While the initial investment for solar power generation is high, it has cost-saving effects in the long term." Subsequently, the server analyzes the entire rebate, generates a summarized meeting minutes, and creates a presentation template. Based on this material, the user (student) reviews the final slides with the robot and makes revisions.

[0045] As described above, the present invention is a system that provides consistent support for rebates, from preparation and implementation to summarization and presentation preparation, thereby improving the quality of rebates and enabling efficient class management.

[0046] The following describes the processing flow.

[0047] Step 1:

[0048] The server receives a rebate topic from the user (teacher). The user (teacher) enters the rebate topic "The pros and cons of promoting renewable energy" into the server.

[0049] Step 2:

[0050] The server collects relevant information based on the received rebate topic. This collection involves accessing APIs to retrieve data from the internet and accessing internal databases. However, the collected raw information is not usable as is, so natural language processing (NLP) algorithms are used to summarize the key points and convert it into an easily understandable format.

[0051] Step 3:

[0052] The device (tablet or rebate robot) prepares to provide summarized information to the user (student). At this point, the information is displayed on the tablet or robot's screen.

[0053] Step 4:

[0054] The user (student) begins the rebate. The server uses speech recognition technology to convert each student's statements into text data in real time and records them. During this process, a time stamp of the spoken content is also saved.

[0055] Step 5:

[0056] The server analyzes the user's statements. Using a keyword extraction algorithm, it identifies important keywords from the statements. For example, from the statement "Renewable energy is expensive," it extracts "cost" as an important keyword.

[0057] Step 6:

[0058] The terminal (rebate robot) generates appropriate responses based on analysis information sent from the server. It provides opinions from various perspectives and deepens the discussion. For example, it might generate a response such as, "The initial investment for solar power generation is high, but it has cost-saving effects in the long term."

[0059] Step 7:

[0060] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. Using natural language processing algorithms, it extracts the main points of the discussion and summarizes the key takeaways.

[0061] Step 8:

[0062] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make corrections as needed.

[0063] Step 9:

[0064] The server generates a presentation template based on the meeting minutes. This template includes key points, data, charts, and graphs discussed during the discussion.

[0065] Step 10:

[0066] The terminal displays the generated document template to the user (student). The user (student) reviews the document content together with the rebate robot and makes improvements. For example, they can add data or charts and change the format to make it easier to understand.

[0067] Step 11:

[0068] The user (student) reviews the completed materials and saves them as the final presentation materials. The device provides the completed materials for the presentation day.

[0069] The above is a detailed explanation of the processing steps of the rebate robot system. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates.

[0070] (Example 1)

[0071] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0072] Conventional rebate support systems do not consistently handle the entire process, from collecting and summarizing information related to rebate topics, to recording and analyzing statements, generating appropriate responses, creating meeting minutes, and preparing presentation materials. As a result, the process from rebate preparation to implementation and summarization is inefficient, leading to a decline in the quality of rebates.

[0073] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0074] In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the generated presentation materials with the user, means for collecting relevant information from the internet or an internal database based on the rebate topic entered by the user, means for summarizing the collected information and storing it in a database, means for using natural language processing technology in the collection and summarization process, means for accessing external data sources via an internet connection, means for recording each user's statements as text data in real time using speech recognition technology, means for analyzing the recorded statements with a keyword extraction algorithm, means for using a generative AI model to generate responses based on keywords, and means for inputting prompt sentences into the generative AI model to generate appropriate responses. This makes it possible to consistently and efficiently carry out everything from rebate preparation and implementation to summarization and presentation preparation.

[0075] A "rebate topic" is the subject or theme discussed during a rebate (debate).

[0076] "Users" refer to the people who use the system, specifically teachers and students.

[0077] "Means of information gathering" refers to methods and technologies for obtaining information related to rebate topics from the internet or internal databases.

[0078] "Means of summarization" refer to methods and techniques for concisely organizing collected information.

[0079] "Means of display" refers to methods and techniques for presenting collected and summarized information to the user.

[0080] "Means of recognition and recording" refers to methods and technologies for real-time speech recognition of a user's speech and saving it as text data.

[0081] "Means of analysis" refer to methods and techniques for analyzing recorded text data and extracting meaning and keywords.

[0082] "Means for generating appropriate responses" refer to methods and technologies for responding to users based on analysis results.

[0083] "Means of analyzing audio recordings" refers to methods and techniques for analyzing audio recordings of rebates and extracting important statements and key points.

[0084] "Means of displaying meeting minutes" refers to methods and technologies for displaying summarized meeting minutes to users.

[0085] "Means for generating presentation materials" refers to methods and techniques for creating presentation materials based on meeting minutes data.

[0086] "Methods for refinement" refer to methods and techniques for reviewing generated presentation materials together with the user and making corrections and improvements as needed.

[0087] An "external data source" refers to an information source located outside the system, such as publicly available information on the internet or external databases.

[0088] "Natural language processing technology" refers to artificial intelligence technology used to analyze and summarize collected information and user statements.

[0089] "Speech recognition technology" refers to the technology used to convert speech data into text data.

[0090] A "keyword extraction algorithm" is a computational method or technique for selecting important words and phrases from text data.

[0091] A "generative AI model" is an artificial intelligence model designed to generate appropriate responses based on user statements.

[0092] A "prompt statement" is a sentence input to a generative AI model that provides instructions or context for generating a specific response.

[0093] The present invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. Specific embodiments of the present invention are described below.

[0094] 1. Basic System Configuration

[0095] This system consists of a server, terminals (tablets and robots), and users (teachers and students). The server plays a central role in information collection and analysis, while the terminals provide information to users and accept user input.

[0096] Hardware and software to be used

[0097] Server: A server computer with a high-performance processor and large memory capacity is recommended. The server accesses external data sources via an internet connection and manages the internal database.

[0098] Internet connection: A high-speed internet connection is required to facilitate data collection and access to external data sources.

[0099] Speech recognition technology: Google's Speech-to-Text API is used to convert user speech into text data in real time.

[0100] Natural Language Processing (NLP) technologies: Google Cloud's Natural Language API and Python libraries (e.g., BeautifulSoup, spaCy) are used. This is used to summarize relevant information and analyze user utterances.

[0101] Keyword extraction algorithm: TF-IDF and the spaCy library are used to extract important keywords from user utterances.

[0102] Generative AI Model: Generative AI models such as OpenAI's GPT-3 (registered trademark) are used. This is to generate appropriate responses based on user statements.

[0103] Database: Relational databases such as MySQL (registered trademark) and PostgreSQL are used to manage collected information, summary data, and meeting minutes.

[0104] 2. System Operation

[0105] The server first receives a rebate topic as input from the user (teacher). Based on this input, the server collects relevant information from the internet and its internal database. Google's search API and queries from the internal database are used for information collection.

[0106] The collected information is summarized using the Google Cloud Natural Language API. The summarized information is stored in the server's database.

[0107] Next, the device (tablet or robot) displays this summary information to the user (student). This allows the student to prepare for their rebate in advance.

[0108] Rebate start:

[0109] During the rebate process, the server records each user's speech in real time as text data using the Google Speech-to-Text API. This recorded data is then analyzed using TF-IDF and the spaCy library to extract important keywords.

[0110] Based on the analyzed data, the server inputs prompts into a generative AI model (e.g., OpenAI's GPT-3) to generate an appropriate response. For example, if a user says, "Renewable energy is expensive," the server inputs the prompt, "Generate a response regarding the cost of renewable energy," into the generative AI model, which then generates an appropriate response.

[0111] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. This summary is compiled into meeting minutes and displayed on the terminal. The user (teacher) reviews the minutes and makes corrections as needed.

[0112] Finally, the server generates a presentation template based on the meeting minutes. This template includes charts and graphs, and uses libraries such as Matplotlib and Pandas. This material is displayed on the terminal, and the user (student) reviews the content together with the rebate robot and makes final adjustments.

[0113] 3. Specific Examples

[0114] If the rebate topic is "The pros and cons of promoting renewable energy," the user (teacher) enters this topic into the server. The server uses the internet and its internal database to collect relevant information, summarizes data such as "data on the costs and efficiency of renewable energy implementation," and stores it in the database. The terminal displays this information to the students to help them prepare for their rebates.

[0115] During the rebate process, the user's (student's) statement, "Renewable energy is expensive," is recorded in real time as text data by the server, and the keyword "cost" is analyzed by a keyword extraction algorithm. The prompt "Generate a response about the cost of renewable energy" is input to the generative AI model (GPT-3), and the response "The initial investment for solar power generation is high, but it has cost-saving effects in the long term" is generated.

[0116] After the rebate process is complete, the server re-analyzes the audio data and creates meeting minutes. It then generates a presentation template, which the user reviews and revises together with the rebate robot.

[0117] Thus, this system makes it possible to efficiently handle the entire process of rebate preparation, from preparation and execution to summarization and presentation preparation.

[0118] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0119] Step 1:

[0120] The server receives a rebate topic as input from the user (teacher).

[0121] Input: Users enter rebate topics such as "The pros and cons of promoting renewable energy."

[0122] Data processing: The server saves this topic and begins searching for related information.

[0123] Output: Ready to explore

[0124] Step 2:

[0125] The server collects relevant information from the internet and internal databases based on the rebate topic.

[0126] Input: Rebate Topic

[0127] Data Processing: We collect the latest information from the internet using Google's search API and other tools, and also retrieve relevant information from our internal database.

[0128] Output: List of collected related information

[0129] Step 3:

[0130] The server summarizes the collected information and stores it in a database.

[0131] Input: List of collected information

[0132] Data processing: Text summarization is performed using the Google Cloud Natural Language API.

[0133] Output: Summarized information is saved to the database.

[0134] Step 4:

[0135] The device (tablet or robot) provides summary information to the user (student).

[0136] Input: Summary information stored in the database

[0137] Data processing: None

[0138] Output: Summary information will be displayed on the terminal screen.

[0139] Step 5:

[0140] The server recognizes and records each user's statements in real time after the rebate program begins.

[0141] Input: User's voice utterance

[0142] Data processing: Convert speech to text using the Google Speech-to-Text API.

[0143] Output: Real-time recorded text data

[0144] Step 6:

[0145] The server analyzes the recorded statements and generates an appropriate response.

[0146] Input: Text data recorded in real time

[0147] Data processing: Extract keywords using TF-IDF or the spaCy library.

[0148] Output: Keyword extraction results

[0149] Step 7:

[0150] The server uses a generative AI model to generate an appropriate response.

[0151] Input: Keyword extraction results

[0152] Data processing: Input prompt text into a generative AI model (e.g., OpenAI GPT-3) and generate a response.

[0153] Output: Response generated by the generative AI model

[0154] Specific example: If a student says "Renewable energy is expensive" during a rebate session, and the prompt is "Generate a response about the cost of renewable energy," the response will be "The initial investment for solar power is high, but it has cost-saving effects in the long term."

[0155] Step 8:

[0156] After the rebate is completed, the server analyzes the recorded data, summarizes it, and creates meeting minutes.

[0157] Input: Rebate recording data

[0158] Data processing: Summarize the audio data using a text analysis algorithm.

[0159] Output: Summarized meeting minutes

[0160] Step 9:

[0161] The terminal displays summarized meeting minutes to the user (teacher) and assists with review and correction.

[0162] Input: Summarized meeting minutes

[0163] Data processing: None

[0164] Output: The meeting minutes will be displayed on the terminal screen.

[0165] Step 10:

[0166] The server generates presentation material templates based on the meeting minutes.

[0167] Input: Summarized meeting minutes

[0168] Data processing: Create charts and graphs using Matplotlib and Pandas.

[0169] Output: Presentation material template

[0170] Step 11:

[0171] The device provides the generated presentation materials to the user (student) and supports them in refining their work.

[0172] Input: Presentation material template

[0173] Data processing: Provides an interface to reflect necessary modifications.

[0174] Output: Finalized presentation materials

[0175] Through the above processing steps, this system can efficiently handle everything from preparing and executing rebates to summarizing and preparing presentations.

[0176] (Application Example 1)

[0177] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0178] In modern factory settings, there is a need to quickly identify problems on the production line and propose improvements. However, there is a lack of means to identify problems in real time and provide appropriate information, which can lead to decreased production efficiency. In particular, there is a need for a system that provides quick and concrete solutions to production problems and troubles faced by operators.

[0179] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0180] In this invention, the server includes means for inputting rebate topics, means for collecting and summarizing information related to rebate topics, means for displaying information related to rebate topics, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on meeting minutes, means for refining presentation materials with the user, means for collecting problems and improvement suggestions at the production site in real time using speech recognition technology, means for extracting keywords from speech-recognized text and collecting related information, and means for summarizing and displaying the collected information. This enables the rapid and specific provision of information at the production site.

[0181] A "rebuttal topic" is a theme or issue that is discussed in a debate.

[0182] "Means of collecting and summarizing information" refers to methods and devices for obtaining relevant information from sources such as the internet or internal databases and compiling it into a concise format.

[0183] "Means of displaying information" refers to devices and software that visually provide collected and summarized information to users.

[0184] "Means for recognizing and recording user speech in real time" refers to a device or method that uses speech recognition technology to instantly save user speech as text data.

[0185] "Means for analyzing user utterances and generating appropriate responses" refers to methods or devices that extract keywords from speech-recognized text, analyze their content, and automatically create appropriate responses.

[0186] "Means for analyzing and summarizing debate recordings" refers to devices or methods for analyzing debate session recordings and concisely summarizing the main points of the discussion.

[0187] "Means of displaying meeting minutes" refers to devices or software that visually provide users with a summarized version of the discussion.

[0188] "Means for generating presentation materials" refers to devices or methods that automatically create presentation-appropriate forms and materials based on meeting minutes data.

[0189] "Means of refinement" refer to devices or methods that allow users to modify or improve the generated presentation materials.

[0190] "Methods for collecting problems and improvement suggestions in the production field in real time using voice recognition technology" refers to devices and technologies that recognize voice communication on the production line and collect that information immediately.

[0191] "Means for extracting keywords and collecting related information" refers to methods or devices that extract important words from speech-recognized text and search for and obtain related information based on those words.

[0192] "Means for summarizing and displaying collected information" refers to devices or software that condense acquired information into a concise format and provide it to the user visually.

[0193] This invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user comments in real time. Applying this to factory robots, the following system is realized to collect problems and improvement suggestions in real time on the production floor.

[0194] 1. System Program

[0195] The system consists of the following main processing steps:

[0196] 1. Input a rebate topic (problems or improvement suggestions in the production environment), and collect, summarize, and display related information.

[0197] 2. The system performs speech recognition on the user's speech in real time and records the resulting text data.

[0198] 3. Extract keywords from the speech-recognized text, collect and analyze related information, and generate an appropriate response.

[0199] 4. Summarize the collected information and present it to the user visually.

[0200] 2. Explain the program's processing in natural language.

[0201] Input, gather, summarize, and display information on rebate topics.

[0202] The server collects relevant information from the internet and internal databases based on the rebate topic entered by the user. This process utilizes major search engine APIs and internal database querying techniques. The collected information is summarized using a transformer model and stored in the database. Subsequently, the terminal (factory robot) displays this summarized information and provides it to the on-site operator.

[0203] Speech recognition and text conversion of user speech.

[0204] The user's (operator's) speech is collected through the robot's microphone and converted into text data using the speech_recognition library. By utilizing Google's speech recognition service, highly accurate speech-to-text conversion is achieved.

[0205] Keyword extraction, collection and analysis of related information

[0206] Key keywords are extracted from the speech-recognized text. This is done using natural language processing techniques and simple text segmentation algorithms. Based on the extracted keywords, relevant information is collected from the Google Search API and other databases. The collected information is then summarized again using a transformer model to generate an appropriate response.

[0207] Display summary information

[0208] Ultimately, the server generates summarized information and provides it to the user through the terminal's display. This process allows the user to obtain specific and immediate solutions.

[0209] 3. Add specific examples

[0210] As a concrete example, let's consider the problem of frequent defective products on a production line. When an operator asks the robot, "The defect rate of our products has been increasing recently. What could be the cause?", the robot analyzes the statement and gathers relevant information. Finally, it responds, "It may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening the inspection process."

[0211] Example of a prompt

[0212] "We've encountered a problem where the number of defective products has suddenly increased during the manufacturing process. Please gather information related to this issue and propose appropriate solutions."

[0213] This will enable the rapid and specific provision of information at the production site.

[0214] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0215] Step 1:

[0216] The server receives a rebate topic as input from the user. This rebate topic serves as the basis for collecting relevant information from the internet and internal databases. For example, if a user enters "causes of defective products" on a production line, information collection will begin based on this topic.

[0217] Step 2:

[0218] The server collects relevant information based on rebate topics. This information collection utilizes the Google Search API and internal database querying techniques. The collected data is summarized using natural language processing techniques. Specifically, key points are extracted using a transformer model.

[0219] Step 3:

[0220] The server stores the summarized information in a database and then sends it to the terminal (factory robot). This summarized information is displayed on the terminal's screen, allowing the user (operator) to visually confirm it.

[0221] Step 4:

[0222] The user speaks about problems and suggestions for improvement. This speech is collected by the robot's microphone. For example, they might ask, "The product defect rate has been increasing recently. What is the reason?"

[0223] Step 5:

[0224] The server converts the voice data transmitted from the robot into text data using speech recognition technology (e.g., the speech_recognition library). This conversion process uses Google's speech recognition service to achieve high accuracy in text transcription.

[0225] Step 6:

[0226] The server extracts important keywords from the speech-recognized text. This keyword extraction uses natural language processing techniques and simple text segmentation algorithms. For example, words like "defect rate" and "cause" might be extracted.

[0227] Step 7:

[0228] The server then collects information again based on the extracted keywords. This information gathering again uses the Google Search API and internal database queries. The collected information is similarly summarized using a transformer model.

[0229] Step 8:

[0230] The server analyzes the summarized information and generates an appropriate response. The generated response is sent to the terminal in text format. For example, it might offer specific improvement suggestions such as, "This may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening your inspection process."

[0231] Step 9:

[0232] The terminal displays the generated response on its screen for the user to review. Based on the displayed information, the user considers and implements specific countermeasures.

[0233] Step 10:

[0234] The server analyzes the entire rebate process from the audio recording and generates a summary as meeting minutes. This summary is then saved back to the database and can be viewed and modified by the user on their terminal.

[0235] Step 11:

[0236] The server generates presentation materials based on the meeting minutes. The generated materials are provided in a format that includes charts and graphs, and users can make final reviews and revisions.

[0237] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0238] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement with an emotion engine that recognizes the user's emotions in real time. This makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[0239] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects and summarizes relevant information from the internet and internal databases. The collected information is summarized and stored in the database. Subsequently, the device (tablet or robot) prepares to provide the summarized information to the user (student).

[0240] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. The user's speech is then analyzed using a keyword extraction algorithm to identify important keywords.

[0241] In addition, the emotion engine, along with the user's statements,

[0242] The system also recognizes and records user emotions in real time. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis, and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[0243] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, if a user says, "Renewable energy is expensive," and the emotion engine detects "anxiety," the robot will provide a response such as, "While the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[0244] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[0245] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final presentation material.

[0246] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers long-term cost savings and is therefore reassuring," which the terminal (rebate robot) responds to. Subsequently, the entire rebate is analyzed, and a summarized meeting transcript and presentation materials are generated. Finally, the user (student) and the rebate robot refine the materials to complete them.

[0247] The above is a specific embodiment of a rebate robot system that incorporates an emotion engine. This system allows for efficient preparation, execution, summarization, and presentation preparation of rebates, and because it also takes the user's emotional state into consideration, it can achieve a deeper understanding and empathy.

[0248] The following describes the processing flow.

[0249] Step 1:

[0250] The server receives the rebate topic entered by the user (teacher). For example, the user (teacher) enters the topic "The pros and cons of promoting renewable energy."

[0251] Step 2:

[0252] The server collects relevant information from the internet and internal databases based on the received rebate topic. This process utilizes APIs and web scraping techniques. The collected information is summarized using natural language processing (NLP) algorithms.

[0253] Step 3:

[0254] The device (tablet or rebate robot) displays summarized information to the user (student), allowing the student to prepare for the rebate.

[0255] Step 4:

[0256] At the start of the rebate, the server uses speech recognition technology in real time to convert the user's (student's) statements into text data and begins recording.

[0257] Step 5:

[0258] The emotion engine analyzes the user's (student's) speech, recognizing and recording their emotional state (e.g., joy, anxiety, anger) in real time. The analysis uses data such as voice tone, facial expression analysis, and heart rate.

[0259] Step 6:

[0260] The server analyzes the text and sentiment data of the statements. It uses a keyword extraction algorithm to identify important keywords from the statements.

[0261] Step 7:

[0262] The terminal (rebate robot) generates an appropriate response based on the analysis results sent from the server. For example, in response to the statement "Renewable energy is expensive" and the emotional state "anxiety," it generates a response such as "Although the initial investment in solar power generation is high, it has cost-saving effects in the long term, so you can rest assured."

[0263] Step 8:

[0264] After the rebate session ends, the server analyzes the entire audio recording and summarizes the key points and the user's emotional state. The summary is then compiled into meeting minutes.

[0265] Step 9:

[0266] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make any necessary corrections.

[0267] Step 10:

[0268] The server generates a presentation template based on the reviewed and revised meeting minutes. This template includes key points of the discussion, data, charts, graphs, and information on sentiment.

[0269] Step 11:

[0270] The terminal displays a generated document template to the user (student). The user (student) then collaborates with the rebate robot to refine the document's content. For example, they might input additional data or change the format of graphs.

[0271] Step 12:

[0272] The user (student) reviews and saves the final presentation materials. The device then provides these final materials for presentation.

[0273] The above outlines the specific processing steps of the rebate robot system that incorporates an emotion engine. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates while taking into account the user's emotional state.

[0274] (Example 2)

[0275] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0276] In rebate processes, it is difficult to analyze users' statements and emotional states in real time and provide appropriate responses and feedback based on that analysis. Furthermore, tasks such as summarizing the content after the rebate, generating meeting minutes, creating presentation materials, and refining them are time-consuming and labor-intensive, making them difficult to perform efficiently.

[0277] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying the summarized information, means for recognizing user statements in real time and converting them into text data, means for analyzing user statements and extracting keywords, means for recognizing and recording emotional states, means for generating appropriate responses based on user statements and emotional data, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation material templates based on meeting minutes, and means for refining presentation material templates. This makes it possible to provide efficient and appropriate responses while considering the user's statements and emotional states during the rebate process, and to efficiently carry out post-rebate work.

[0278] A "rebuttal topic" refers to the subject or theme of a rebuttal, and the content that will be discussed in that rebuttal.

[0279] "Means for collecting and summarizing information" refers to a system or method for searching for relevant information from the internet or internal databases and summarizing that information concisely using natural language processing technology.

[0280] "Means of displaying information" refers to an interface that visually provides users with access to collected and summarized information.

[0281] The "voice recognition means" is a system or method that receives voice input and converts it into text data.

[0282] The "keyword extraction means" is a system or method that automatically identifies and extracts important words and phrases from text data.

[0283] The "emotion recognition means" is a system or method that identifies and records the user's emotional state from multiple information sources such as voice, facial expressions, and biometric data.

[0284] The "response generation means" is a system or method that automatically creates appropriate responses or feedback based on the user's speech content and emotional state.

[0285] The "means for analyzing and summarizing recorded data" is a system or method for analyzing voice data and concisely summarizing its content.

[0286] The "means for displaying meeting minutes" is an interface for presenting the summarized meeting content in a form that can be viewed by the user.

[0287] The "means for generating presentation material templates" is a system or method that automatically creates templates containing key points, data, graphs, and charts required for presentations based on the summarized meeting minutes. s

[0288] The "means for enhancing presentation material templates" is a support system or method for the user to edit the generated templates and complete them as final presentation materials.

[0289] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement, with an emotion engine that recognizes the user's emotions in real time. This system makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[0290] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and internal databases and summarizes it using natural language processing techniques. This technique uses common natural language processing models, such as BERT or GPT-4®. The collected information is summarized and stored in the database. Subsequently, the terminal (tablet or robot) prepares to provide the summarized information to the user (student).

[0291] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. For speech recognition, a service such as Google Speech-to-Text is used. The user's speech is analyzed, and key keywords are identified using keyword extraction algorithms (e.g., TF-IDF or word cloud).

[0292] In addition, the emotion engine recognizes and records the user's emotional state in real time along with their statements. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis (e.g., Azure® Face API), and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[0293] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, a response generated by the server might include something like, "Although the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[0294] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[0295] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final version.

[0296] To give a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers cost-saving benefits in the long term, so it's reassuring," and the terminal (rebate robot) responds. Subsequently, the entire rebate is analyzed, a summarized meeting minute and presentation materials are generated, and finally, the user (student) and the rebate robot refine the materials to complete them.

[0297] Examples of prompts for a generative AI model include:

[0298] "Regarding the rebate topic 'The pros and cons of widespread renewable energy,' generate an appropriate response using student statements and sentiment data. A student stated, 'Renewable energy is expensive,' and the sentiment engine detected 'anxiety.' Provide the optimal response."

[0299] It can be used as such.

[0300] The flow of the specific process in Example 2 will be described with reference to FIG. 13.

[0301] Step 1: Input of Rebate Topic

[0302] Input: The user (teacher) uses the terminal to input a rebate topic.

[0303] Specific operation: The user (teacher) inputs a topic such as "The pros and cons of the popularization of renewable energy" into the interface of the terminal. This information is sent from the terminal to the server.

[0304] Output: The input topic information is received by the server.

[0305] Step 2: Collection and Summarization of Related Information

[0306] Input: The rebate topic information received by the server.

[0307] Specific operation: The server searches for related information from the Internet or internal databases and summarizes the information using natural language processing technologies (e.g., BERT or GPT-4).

[0308] <​​​​​​​​​​​​​​​​​​​​Output: Summary information in a format viewable by the user (student).

[0314] Step 4: Initiating the rebate and recognizing the statement

[0315] Input: The rebate process begins, and the user's (student's) comments are collected via the microphone.

[0316] Specific operation: The server uses speech recognition technology (e.g., Google Speech-to-Text) to convert and record speech data into text data in real time.

[0317] Data processing: Real-time text conversion of audio data.

[0318] Output: Text data of the user's (student's) statements.

[0319] Step 5: Recognizing the user's emotions

[0320] Input: User (student) speech data and biometric data (voice tone, facial expressions, heart rate).

[0321] Specific operation: The emotion engine analyzes this data to determine the user's (student's) emotional state. Sentiment Analysis technology is used for voice tone analysis, facial expression analysis APIs for facial expression analysis, and biosensors for biometric data analysis.

[0322] Output: User (student) emotion recognition data.

[0323] Step 6: Generating the response

[0324] Input: Transcribed speech data and sentiment recognition data.

[0325] Specific operation: The terminal (rebate robot) uses an NLP model to generate an appropriate response based on statements and sentiment data from the server.

[0326] Data processing: A response generation process that takes emotional states into consideration.

[0327] Output: The generated response.

[0328] Step 7: Analysis and summary after the rebate is completed.

[0329] Input: Voice and emotion data collected during rebates.

[0330] Specific operation: The server analyzes the recorded data and summarizes the rebate details and emotional state. The recorded data is then subjected to speech recognition again, converted to text, and analyzed.

[0331] Data processing: The process of summarizing from multiple data sources.

[0332] Output: Summarized meeting minutes data.

[0333] Step 8: View and review the meeting minutes.

[0334] Input: Summarized meeting minutes data.

[0335] Specific operation: The terminal displays the meeting minutes, and the user (teacher) reviews the content and makes corrections as needed.

[0336] Output: Reviewed and revised meeting minutes.

[0337] Step 9: Generating presentation materials

[0338] Input: Reviewed and revised meeting minutes.

[0339] Specific operation: The server automatically generates a presentation template based on the meeting minutes, including key points, data, graphs, and charts.

[0340] Data processing: The process of visualizing data and creating templates.

[0341] Output: Generated presentation template.

[0342] Step 10: Refining the presentation materials

[0343] Input: Presentation material template.

[0344] Specific operation: The user (student) edits materials together with the rebate robot and refines them into final presentation materials.

[0345] Output: Final presentation materials.

[0346] (Application Example 2)

[0347] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0348] Traditional rebate systems and in-store customer service systems have difficulty recognizing and analyzing user statements in real time and generating responses that take into account the user's emotional state. This has resulted in challenges in providing services that accurately capture user needs and emotions, hindering improvements in customer satisfaction and efficient business negotiations.

[0349] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the presentation materials with the user, means for analyzing conversations with customers in real time during business negotiations and converting customer statements into text data, means for determining the content of customer statements and emotional state and generating appropriate responses based thereon, and means for displaying the generated responses on the store clerk's terminal. This makes it possible to analyze user statements and emotions in real time and provide accurate responses in rebate and in-store customer service.

[0350] A "rebate topic" refers to the subject matter that becomes the theme of a discussion or debate.

[0351] "Real-time recognition" refers to the immediate recording and analysis of speech and actions.

[0352] "Text data" refers to a data format in which audio is converted into written information.

[0353] "Emotional state" refers to the user's emotional and psychological state.

[0354] "Appropriate response" refers to providing the most suitable answer based on the user's statements and emotions.

[0355] "Meeting minutes" refers to a record that summarizes the content of a meeting or discussion.

[0356] "Presentation materials" refer to explanatory documents created for purposes such as rebates or presentations.

[0357] "Brush-up" refers to the process of improving existing materials and information to create a higher-quality version.

[0358] "Business negotiation" refers to dialogue and negotiation with customers in a physical store.

[0359] "Device" refers to a device such as a smartphone, tablet, or smart glasses.

[0360] Modes for carrying out the invention

[0361] This invention is a system that recognizes the customer's statements and emotional state in real time during business negotiations with customers who visit the store, and generates appropriate responses. First, the server has a means of inputting a rebate topic and collects and summarizes information related to the topic from the internet or an internal database. The server sends this summarized information to a terminal, and the terminal displays the information.

[0362] Next, the device converts the customer's speech into text data in real time using speech recognition technology. The Python library `speech_recognition` is used for this speech recognition. Furthermore, the device analyzes the user's speech and uses an emotion engine to determine the customer's emotions. This emotion engine analyzes the user's emotional state based on voice tone analysis, facial expression analysis, and biometric data such as heart rate.

[0363] Based on the analyzed utterances and emotional state, the device generates an appropriate response. A generative AI model is used for response generation, ensuring the user receives the most suitable reply. These processes are seamlessly coordinated between the server and the device, utilizing devices such as smartphones, tablets, or smart glasses.

[0364] For example, if a customer visiting an electronics store asks, "Isn't this TV too expensive?", the terminal uses speech recognition technology to convert this statement into text data, and an emotion engine detects "anxiety" from the customer's tone of voice. Subsequently, a generative AI model determines that the customer's statement, "I feel the price of the new TV is too high," is anxiety-inducing and generates an appropriate response: "The initial price is on the higher side, but in the long run, you can save on electricity costs, making it a good deal overall." This response is immediately displayed on the terminal, allowing store employees to refer to it while assisting the customer.

[0365] A concrete example of a prompt message is: "I feel that the price of the new TV is too high." This prompt should be interpreted as a sign of concern, and an appropriate response should be generated to reassure the customer. Using such concrete examples enables the effective implementation of this system.

[0366] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0367] Step 1:

[0368] The server receives rebate topics from users (store clerks). The rebate topics entered by the user are sent to the server. The input data includes the topic name and a detailed description.

[0369] Step 2:

[0370] The server collects and summarizes relevant information from the internet and internal databases based on the received rebate topic. The collected data is summarized in text format, with key points extracted by a summarization algorithm. The summarized information is stored on the server. Input data is the raw data of the relevant information, while output data is the summarized text data.

[0371] Step 3:

[0372] The server sends summarized information to a device (smartphone, tablet, smart glasses, etc.). The device displays the received summarized information on its screen. The input data is summarized text data, and the output data is data in a format that can be displayed on the screen.

[0373] Step 4:

[0374] The device uses its microphone to capture what the user (customer) says aloud. The input data is raw audio data.

[0375] Step 5:

[0376] The device converts the acquired audio data into text data using the speech_recognition library. This speech recognition process captures the customer's spoken content in text format. The input data is audio data, and the output data is text data.

[0377] Step 6:

[0378] The device uses a generative AI model and an emotion engine to analyze acquired text data and determine the customer's emotional state. This analysis utilizes data such as voice tone, facial expression analysis, and heart rate. Input data consists of text data and biometric data, while output data is the analysis result (spoken content and emotional state).

[0379] Step 7:

[0380] The device generates an appropriate response using a generative AI model based on the analyzed utterance and emotional state. For example, if the utterance includes the keyword "too expensive" and the emotion of anxiety is detected, the response generation algorithm will generate an appropriate response such as "The initial price is a bit high, but it's a good deal in the long run." The input data is the analysis result, and the output data is the generated response.

[0381] Step 8:

[0382] The terminal displays the generated response on the user's (store clerk's) screen. This allows the clerk to immediately confirm and respond to the customer appropriately. The input data is the generated response, and the output data is the displayed text information.

[0383] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0384] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0385] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0386] [Second Embodiment]

[0387] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0388] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0389] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0390] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0391] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0392] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0393] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0394] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0395] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0396] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0397] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0398] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0399] This invention relates to a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. This embodiment details how the system achieves these functions.

[0400] The server first receives a rebate topic entered by the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and its internal database. The collected information is summarized and stored in the server's database. Subsequently, the terminal (tablet or robot) displays this summarized information and provides it to the user (student).

[0401] When the rebate program begins, the server records each user's statements as text data in real time using speech recognition technology. The recorded statements are analyzed by a keyword extraction algorithm, and the server generates an appropriate response. For example, if a user (student) says, "Renewable energy is expensive," the server analyzes this statement and, based on the keyword "cost," generates a response such as, "While the initial investment in solar power generation is high, it has cost-saving effects in the long term."

[0402] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate. This summary is compiled into meeting minutes and displayed on the terminal. The meeting minutes can be reviewed and edited by the user (teacher).

[0403] Next, the server generates a presentation template based on the meeting minutes data. This template is provided in a format that includes charts and graphs and is displayed on the device (tablet or robot). The user (student) reviews the content of the material together with the rebate robot using this template and makes improvements.

[0404] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive." In response to this statement, the server generates an appropriate response, and the terminal (rebate robot) replies, "While the initial investment for solar power generation is high, it has cost-saving effects in the long term." Subsequently, the server analyzes the entire rebate, generates a summarized meeting minutes, and creates a presentation template. Based on this material, the user (student) reviews the final slides with the robot and makes revisions.

[0405] As described above, the present invention is a system that provides consistent support for rebates, from preparation and implementation to summarization and presentation preparation, thereby improving the quality of rebates and enabling efficient class management.

[0406] The following describes the processing flow.

[0407] Step 1:

[0408] The server receives a rebate topic from the user (teacher). The user (teacher) enters the rebate topic "The pros and cons of promoting renewable energy" into the server.

[0409] Step 2:

[0410] The server collects relevant information based on the received rebate topic. This collection involves accessing APIs to retrieve data from the internet and accessing internal databases. However, the collected raw information is not usable as is, so natural language processing (NLP) algorithms are used to summarize the key points and convert it into an easily understandable format.

[0411] Step 3:

[0412] The device (tablet or rebate robot) prepares to provide summarized information to the user (student). At this point, the information is displayed on the tablet or robot's screen.

[0413] Step 4:

[0414] The user (student) begins the rebate. The server uses speech recognition technology to convert each student's statements into text data in real time and records them. During this process, a time stamp of the spoken content is also saved.

[0415] Step 5:

[0416] The server analyzes the user's statements. Using a keyword extraction algorithm, it identifies important keywords from the statements. For example, from the statement "Renewable energy is expensive," it extracts "cost" as an important keyword.

[0417] Step 6:

[0418] The terminal (rebate robot) generates appropriate responses based on analysis information sent from the server. It provides opinions from various perspectives and deepens the discussion. For example, it might generate a response such as, "The initial investment for solar power generation is high, but it has cost-saving effects in the long term."

[0419] Step 7:

[0420] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. Using natural language processing algorithms, it extracts the main points of the discussion and summarizes the key takeaways.

[0421] Step 8:

[0422] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make corrections as needed.

[0423] Step 9:

[0424] The server generates a presentation template based on the meeting minutes. This template includes key points, data, charts, and graphs discussed during the discussion.

[0425] Step 10:

[0426] The terminal displays the generated document template to the user (student). The user (student) reviews the document content together with the rebate robot and makes improvements. For example, they can add data or charts and change the format to make it easier to understand.

[0427] Step 11:

[0428] The user (student) reviews the completed materials and saves them as the final presentation materials. The device provides the completed materials for the presentation day.

[0429] The above is a detailed explanation of the processing steps of the rebate robot system. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates.

[0430] (Example 1)

[0431] Next, we will describe Example 1. 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".

[0432] Conventional rebate support systems do not consistently handle the entire process, from collecting and summarizing information related to rebate topics, to recording and analyzing statements, generating appropriate responses, creating meeting minutes, and preparing presentation materials. As a result, the process from rebate preparation to implementation and summarization is inefficient, leading to a decline in the quality of rebates.

[0433] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0434] In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the generated presentation materials with the user, means for collecting relevant information from the internet or an internal database based on the rebate topic entered by the user, means for summarizing the collected information and storing it in a database, means for using natural language processing technology in the collection and summarization process, means for accessing external data sources via an internet connection, means for recording each user's statements as text data in real time using speech recognition technology, means for analyzing the recorded statements with a keyword extraction algorithm, means for using a generative AI model to generate responses based on keywords, and means for inputting prompt sentences into the generative AI model to generate appropriate responses. This makes it possible to consistently and efficiently carry out everything from rebate preparation and implementation to summarization and presentation preparation.

[0435] A "rebate topic" is the subject or theme discussed during a rebate (debate).

[0436] "Users" refer to the people who use the system, specifically teachers and students.

[0437] "Means of information gathering" refers to methods and technologies for obtaining information related to rebate topics from the internet or internal databases.

[0438] "Means of summarization" refer to methods and techniques for concisely organizing collected information.

[0439] "Means of display" refers to methods and techniques for presenting collected and summarized information to the user.

[0440] "Means of recognition and recording" refers to methods and technologies for real-time speech recognition of a user's speech and saving it as text data.

[0441] "Means of analysis" refer to methods and techniques for analyzing recorded text data and extracting meaning and keywords.

[0442] "Means for generating appropriate responses" refer to methods and technologies for responding to users based on analysis results.

[0443] "Means of analyzing audio recordings" refers to methods and techniques for analyzing audio recordings of rebates and extracting important statements and key points.

[0444] "Means of displaying meeting minutes" refers to methods and technologies for displaying summarized meeting minutes to users.

[0445] "Means for generating presentation materials" refers to methods and techniques for creating presentation materials based on meeting minutes data.

[0446] "Methods for refinement" refer to methods and techniques for reviewing generated presentation materials together with the user and making corrections and improvements as needed.

[0447] An "external data source" refers to an information source located outside the system, such as publicly available information on the internet or external databases.

[0448] "Natural language processing technology" refers to artificial intelligence technology used to analyze and summarize collected information and user statements.

[0449] "Speech recognition technology" refers to the technology used to convert speech data into text data.

[0450] A "keyword extraction algorithm" is a computational method or technique for selecting important words and phrases from text data.

[0451] A "generative AI model" is an artificial intelligence model designed to generate appropriate responses based on user statements.

[0452] A "prompt statement" is a sentence input to a generative AI model that provides instructions or context for generating a specific response.

[0453] The present invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. Specific embodiments of the present invention are described below.

[0454] 1. Basic System Configuration

[0455] This system consists of a server, terminals (tablets and robots), and users (teachers and students). The server plays a central role in information collection and analysis, while the terminals provide information to users and accept user input.

[0456] Hardware and software to be used

[0457] Server: A server computer with a high-performance processor and large memory capacity is recommended. The server accesses external data sources via an internet connection and manages the internal database.

[0458] Internet connection: A high-speed internet connection is required to facilitate data collection and access to external data sources.

[0459] Speech recognition technology: The Google Speech-to-Text API is used to convert user speech into text data in real time.

[0460] Natural Language Processing (NLP) technologies: Google Cloud's Natural Language API and Python libraries (e.g., BeautifulSoup, spaCy) are used. This is used to summarize relevant information and analyze user utterances.

[0461] Keyword extraction algorithm: TF-IDF and the spaCy library are used to extract important keywords from user utterances.

[0462] Generative AI Model: Generative AI models such as OpenAI's GPT-3 are used. This is to generate appropriate responses based on user statements.

[0463] Database: Relational databases such as MySQL and PostgreSQL are used to manage collected information, summary data, and meeting minutes.

[0464] 2. System Operation

[0465] The server first receives a rebate topic as input from the user (teacher). Based on this input, the server collects relevant information from the internet and its internal database. Google's search API and queries from the internal database are used for information collection.

[0466] The collected information is summarized using the Google Cloud Natural Language API. The summarized information is stored in the server's database.

[0467] Next, the device (tablet or robot) displays this summary information to the user (student). This allows the student to prepare for their rebate in advance.

[0468] Rebate start:

[0469] During the rebate process, the server records each user's speech in real time as text data using the Google Speech-to-Text API. This recorded data is then analyzed using TF-IDF and the spaCy library to extract important keywords.

[0470] Based on the analyzed data, the server inputs prompts into a generative AI model (e.g., OpenAI's GPT-3) to generate an appropriate response. For example, if a user says, "Renewable energy is expensive," the server inputs the prompt, "Generate a response regarding the cost of renewable energy," into the generative AI model, which then generates an appropriate response.

[0471] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. This summary is compiled into meeting minutes and displayed on the terminal. The user (teacher) reviews the minutes and makes corrections as needed.

[0472] Finally, the server generates a presentation template based on the meeting minutes. This template includes charts and graphs, and uses libraries such as Matplotlib and Pandas. This material is displayed on the terminal, and the user (student) reviews the content together with the rebate robot and makes final adjustments.

[0473] 3. Specific Examples

[0474] If the rebate topic is "The pros and cons of promoting renewable energy," the user (teacher) enters this topic into the server. The server uses the internet and its internal database to collect relevant information, summarizes data such as "data on the costs and efficiency of renewable energy implementation," and stores it in the database. The terminal displays this information to the students to help them prepare for their rebates.

[0475] During the rebate process, the user's (student's) statement, "Renewable energy is expensive," is recorded in real time as text data by the server, and the keyword "cost" is analyzed by a keyword extraction algorithm. The prompt "Generate a response about the cost of renewable energy" is input to the generative AI model (GPT-3), and the response "The initial investment for solar power generation is high, but it has cost-saving effects in the long term" is generated.

[0476] After the rebate process is complete, the server re-analyzes the audio data and creates meeting minutes. It then generates a presentation template, which the user reviews and revises together with the rebate robot.

[0477] Thus, this system makes it possible to efficiently handle the entire process of rebate preparation, from preparation and execution to summarization and presentation preparation.

[0478] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0479] Step 1:

[0480] The server receives a rebate topic as input from the user (teacher).

[0481] Input: Users enter rebate topics such as "The pros and cons of promoting renewable energy."

[0482] Data processing: The server saves this topic and begins searching for related information.

[0483] Output: Ready to explore

[0484] Step 2:

[0485] The server collects relevant information from the internet and internal databases based on the rebate topic.

[0486] Input: Rebate Topic

[0487] Data Processing: We collect the latest information from the internet using Google's search API and other tools, and also retrieve relevant information from our internal database.

[0488] Output: List of collected related information

[0489] Step 3:

[0490] The server summarizes the collected information and stores it in a database.

[0491] Input: List of collected information

[0492] Data processing: Text summarization is performed using the Google Cloud Natural Language API.

[0493] Output: Summarized information is saved to the database.

[0494] Step 4:

[0495] The device (tablet or robot) provides summary information to the user (student).

[0496] Input: Summary information stored in the database

[0497] Data processing: None

[0498] Output: Summary information will be displayed on the terminal screen.

[0499] Step 5:

[0500] The server recognizes and records each user's statements in real time after the rebate program begins.

[0501] Input: User's voice utterance

[0502] Data processing: Convert speech to text using the Google Speech-to-Text API.

[0503] Output: Real-time recorded text data

[0504] Step 6:

[0505] The server analyzes the recorded statements and generates an appropriate response.

[0506] Input: Text data recorded in real time

[0507] Data processing: Extract keywords using TF-IDF or the spaCy library.

[0508] Output: Keyword extraction results

[0509] Step 7:

[0510] The server uses a generative AI model to generate an appropriate response.

[0511] Input: Keyword extraction results

[0512] Data processing: Input prompt text into a generative AI model (e.g., OpenAI GPT-3) and generate a response.

[0513] Output: Response generated by the generative AI model

[0514] Specific example: If a student says "Renewable energy is expensive" during a rebate session, and the prompt is "Generate a response about the cost of renewable energy," the response will be "The initial investment for solar power is high, but it has cost-saving effects in the long term."

[0515] Step 8:

[0516] After the rebate is completed, the server analyzes the recorded data, summarizes it, and creates meeting minutes.

[0517] Input: Rebate recording data

[0518] Data processing: Summarize the audio data using a text analysis algorithm.

[0519] Output: Summarized meeting minutes

[0520] Step 9:

[0521] The terminal displays summarized meeting minutes to the user (teacher) and assists with review and correction.

[0522] Input: Summarized meeting minutes

[0523] Data processing: None

[0524] Output: The meeting minutes will be displayed on the terminal screen.

[0525] Step 10:

[0526] The server generates presentation material templates based on the meeting minutes.

[0527] Input: Summarized meeting minutes

[0528] Data processing: Create charts and graphs using Matplotlib and Pandas.

[0529] Output: Presentation material template

[0530] Step 11:

[0531] The device provides the generated presentation materials to the user (student) and supports them in refining their work.

[0532] Input: Presentation material template

[0533] Data processing: Provides an interface to reflect necessary modifications.

[0534] Output: Finalized presentation materials

[0535] Through the above processing steps, this system can efficiently handle everything from preparing and executing rebates to summarizing and preparing presentations.

[0536] (Application Example 1)

[0537] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0538] In modern factory settings, there is a need to quickly identify problems on the production line and propose improvements. However, there is a lack of means to identify problems in real time and provide appropriate information, which can lead to decreased production efficiency. In particular, there is a need for a system that provides quick and concrete solutions to production problems and troubles faced by operators.

[0539] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0540] In this invention, the server includes means for inputting rebate topics, means for collecting and summarizing information related to rebate topics, means for displaying information related to rebate topics, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on meeting minutes, means for refining presentation materials with the user, means for collecting problems and improvement suggestions at the production site in real time using speech recognition technology, means for extracting keywords from speech-recognized text and collecting related information, and means for summarizing and displaying the collected information. This enables the rapid and specific provision of information at the production site.

[0541] A "rebuttal topic" is a theme or issue that is discussed in a debate.

[0542] "Means of collecting and summarizing information" refers to methods and devices for obtaining relevant information from sources such as the internet or internal databases and compiling it into a concise format.

[0543] "Means of displaying information" refers to devices and software that visually provide collected and summarized information to users.

[0544] "Means for recognizing and recording user speech in real time" refers to a device or method that uses speech recognition technology to instantly save user speech as text data.

[0545] "Means for analyzing user utterances and generating appropriate responses" refers to methods or devices that extract keywords from speech-recognized text, analyze their content, and automatically create appropriate responses.

[0546] "Means for analyzing and summarizing debate recordings" refers to devices or methods for analyzing debate session recordings and concisely summarizing the main points of the discussion.

[0547] "Means of displaying meeting minutes" refers to devices or software that visually provide users with a summarized version of the discussion.

[0548] "Means for generating presentation materials" refers to devices or methods that automatically create presentation-appropriate forms and materials based on meeting minutes data.

[0549] "Means of refinement" refer to devices or methods that allow users to modify or improve the generated presentation materials.

[0550] "Methods for collecting problems and improvement suggestions in the production field in real time using voice recognition technology" refers to devices and technologies that recognize voice communication on the production line and collect that information immediately.

[0551] "Means for extracting keywords and collecting related information" refers to methods or devices that extract important words from speech-recognized text and search for and obtain related information based on those words.

[0552] "Means for summarizing and displaying collected information" refers to devices or software that condense acquired information into a concise format and provide it to the user visually.

[0553] This invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user comments in real time. Applying this to factory robots, the following system is realized to collect problems and improvement suggestions in real time on the production floor.

[0554] 1. System Program

[0555] The system consists of the following main processing steps:

[0556] 1. Input a rebate topic (problems or improvement suggestions in the production environment), and collect, summarize, and display related information.

[0557] 2. The system performs speech recognition on the user's speech in real time and records the resulting text data.

[0558] 3. Extract keywords from the speech-recognized text, collect and analyze related information, and generate an appropriate response.

[0559] 4. Summarize the collected information and present it to the user visually.

[0560] 2. Explain the program's processing in natural language.

[0561] Input, gather, summarize, and display information on rebate topics.

[0562] The server collects relevant information from the internet and internal databases based on the rebate topic entered by the user. This process utilizes major search engine APIs and internal database querying techniques. The collected information is summarized using a transformer model and stored in the database. Subsequently, the terminal (factory robot) displays this summarized information and provides it to the on-site operator.

[0563] Speech recognition and text conversion of user speech.

[0564] The user's (operator's) speech is collected through the robot's microphone and converted into text data using the speech_recognition library. By utilizing Google's speech recognition service, highly accurate speech-to-text conversion is achieved.

[0565] Keyword extraction, collection and analysis of related information

[0566] Key keywords are extracted from the speech-recognized text. This is done using natural language processing techniques and simple text segmentation algorithms. Based on the extracted keywords, relevant information is collected from the Google Search API and other databases. The collected information is then summarized again using a transformer model to generate an appropriate response.

[0567] Display summary information

[0568] Ultimately, the server generates summarized information and provides it to the user through the terminal's display. This process allows the user to obtain specific and immediate solutions.

[0569] 3. Add specific examples

[0570] As a concrete example, let's consider the problem of frequent defective products on a production line. When an operator asks the robot, "The defect rate of our products has been increasing recently. What could be the cause?", the robot analyzes the statement and gathers relevant information. Finally, it responds, "It may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening the inspection process."

[0571] Example of a prompt

[0572] "We've encountered a problem where the number of defective products has suddenly increased during the manufacturing process. Please gather information related to this issue and propose appropriate solutions."

[0573] This will enable the rapid and specific provision of information at the production site.

[0574] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0575] Step 1:

[0576] The server receives a rebate topic as input from the user. This rebate topic serves as the basis for collecting relevant information from the internet and internal databases. For example, if a user enters "causes of defective products" on a production line, information collection will begin based on this topic.

[0577] Step 2:

[0578] The server collects relevant information based on rebate topics. This information collection utilizes the Google Search API and internal database querying techniques. The collected data is summarized using natural language processing techniques. Specifically, key points are extracted using a transformer model.

[0579] Step 3:

[0580] The server stores the summarized information in a database and then sends it to the terminal (factory robot). This summarized information is displayed on the terminal's screen, allowing the user (operator) to visually confirm it.

[0581] Step 4:

[0582] The user speaks about problems and suggestions for improvement. This speech is collected by the robot's microphone. For example, they might ask, "The product defect rate has been increasing recently. What is the reason?"

[0583] Step 5:

[0584] The server converts the voice data transmitted from the robot into text data using speech recognition technology (e.g., the speech_recognition library). This conversion process uses Google's speech recognition service to achieve high accuracy in text transcription.

[0585] Step 6:

[0586] The server extracts important keywords from the speech-recognized text. This keyword extraction uses natural language processing techniques and simple text segmentation algorithms. For example, words like "defect rate" and "cause" might be extracted.

[0587] Step 7:

[0588] The server then collects information again based on the extracted keywords. This information gathering again uses the Google Search API and internal database queries. The collected information is similarly summarized using a transformer model.

[0589] Step 8:

[0590] The server analyzes the summarized information and generates an appropriate response. The generated response is sent to the terminal in text format. For example, it might offer specific improvement suggestions such as, "This may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening your inspection process."

[0591] Step 9:

[0592] The terminal displays the generated response on its screen for the user to review. Based on the displayed information, the user considers and implements specific countermeasures.

[0593] Step 10:

[0594] The server analyzes the entire rebate process from the audio recording and generates a summary as meeting minutes. This summary is then saved back to the database and can be viewed and modified by the user on their terminal.

[0595] Step 11:

[0596] The server generates presentation materials based on the meeting minutes. The generated materials are provided in a format that includes charts and graphs, and users can make final reviews and revisions.

[0597] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0598] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement with an emotion engine that recognizes the user's emotions in real time. This makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[0599] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects and summarizes relevant information from the internet and internal databases. The collected information is summarized and stored in the database. Subsequently, the device (tablet or robot) prepares to provide the summarized information to the user (student).

[0600] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. The user's speech is then analyzed using a keyword extraction algorithm to identify important keywords.

[0601] In addition, the emotion engine, along with the user's statements,

[0602] The system also recognizes and records user emotions in real time. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis, and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[0603] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, if a user says, "Renewable energy is expensive," and the emotion engine detects "anxiety," the robot will provide a response such as, "While the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[0604] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[0605] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final presentation material.

[0606] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers long-term cost savings and is therefore reassuring," which the terminal (rebate robot) responds to. Subsequently, the entire rebate is analyzed, and a summarized meeting transcript and presentation materials are generated. Finally, the user (student) and the rebate robot refine the materials to complete them.

[0607] The above is a specific embodiment of a rebate robot system that incorporates an emotion engine. This system allows for efficient preparation, execution, summarization, and presentation preparation of rebates, and because it also takes the user's emotional state into consideration, it can achieve a deeper understanding and empathy.

[0608] The following describes the processing flow.

[0609] Step 1:

[0610] The server receives the rebate topic entered by the user (teacher). For example, the user (teacher) enters the topic "The pros and cons of promoting renewable energy."

[0611] Step 2:

[0612] The server collects relevant information from the internet and internal databases based on the received rebate topic. This process utilizes APIs and web scraping techniques. The collected information is summarized using natural language processing (NLP) algorithms.

[0613] Step 3:

[0614] The device (tablet or rebate robot) displays summarized information to the user (student), allowing the student to prepare for the rebate.

[0615] Step 4:

[0616] At the start of the rebate, the server uses speech recognition technology in real time to convert the user's (student's) statements into text data and begins recording.

[0617] Step 5:

[0618] The emotion engine analyzes the user's (student's) speech, recognizing and recording their emotional state (e.g., joy, anxiety, anger) in real time. The analysis uses data such as voice tone, facial expression analysis, and heart rate.

[0619] Step 6:

[0620] The server analyzes the text and sentiment data of the statements. It uses a keyword extraction algorithm to identify important keywords from the statements.

[0621] Step 7:

[0622] The terminal (rebate robot) generates an appropriate response based on the analysis results sent from the server. For example, in response to the statement "Renewable energy is expensive" and the emotional state "anxiety," it generates a response such as "Although the initial investment in solar power generation is high, it has cost-saving effects in the long term, so you can rest assured."

[0623] Step 8:

[0624] After the rebate session ends, the server analyzes the entire audio recording and summarizes the key points and the user's emotional state. The summary is then compiled into meeting minutes.

[0625] Step 9:

[0626] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make any necessary corrections.

[0627] Step 10:

[0628] The server generates a presentation template based on the reviewed and revised meeting minutes. This template includes key points of the discussion, data, charts, graphs, and information on sentiment.

[0629] Step 11:

[0630] The terminal displays a generated document template to the user (student). The user (student) then collaborates with the rebate robot to refine the document's content. For example, they might input additional data or change the format of graphs.

[0631] Step 12:

[0632] The user (student) reviews and saves the final presentation materials. The device then provides these final materials for presentation.

[0633] The above outlines the specific processing steps of the rebate robot system that incorporates an emotion engine. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates while taking into account the user's emotional state.

[0634] (Example 2)

[0635] Next, we will describe Example 2. 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".

[0636] In rebate processes, it is difficult to analyze users' statements and emotional states in real time and provide appropriate responses and feedback based on that analysis. Furthermore, tasks such as summarizing the content after the rebate, generating meeting minutes, creating presentation materials, and refining them are time-consuming and labor-intensive, making them difficult to perform efficiently.

[0637] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying the summarized information, means for recognizing user statements in real time and converting them into text data, means for analyzing user statements and extracting keywords, means for recognizing and recording emotional states, means for generating appropriate responses based on user statements and emotional data, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation material templates based on meeting minutes, and means for refining presentation material templates. This makes it possible to provide efficient and appropriate responses while considering the user's statements and emotional states during the rebate process, and to efficiently carry out post-rebate work.

[0638] A "rebuttal topic" refers to the subject or theme of a rebuttal, and the content that will be discussed in that rebuttal.

[0639] "Means for collecting and summarizing information" refers to a system or method for searching for relevant information from the internet or internal databases and summarizing that information concisely using natural language processing technology.

[0640] "Means of displaying information" refers to an interface that visually provides users with access to collected and summarized information.

[0641] "Speech recognition means" refers to a system or method that receives speech input and converts it into text data.

[0642] A "keyword extraction method" is a system or method that automatically identifies and extracts important words and phrases from text data.

[0643] "Emotion recognition means" refers to a system or method that identifies and records a user's emotional state from multiple information sources such as voice, facial expressions, and biometric data.

[0644] A "response generation means" is a system or method that automatically generates an appropriate response or feedback based on the user's statements and emotional state.

[0645] "Means for analyzing and summarizing recorded data" refers to a system or method for analyzing audio data and concisely summarizing its contents.

[0646] "Means for displaying meeting minutes" refers to an interface for presenting summarized rebate details in a format that users can view.

[0647] "Means for generating presentation material templates" refers to a system or method that automatically creates a template containing the necessary points, data, graphs, and charts for a presentation, based on summarized meeting minutes.

[0648] "Methods for refining presentation material templates" refers to support systems or methods that allow users to edit generated templates and complete them as final presentation materials.

[0649] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement, with an emotion engine that recognizes the user's emotions in real time. This system makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[0650] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and internal databases and summarizes it using natural language processing techniques. This technique uses common natural language processing models, such as BERT or GPT-4. The collected information is summarized and stored in the database. The terminal (tablet or robot) then prepares to provide the summarized information to the user (student).

[0651] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. For speech recognition, a service such as Google Speech-to-Text is used. The user's speech is analyzed, and key keywords are identified using keyword extraction algorithms (e.g., TF-IDF or word cloud).

[0652] In addition, the emotion engine recognizes and records the user's emotional state in real time along with their statements. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis (e.g., Azure Face API), and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[0653] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, a response generated by the server might include something like, "Although the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[0654] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[0655] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final version.

[0656] To give a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers cost-saving benefits in the long term, so it's reassuring," and the terminal (rebate robot) responds. Subsequently, the entire rebate is analyzed, a summarized meeting minute and presentation materials are generated, and finally, the user (student) and the rebate robot refine the materials to complete them.

[0657] Examples of prompts for a generative AI model include:

[0658] "Regarding the rebate topic 'The pros and cons of widespread renewable energy,' generate an appropriate response using student statements and sentiment data. A student stated, 'Renewable energy is expensive,' and the sentiment engine detected 'anxiety.' Provide the optimal response."

[0659] It can be used as such.

[0660] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0661] Step 1: Enter the rebate topic.

[0662] Input: The user (teacher) enters the rebate topic using a terminal.

[0663] Specific operation: The user (teacher) enters a topic such as "the pros and cons of promoting renewable energy" into the terminal interface. This information is sent from the terminal to the server.

[0664] Output: The input topic information is received by the server.

[0665] Step 2: Gather and summarize relevant information

[0666] Input: Rebate topic information received by the server.

[0667] Specific operation: The server searches for relevant information from the internet and internal databases, and summarizes the information using natural language processing techniques (e.g., BERT or GPT-4).

[0668] Data processing: Analyze data from the sources used, extract necessary information, and summarize it.

[0669] Output: The summarized information is saved to the database.

[0670] Step 3: Prepare to provide summary information

[0671] Input: Summarized information.

[0672] Specific operation: The terminal receives summary information from the server and prepares it to display the information in a format that is easy for the user (student) to use.

[0673] Output: Summary information in a format viewable by the user (student).

[0674] Step 4: Initiating the rebate and recognizing the statement

[0675] Input: The rebate process begins, and the user's (student's) comments are collected via the microphone.

[0676] Specific operation: The server uses speech recognition technology (e.g., Google Speech-to-Text) to convert and record speech data into text data in real time.

[0677] Data processing: Real-time text conversion of audio data.

[0678] Output: Text data of the user's (student's) statements.

[0679] Step 5: Recognizing the user's emotions

[0680] Input: User (student) speech data and biometric data (voice tone, facial expressions, heart rate).

[0681] Specific operation: The emotion engine analyzes this data to determine the user's (student's) emotional state. Sentiment Analysis technology is used for voice tone analysis, facial expression analysis APIs for facial expression analysis, and biosensors for biometric data analysis.

[0682] Output: User (student) emotion recognition data.

[0683] Step 6: Generating the response

[0684] Input: Transcribed speech data and sentiment recognition data.

[0685] Specific operation: The terminal (rebate robot) uses an NLP model to generate an appropriate response based on statements and sentiment data from the server.

[0686] Data processing: A response generation process that takes emotional states into consideration.

[0687] Output: The generated response.

[0688] Step 7: Analysis and summary after the rebate is completed.

[0689] Input: Voice and emotion data collected during rebates.

[0690] Specific operation: The server analyzes the recorded data and summarizes the rebate details and emotional state. The recorded data is then subjected to speech recognition again, converted to text, and analyzed.

[0691] Data processing: The process of summarizing from multiple data sources.

[0692] Output: Summarized meeting minutes data.

[0693] Step 8: View and review the meeting minutes.

[0694] Input: Summarized meeting minutes data.

[0695] Specific operation: The terminal displays the meeting minutes, and the user (teacher) reviews the content and makes corrections as needed.

[0696] Output: Reviewed and revised meeting minutes.

[0697] Step 9: Generating presentation materials

[0698] Input: Reviewed and revised meeting minutes.

[0699] Specific operation: The server automatically generates a presentation template based on the meeting minutes, including key points, data, graphs, and charts.

[0700] Data processing: The process of visualizing data and creating templates.

[0701] Output: Generated presentation template.

[0702] Step 10: Refining the presentation materials

[0703] Input: Presentation material template.

[0704] Specific operation: The user (student) edits materials together with the rebate robot and refines them into final presentation materials.

[0705] Output: Final presentation materials.

[0706] (Application Example 2)

[0707] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0708] Traditional rebate systems and in-store customer service systems have difficulty recognizing and analyzing user statements in real time and generating responses that take into account the user's emotional state. This has resulted in challenges in providing services that accurately capture user needs and emotions, hindering improvements in customer satisfaction and efficient business negotiations.

[0709] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the presentation materials with the user, means for analyzing conversations with customers in real time during business negotiations and converting customer statements into text data, means for determining the content of customer statements and emotional state and generating appropriate responses based thereon, and means for displaying the generated responses on the store clerk's terminal. This makes it possible to analyze user statements and emotions in real time and provide accurate responses in rebate and in-store customer service.

[0710] A "rebate topic" refers to the subject matter that becomes the theme of a discussion or debate.

[0711] "Real-time recognition" refers to the immediate recording and analysis of speech and actions.

[0712] "Text data" refers to a data format in which audio is converted into written information.

[0713] "Emotional state" refers to the user's emotional and psychological state.

[0714] "Appropriate response" refers to providing the most suitable answer based on the user's statements and emotions.

[0715] "Meeting minutes" refers to a record that summarizes the content of a meeting or discussion.

[0716] "Presentation materials" refer to explanatory documents created for purposes such as rebates or presentations.

[0717] "Brush-up" refers to the process of improving existing materials and information to create a higher-quality version.

[0718] "Business negotiation" refers to dialogue and negotiation with customers in a physical store.

[0719] "Device" refers to a device such as a smartphone, tablet, or smart glasses.

[0720] Modes for carrying out the invention

[0721] This invention is a system that recognizes the customer's statements and emotional state in real time during business negotiations with customers who visit the store, and generates appropriate responses. First, the server has a means of inputting a rebate topic and collects and summarizes information related to the topic from the internet or an internal database. The server sends this summarized information to a terminal, and the terminal displays the information.

[0722] Next, the device converts the customer's speech into text data in real time using speech recognition technology. The Python library `speech_recognition` is used for this speech recognition. Furthermore, the device analyzes the user's speech and uses an emotion engine to determine the customer's emotions. This emotion engine analyzes the user's emotional state based on voice tone analysis, facial expression analysis, and biometric data such as heart rate.

[0723] Based on the analyzed utterances and emotional state, the device generates an appropriate response. A generative AI model is used for response generation, ensuring the user receives the most suitable reply. These processes are seamlessly coordinated between the server and the device, utilizing devices such as smartphones, tablets, or smart glasses.

[0724] For example, if a customer visiting an electronics store asks, "Isn't this TV too expensive?", the terminal uses speech recognition technology to convert this statement into text data, and an emotion engine detects "anxiety" from the customer's tone of voice. Subsequently, a generative AI model determines that the customer's statement, "I feel the price of the new TV is too high," is anxiety-inducing and generates an appropriate response: "The initial price is on the higher side, but in the long run, you can save on electricity costs, making it a good deal overall." This response is immediately displayed on the terminal, allowing store employees to refer to it while assisting the customer.

[0725] A concrete example of a prompt message is: "I feel that the price of the new TV is too high." This prompt should be interpreted as a sign of concern, and an appropriate response should be generated to reassure the customer. Using such concrete examples enables the effective implementation of this system.

[0726] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0727] Step 1:

[0728] The server receives rebate topics from users (store clerks). The rebate topics entered by the user are sent to the server. The input data includes the topic name and a detailed description.

[0729] Step 2:

[0730] The server collects and summarizes relevant information from the internet and internal databases based on the received rebate topic. The collected data is summarized in text format, with key points extracted by a summarization algorithm. The summarized information is stored on the server. Input data is the raw data of the relevant information, while output data is the summarized text data.

[0731] Step 3:

[0732] The server sends summarized information to a device (smartphone, tablet, smart glasses, etc.). The device displays the received summarized information on its screen. The input data is summarized text data, and the output data is data in a format that can be displayed on the screen.

[0733] Step 4:

[0734] The device uses its microphone to capture what the user (customer) says aloud. The input data is raw audio data.

[0735] Step 5:

[0736] The device converts the acquired audio data into text data using the speech_recognition library. This speech recognition process captures the customer's spoken content in text format. The input data is audio data, and the output data is text data.

[0737] Step 6:

[0738] The device uses a generative AI model and an emotion engine to analyze acquired text data and determine the customer's emotional state. This analysis utilizes data such as voice tone, facial expression analysis, and heart rate. Input data consists of text data and biometric data, while output data is the analysis result (spoken content and emotional state).

[0739] Step 7:

[0740] The device generates an appropriate response using a generative AI model based on the analyzed utterance and emotional state. For example, if the utterance includes the keyword "too expensive" and the emotion of anxiety is detected, the response generation algorithm will generate an appropriate response such as "The initial price is a bit high, but it's a good deal in the long run." The input data is the analysis result, and the output data is the generated response.

[0741] Step 8:

[0742] The terminal displays the generated response on the user's (store clerk's) screen. This allows the clerk to immediately confirm and respond to the customer appropriately. The input data is the generated response, and the output data is the displayed text information.

[0743] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0744] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0745] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0746] [Third Embodiment]

[0747] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0748] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0749] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0750] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0751] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0752] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0753] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0754] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0755] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0756] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0757] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0758] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0759] This invention relates to a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. This embodiment details how the system achieves these functions.

[0760] The server first receives a rebate topic entered by the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and its internal database. The collected information is summarized and stored in the server's database. Subsequently, the terminal (tablet or robot) displays this summarized information and provides it to the user (student).

[0761] When the rebate program begins, the server records each user's statements as text data in real time using speech recognition technology. The recorded statements are analyzed by a keyword extraction algorithm, and the server generates an appropriate response. For example, if a user (student) says, "Renewable energy is expensive," the server analyzes this statement and, based on the keyword "cost," generates a response such as, "While the initial investment in solar power generation is high, it has cost-saving effects in the long term."

[0762] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate. This summary is compiled into meeting minutes and displayed on the terminal. The meeting minutes can be reviewed and edited by the user (teacher).

[0763] Next, the server generates a presentation template based on the meeting minutes data. This template is provided in a format that includes charts and graphs and is displayed on the device (tablet or robot). The user (student) reviews the content of the material together with the rebate robot using this template and makes improvements.

[0764] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive." In response to this statement, the server generates an appropriate response, and the terminal (rebate robot) replies, "While the initial investment for solar power generation is high, it has cost-saving effects in the long term." Subsequently, the server analyzes the entire rebate, generates a summarized meeting minutes, and creates a presentation template. Based on this material, the user (student) reviews the final slides with the robot and makes revisions.

[0765] As described above, the present invention is a system that provides consistent support for rebates, from preparation and implementation to summarization and presentation preparation, thereby improving the quality of rebates and enabling efficient class management.

[0766] The following describes the processing flow.

[0767] Step 1:

[0768] The server receives a rebate topic from the user (teacher). The user (teacher) enters the rebate topic "The pros and cons of promoting renewable energy" into the server.

[0769] Step 2:

[0770] The server collects relevant information based on the received rebate topic. This collection involves accessing APIs to retrieve data from the internet and accessing internal databases. However, the collected raw information is not usable as is, so natural language processing (NLP) algorithms are used to summarize the key points and convert it into an easily understandable format.

[0771] Step 3:

[0772] The device (tablet or rebate robot) prepares to provide summarized information to the user (student). At this point, the information is displayed on the tablet or robot's screen.

[0773] Step 4:

[0774] The user (student) begins the rebate. The server uses speech recognition technology to convert each student's statements into text data in real time and records them. During this process, a time stamp of the spoken content is also saved.

[0775] Step 5:

[0776] The server analyzes the user's statements. Using a keyword extraction algorithm, it identifies important keywords from the statements. For example, from the statement "Renewable energy is expensive," it extracts "cost" as an important keyword.

[0777] Step 6:

[0778] The terminal (rebate robot) generates appropriate responses based on analysis information sent from the server. It provides opinions from various perspectives and deepens the discussion. For example, it might generate a response such as, "The initial investment for solar power generation is high, but it has cost-saving effects in the long term."

[0779] Step 7:

[0780] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. Using natural language processing algorithms, it extracts the main points of the discussion and summarizes the key takeaways.

[0781] Step 8:

[0782] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make corrections as needed.

[0783] Step 9:

[0784] The server generates a presentation template based on the meeting minutes. This template includes key points, data, charts, and graphs discussed during the discussion.

[0785] Step 10:

[0786] The terminal displays the generated document template to the user (student). The user (student) reviews the document content together with the rebate robot and makes improvements. For example, they can add data or charts and change the format to make it easier to understand.

[0787] Step 11:

[0788] The user (student) reviews the completed materials and saves them as the final presentation materials. The device provides the completed materials for the presentation day.

[0789] The above is a detailed explanation of the processing steps of the rebate robot system. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates.

[0790] (Example 1)

[0791] Next, we will describe Example 1. 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."

[0792] Conventional rebate support systems do not consistently handle the entire process, from collecting and summarizing information related to rebate topics, to recording and analyzing statements, generating appropriate responses, creating meeting minutes, and preparing presentation materials. As a result, the process from rebate preparation to implementation and summarization is inefficient, leading to a decline in the quality of rebates.

[0793] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0794] In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the generated presentation materials with the user, means for collecting relevant information from the internet or an internal database based on the rebate topic entered by the user, means for summarizing the collected information and storing it in a database, means for using natural language processing technology in the collection and summarization process, means for accessing external data sources via an internet connection, means for recording each user's statements as text data in real time using speech recognition technology, means for analyzing the recorded statements with a keyword extraction algorithm, means for using a generative AI model to generate responses based on keywords, and means for inputting prompt sentences into the generative AI model to generate appropriate responses. This makes it possible to consistently and efficiently carry out everything from rebate preparation and implementation to summarization and presentation preparation.

[0795] A "rebate topic" is the subject or theme discussed during a rebate (debate).

[0796] "Users" refer to the people who use the system, specifically teachers and students.

[0797] "Means of information gathering" refers to methods and technologies for obtaining information related to rebate topics from the internet or internal databases.

[0798] "Means of summarization" refer to methods and techniques for concisely organizing collected information.

[0799] "Means of display" refers to methods and techniques for presenting collected and summarized information to the user.

[0800] "Means of recognition and recording" refers to methods and technologies for real-time speech recognition of a user's speech and saving it as text data.

[0801] "Means of analysis" refer to methods and techniques for analyzing recorded text data and extracting meaning and keywords.

[0802] "Means for generating appropriate responses" refer to methods and technologies for responding to users based on analysis results.

[0803] "Means of analyzing audio recordings" refers to methods and techniques for analyzing audio recordings of rebates and extracting important statements and key points.

[0804] "Means of displaying meeting minutes" refers to methods and technologies for displaying summarized meeting minutes to users.

[0805] "Means for generating presentation materials" refers to methods and techniques for creating presentation materials based on meeting minutes data.

[0806] "Methods for refinement" refer to methods and techniques for reviewing generated presentation materials together with the user and making corrections and improvements as needed.

[0807] An "external data source" refers to an information source located outside the system, such as publicly available information on the internet or external databases.

[0808] "Natural language processing technology" refers to artificial intelligence technology used to analyze and summarize collected information and user statements.

[0809] "Speech recognition technology" refers to the technology used to convert speech data into text data.

[0810] A "keyword extraction algorithm" is a computational method or technique for selecting important words and phrases from text data.

[0811] A "generative AI model" is an artificial intelligence model designed to generate appropriate responses based on user statements.

[0812] A "prompt statement" is a sentence input to a generative AI model that provides instructions or context for generating a specific response.

[0813] The present invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. Specific embodiments of the present invention are described below.

[0814] 1. Basic System Configuration

[0815] This system consists of a server, terminals (tablets and robots), and users (teachers and students). The server plays a central role in information collection and analysis, while the terminals provide information to users and accept user input.

[0816] Hardware and software to be used

[0817] Server: A server computer with a high-performance processor and large memory capacity is recommended. The server accesses external data sources via an internet connection and manages the internal database.

[0818] Internet connection: A high-speed internet connection is required to facilitate data collection and access to external data sources.

[0819] Speech recognition technology: The Google Speech-to-Text API is used to convert user speech into text data in real time.

[0820] Natural Language Processing (NLP) technologies: Google Cloud's Natural Language API and Python libraries (e.g., BeautifulSoup, spaCy) are used. This is used to summarize relevant information and analyze user utterances.

[0821] Keyword extraction algorithm: TF-IDF and the spaCy library are used to extract important keywords from user utterances.

[0822] Generative AI Model: Generative AI models such as OpenAI's GPT-3 are used. This is to generate appropriate responses based on user statements.

[0823] Database: Relational databases such as MySQL and PostgreSQL are used to manage collected information, summary data, and meeting minutes.

[0824] 2. System Operation

[0825] The server first receives a rebate topic as input from the user (teacher). Based on this input, the server collects relevant information from the internet and its internal database. Google's search API and queries from the internal database are used for information collection.

[0826] The collected information is summarized using the Google Cloud Natural Language API. The summarized information is stored in the server's database.

[0827] Next, the device (tablet or robot) displays this summary information to the user (student). This allows the student to prepare for their rebate in advance.

[0828] Rebate start:

[0829] During the rebate process, the server records each user's speech in real time as text data using the Google Speech-to-Text API. This recorded data is then analyzed using TF-IDF and the spaCy library to extract important keywords.

[0830] Based on the analyzed data, the server inputs prompts into a generative AI model (e.g., OpenAI's GPT-3) to generate an appropriate response. For example, if a user says, "Renewable energy is expensive," the server inputs the prompt, "Generate a response regarding the cost of renewable energy," into the generative AI model, which then generates an appropriate response.

[0831] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. This summary is compiled into meeting minutes and displayed on the terminal. The user (teacher) reviews the minutes and makes corrections as needed.

[0832] Finally, the server generates a presentation template based on the meeting minutes. This template includes charts and graphs, and uses libraries such as Matplotlib and Pandas. This material is displayed on the terminal, and the user (student) reviews the content together with the rebate robot and makes final adjustments.

[0833] 3. Specific Examples

[0834] If the rebate topic is "The pros and cons of promoting renewable energy," the user (teacher) enters this topic into the server. The server uses the internet and its internal database to collect relevant information, summarizes data such as "data on the costs and efficiency of renewable energy implementation," and stores it in the database. The terminal displays this information to the students to help them prepare for their rebates.

[0835] During the rebate process, the user's (student's) statement, "Renewable energy is expensive," is recorded in real time as text data by the server, and the keyword "cost" is analyzed by a keyword extraction algorithm. The prompt "Generate a response about the cost of renewable energy" is input to the generative AI model (GPT-3), and the response "The initial investment for solar power generation is high, but it has cost-saving effects in the long term" is generated.

[0836] After the rebate process is complete, the server re-analyzes the audio data and creates meeting minutes. It then generates a presentation template, which the user reviews and revises together with the rebate robot.

[0837] Thus, this system makes it possible to efficiently handle the entire process of rebate preparation, from preparation and execution to summarization and presentation preparation.

[0838] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0839] Step 1:

[0840] The server receives a rebate topic as input from the user (teacher).

[0841] Input: Users enter rebate topics such as "The pros and cons of promoting renewable energy."

[0842] Data processing: The server saves this topic and begins searching for related information.

[0843] Output: Ready to explore

[0844] Step 2:

[0845] The server collects relevant information from the internet and internal databases based on the rebate topic.

[0846] Input: Rebate Topic

[0847] Data Processing: We collect the latest information from the internet using Google's search API and other tools, and also retrieve relevant information from our internal database.

[0848] Output: List of collected related information

[0849] Step 3:

[0850] The server summarizes the collected information and stores it in a database.

[0851] Input: List of collected information

[0852] Data processing: Text summarization is performed using the Google Cloud Natural Language API.

[0853] Output: Summarized information is saved to the database.

[0854] Step 4:

[0855] The device (tablet or robot) provides summary information to the user (student).

[0856] Input: Summary information stored in the database

[0857] Data processing: None

[0858] Output: Summary information will be displayed on the terminal screen.

[0859] Step 5:

[0860] The server recognizes and records each user's statements in real time after the rebate program begins.

[0861] Input: User's voice utterance

[0862] Data processing: Convert speech to text using the Google Speech-to-Text API.

[0863] Output: Real-time recorded text data

[0864] Step 6:

[0865] The server analyzes the recorded statements and generates an appropriate response.

[0866] Input: Text data recorded in real time

[0867] Data processing: Extract keywords using TF-IDF or the spaCy library.

[0868] Output: Keyword extraction results

[0869] Step 7:

[0870] The server uses a generative AI model to generate an appropriate response.

[0871] Input: Keyword extraction results

[0872] Data processing: Input prompt text into a generative AI model (e.g., OpenAI GPT-3) and generate a response.

[0873] Output: Response generated by the generative AI model

[0874] Specific example: If a student says "Renewable energy is expensive" during a rebate session, and the prompt is "Generate a response about the cost of renewable energy," the response will be "The initial investment for solar power is high, but it has cost-saving effects in the long term."

[0875] Step 8:

[0876] After the rebate is completed, the server analyzes the recorded data, summarizes it, and creates meeting minutes.

[0877] Input: Rebate recording data

[0878] Data processing: Summarize the audio data using a text analysis algorithm.

[0879] Output: Summarized meeting minutes

[0880] Step 9:

[0881] The terminal displays summarized meeting minutes to the user (teacher) and assists with review and correction.

[0882] Input: Summarized meeting minutes

[0883] Data processing: None

[0884] Output: The meeting minutes will be displayed on the terminal screen.

[0885] Step 10:

[0886] The server generates presentation material templates based on the meeting minutes.

[0887] Input: Summarized meeting minutes

[0888] Data processing: Create charts and graphs using Matplotlib and Pandas.

[0889] Output: Presentation material template

[0890] Step 11:

[0891] The device provides the generated presentation materials to the user (student) and supports them in refining their work.

[0892] Input: Presentation material template

[0893] Data processing: Provides an interface to reflect necessary modifications.

[0894] Output: Finalized presentation materials

[0895] Through the above processing steps, this system can efficiently handle everything from preparing and executing rebates to summarizing and preparing presentations.

[0896] (Application Example 1)

[0897] Next, we will explain Application Example 1. In the following explanation, 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."

[0898] In modern factory settings, there is a need to quickly identify problems on the production line and propose improvements. However, there is a lack of means to identify problems in real time and provide appropriate information, which can lead to decreased production efficiency. In particular, there is a need for a system that provides quick and concrete solutions to production problems and troubles faced by operators.

[0899] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0900] In this invention, the server includes means for inputting rebate topics, means for collecting and summarizing information related to rebate topics, means for displaying information related to rebate topics, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on meeting minutes, means for refining presentation materials with the user, means for collecting problems and improvement suggestions at the production site in real time using speech recognition technology, means for extracting keywords from speech-recognized text and collecting related information, and means for summarizing and displaying the collected information. This enables the rapid and specific provision of information at the production site.

[0901] A "rebuttal topic" is a theme or issue that is discussed in a debate.

[0902] "Means of collecting and summarizing information" refers to methods and devices for obtaining relevant information from sources such as the internet or internal databases and compiling it into a concise format.

[0903] "Means of displaying information" refers to devices and software that visually provide collected and summarized information to users.

[0904] "Means for recognizing and recording user speech in real time" refers to a device or method that uses speech recognition technology to instantly save user speech as text data.

[0905] "Means for analyzing user utterances and generating appropriate responses" refers to methods or devices that extract keywords from speech-recognized text, analyze their content, and automatically create appropriate responses.

[0906] "Means for analyzing and summarizing debate recordings" refers to devices or methods for analyzing debate session recordings and concisely summarizing the main points of the discussion.

[0907] "Means of displaying meeting minutes" refers to devices or software that visually provide users with a summarized version of the discussion.

[0908] "Means for generating presentation materials" refers to devices or methods that automatically create presentation-appropriate forms and materials based on meeting minutes data.

[0909] "Means of refinement" refer to devices or methods that allow users to modify or improve the generated presentation materials.

[0910] "Methods for collecting problems and improvement suggestions in the production field in real time using voice recognition technology" refers to devices and technologies that recognize voice communication on the production line and collect that information immediately.

[0911] "Means for extracting keywords and collecting related information" refers to methods or devices that extract important words from speech-recognized text and search for and obtain related information based on those words.

[0912] "Means for summarizing and displaying collected information" refers to devices or software that condense acquired information into a concise format and provide it to the user visually.

[0913] This invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user comments in real time. Applying this to factory robots, the following system is realized to collect problems and improvement suggestions in real time on the production floor.

[0914] 1. System Program

[0915] The system consists of the following main processing steps:

[0916] 1. Input a rebate topic (problems or improvement suggestions in the production environment), and collect, summarize, and display related information.

[0917] 2. The system performs speech recognition on the user's speech in real time and records the resulting text data.

[0918] 3. Extract keywords from the speech-recognized text, collect and analyze related information, and generate an appropriate response.

[0919] 4. Summarize the collected information and present it to the user visually.

[0920] 2. Explain the program's processing in natural language.

[0921] Input, gather, summarize, and display information on rebate topics.

[0922] The server collects relevant information from the internet and internal databases based on the rebate topic entered by the user. This process utilizes major search engine APIs and internal database querying techniques. The collected information is summarized using a transformer model and stored in the database. Subsequently, the terminal (factory robot) displays this summarized information and provides it to the on-site operator.

[0923] Speech recognition and text conversion of user speech.

[0924] The user's (operator's) speech is collected through the robot's microphone and converted into text data using the speech_recognition library. By utilizing Google's speech recognition service, highly accurate speech-to-text conversion is achieved.

[0925] Keyword extraction, collection and analysis of related information

[0926] Key keywords are extracted from the speech-recognized text. This is done using natural language processing techniques and simple text segmentation algorithms. Based on the extracted keywords, relevant information is collected from the Google Search API and other databases. The collected information is then summarized again using a transformer model to generate an appropriate response.

[0927] Display summary information

[0928] Ultimately, the server generates summarized information and provides it to the user through the terminal's display. This process allows the user to obtain specific and immediate solutions.

[0929] 3. Add specific examples

[0930] As a concrete example, let's consider the problem of frequent defective products on a production line. When an operator asks the robot, "The defect rate of our products has been increasing recently. What could be the cause?", the robot analyzes the statement and gathers relevant information. Finally, it responds, "It may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening the inspection process."

[0931] Example of a prompt

[0932] "We've encountered a problem where the number of defective products has suddenly increased during the manufacturing process. Please gather information related to this issue and propose appropriate solutions."

[0933] This will enable the rapid and specific provision of information at the production site.

[0934] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0935] Step 1:

[0936] The server receives a rebate topic as input from the user. This rebate topic serves as the basis for collecting relevant information from the internet and internal databases. For example, if a user enters "causes of defective products" on a production line, information collection will begin based on this topic.

[0937] Step 2:

[0938] The server collects relevant information based on rebate topics. This information collection utilizes the Google Search API and internal database querying techniques. The collected data is summarized using natural language processing techniques. Specifically, key points are extracted using a transformer model.

[0939] Step 3:

[0940] The server stores the summarized information in a database and then sends it to the terminal (factory robot). This summarized information is displayed on the terminal's screen, allowing the user (operator) to visually confirm it.

[0941] Step 4:

[0942] The user speaks about problems and suggestions for improvement. This speech is collected by the robot's microphone. For example, they might ask, "The product defect rate has been increasing recently. What is the reason?"

[0943] Step 5:

[0944] The server converts the voice data transmitted from the robot into text data using speech recognition technology (e.g., the speech_recognition library). This conversion process uses Google's speech recognition service to achieve high accuracy in text transcription.

[0945] Step 6:

[0946] The server extracts important keywords from the speech-recognized text. This keyword extraction uses natural language processing techniques and simple text segmentation algorithms. For example, words like "defect rate" and "cause" might be extracted.

[0947] Step 7:

[0948] The server then collects information again based on the extracted keywords. This information gathering again uses the Google Search API and internal database queries. The collected information is similarly summarized using a transformer model.

[0949] Step 8:

[0950] The server analyzes the summarized information and generates an appropriate response. The generated response is sent to the terminal in text format. For example, it might offer specific improvement suggestions such as, "This may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening your inspection process."

[0951] Step 9:

[0952] The terminal displays the generated response on its screen for the user to review. Based on the displayed information, the user considers and implements specific countermeasures.

[0953] Step 10:

[0954] The server analyzes the entire rebate process from the audio recording and generates a summary as meeting minutes. This summary is then saved back to the database and can be viewed and modified by the user on their terminal.

[0955] Step 11:

[0956] The server generates presentation materials based on the meeting minutes. The generated materials are provided in a format that includes charts and graphs, and users can make final reviews and revisions.

[0957] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0958] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement with an emotion engine that recognizes the user's emotions in real time. This makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[0959] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects and summarizes relevant information from the internet and internal databases. The collected information is summarized and stored in the database. Subsequently, the device (tablet or robot) prepares to provide the summarized information to the user (student).

[0960] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. The user's speech is then analyzed using a keyword extraction algorithm to identify important keywords.

[0961] In addition, the emotion engine, along with the user's statements,

[0962] The system also recognizes and records user emotions in real time. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis, and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[0963] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, if a user says, "Renewable energy is expensive," and the emotion engine detects "anxiety," the robot will provide a response such as, "While the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[0964] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[0965] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final presentation material.

[0966] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers long-term cost savings and is therefore reassuring," which the terminal (rebate robot) responds to. Subsequently, the entire rebate is analyzed, and a summarized meeting transcript and presentation materials are generated. Finally, the user (student) and the rebate robot refine the materials to complete them.

[0967] The above is a specific embodiment of a rebate robot system that incorporates an emotion engine. This system allows for efficient preparation, execution, summarization, and presentation preparation of rebates, and because it also takes the user's emotional state into consideration, it can achieve a deeper understanding and empathy.

[0968] The following describes the processing flow.

[0969] Step 1:

[0970] The server receives the rebate topic entered by the user (teacher). For example, the user (teacher) enters the topic "The pros and cons of promoting renewable energy."

[0971] Step 2:

[0972] The server collects relevant information from the internet and internal databases based on the received rebate topic. This process utilizes APIs and web scraping techniques. The collected information is summarized using natural language processing (NLP) algorithms.

[0973] Step 3:

[0974] The device (tablet or rebate robot) displays summarized information to the user (student), allowing the student to prepare for the rebate.

[0975] Step 4:

[0976] At the start of the rebate, the server uses speech recognition technology in real time to convert the user's (student's) statements into text data and begins recording.

[0977] Step 5:

[0978] The emotion engine analyzes the user's (student's) speech, recognizing and recording their emotional state (e.g., joy, anxiety, anger) in real time. The analysis uses data such as voice tone, facial expression analysis, and heart rate.

[0979] Step 6:

[0980] The server analyzes the text and sentiment data of the statements. It uses a keyword extraction algorithm to identify important keywords from the statements.

[0981] Step 7:

[0982] The terminal (rebate robot) generates an appropriate response based on the analysis results sent from the server. For example, in response to the statement "Renewable energy is expensive" and the emotional state "anxiety," it generates a response such as "Although the initial investment in solar power generation is high, it has cost-saving effects in the long term, so you can rest assured."

[0983] Step 8:

[0984] After the rebate session ends, the server analyzes the entire audio recording and summarizes the key points and the user's emotional state. The summary is then compiled into meeting minutes.

[0985] Step 9:

[0986] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make any necessary corrections.

[0987] Step 10:

[0988] The server generates a presentation template based on the reviewed and revised meeting minutes. This template includes key points of the discussion, data, charts, graphs, and information on sentiment.

[0989] Step 11:

[0990] The terminal displays a generated document template to the user (student). The user (student) then collaborates with the rebate robot to refine the document's content. For example, they might input additional data or change the format of graphs.

[0991] Step 12:

[0992] The user (student) reviews and saves the final presentation materials. The device then provides these final materials for presentation.

[0993] The above outlines the specific processing steps of the rebate robot system that incorporates an emotion engine. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates while taking into account the user's emotional state.

[0994] (Example 2)

[0995] Next, we will describe Example 2. 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."

[0996] In rebate processes, it is difficult to analyze users' statements and emotional states in real time and provide appropriate responses and feedback based on that analysis. Furthermore, tasks such as summarizing the content after the rebate, generating meeting minutes, creating presentation materials, and refining them are time-consuming and labor-intensive, making them difficult to perform efficiently.

[0997] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying the summarized information, means for recognizing user statements in real time and converting them into text data, means for analyzing user statements and extracting keywords, means for recognizing and recording emotional states, means for generating appropriate responses based on user statements and emotional data, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation material templates based on meeting minutes, and means for refining presentation material templates. This makes it possible to provide efficient and appropriate responses while considering the user's statements and emotional states during the rebate process, and to efficiently carry out post-rebate work.

[0998] A "rebuttal topic" refers to the subject or theme of a rebuttal, and the content that will be discussed in that rebuttal.

[0999] "Means for collecting and summarizing information" refers to a system or method for searching for relevant information from the internet or internal databases and summarizing that information concisely using natural language processing technology.

[1000] "Means of displaying information" refers to an interface that visually provides users with access to collected and summarized information.

[1001] "Speech recognition means" refers to a system or method that receives speech input and converts it into text data.

[1002] A "keyword extraction method" is a system or method that automatically identifies and extracts important words and phrases from text data.

[1003] "Emotion recognition means" refers to a system or method that identifies and records a user's emotional state from multiple information sources such as voice, facial expressions, and biometric data.

[1004] A "response generation means" is a system or method that automatically generates an appropriate response or feedback based on the user's statements and emotional state.

[1005] "Means for analyzing and summarizing recorded data" refers to a system or method for analyzing audio data and concisely summarizing its contents.

[1006] "Means for displaying meeting minutes" refers to an interface for presenting summarized rebate details in a format that users can view.

[1007] "Means for generating presentation material templates" refers to a system or method that automatically creates a template containing the necessary points, data, graphs, and charts for a presentation, based on summarized meeting minutes.

[1008] "Methods for refining presentation material templates" refers to support systems or methods that allow users to edit generated templates and complete them as final presentation materials.

[1009] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement, with an emotion engine that recognizes the user's emotions in real time. This system makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[1010] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and internal databases and summarizes it using natural language processing techniques. This technique uses common natural language processing models, such as BERT or GPT-4. The collected information is summarized and stored in the database. The terminal (tablet or robot) then prepares to provide the summarized information to the user (student).

[1011] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. For speech recognition, a service such as Google Speech-to-Text is used. The user's speech is analyzed, and key keywords are identified using keyword extraction algorithms (e.g., TF-IDF or word cloud).

[1012] In addition, the emotion engine recognizes and records the user's emotional state in real time along with their statements. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis (e.g., Azure Face API), and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[1013] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, a response generated by the server might include something like, "Although the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[1014] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[1015] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final version.

[1016] To give a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers cost-saving benefits in the long term, so it's reassuring," and the terminal (rebate robot) responds. Subsequently, the entire rebate is analyzed, a summarized meeting minute and presentation materials are generated, and finally, the user (student) and the rebate robot refine the materials to complete them.

[1017] Examples of prompts for a generative AI model include:

[1018] "Regarding the rebate topic 'The pros and cons of widespread renewable energy,' generate an appropriate response using student statements and sentiment data. A student stated, 'Renewable energy is expensive,' and the sentiment engine detected 'anxiety.' Provide the optimal response."

[1019] It can be used as such.

[1020] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1021] Step 1: Enter the rebate topic.

[1022] Input: The user (teacher) enters the rebate topic using a terminal.

[1023] Specific operation: The user (teacher) enters a topic such as "the pros and cons of promoting renewable energy" into the terminal interface. This information is sent from the terminal to the server.

[1024] Output: The input topic information is received by the server.

[1025] Step 2: Gather and summarize relevant information

[1026] Input: Rebate topic information received by the server.

[1027] Specific operation: The server searches for relevant information from the internet and internal databases, and summarizes the information using natural language processing techniques (e.g., BERT or GPT-4).

[1028] Data processing: Analyze data from the sources used, extract necessary information, and summarize it.

[1029] Output: The summarized information is saved to the database.

[1030] Step 3: Prepare to provide summary information

[1031] Input: Summarized information.

[1032] Specific operation: The terminal receives summary information from the server and prepares it to display the information in a format that is easy for the user (student) to use.

[1033] Output: Summary information in a format viewable by the user (student).

[1034] Step 4: Initiating the rebate and recognizing the statement

[1035] Input: The rebate process begins, and the user's (student's) comments are collected via the microphone.

[1036] Specific operation: The server uses speech recognition technology (e.g., Google Speech-to-Text) to convert and record speech data into text data in real time.

[1037] Data processing: Real-time text conversion of audio data.

[1038] Output: Text data of the user's (student's) statements.

[1039] Step 5: Recognizing the user's emotions

[1040] Input: User (student) speech data and biometric data (voice tone, facial expressions, heart rate).

[1041] Specific operation: The emotion engine analyzes this data to determine the user's (student's) emotional state. Sentiment Analysis technology is used for voice tone analysis, facial expression analysis APIs for facial expression analysis, and biosensors for biometric data analysis.

[1042] Output: User (student) emotion recognition data.

[1043] Step 6: Generating the response

[1044] Input: Transcribed speech data and sentiment recognition data.

[1045] Specific operation: The terminal (rebate robot) uses an NLP model to generate an appropriate response based on statements and sentiment data from the server.

[1046] Data processing: A response generation process that takes emotional states into consideration.

[1047] Output: The generated response.

[1048] Step 7: Analysis and summary after the rebate is completed.

[1049] Input: Voice and emotion data collected during rebates.

[1050] Specific operation: The server analyzes the recorded data and summarizes the rebate details and emotional state. The recorded data is then subjected to speech recognition again, converted to text, and analyzed.

[1051] Data processing: The process of summarizing from multiple data sources.

[1052] Output: Summarized meeting minutes data.

[1053] Step 8: View and review the meeting minutes.

[1054] Input: Summarized meeting minutes data.

[1055] Specific operation: The terminal displays the meeting minutes, and the user (teacher) reviews the content and makes corrections as needed.

[1056] Output: Reviewed and revised meeting minutes.

[1057] Step 9: Generating presentation materials

[1058] Input: Reviewed and revised meeting minutes.

[1059] Specific operation: The server automatically generates a presentation template based on the meeting minutes, including key points, data, graphs, and charts.

[1060] Data processing: The process of visualizing data and creating templates.

[1061] Output: Generated presentation template.

[1062] Step 10: Refining the presentation materials

[1063] Input: Presentation material template.

[1064] Specific operation: The user (student) edits materials together with the rebate robot and refines them into final presentation materials.

[1065] Output: Final presentation materials.

[1066] (Application Example 2)

[1067] Next, we will explain application example 2. In the following explanation, 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."

[1068] Traditional rebate systems and in-store customer service systems have difficulty recognizing and analyzing user statements in real time and generating responses that take into account the user's emotional state. This has resulted in challenges in providing services that accurately capture user needs and emotions, hindering improvements in customer satisfaction and efficient business negotiations.

[1069] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the presentation materials with the user, means for analyzing conversations with customers in real time during business negotiations and converting customer statements into text data, means for determining the content of customer statements and emotional state and generating appropriate responses based thereon, and means for displaying the generated responses on the store clerk's terminal. This makes it possible to analyze user statements and emotions in real time and provide accurate responses in rebate and in-store customer service.

[1070] A "rebate topic" refers to the subject matter that becomes the theme of a discussion or debate.

[1071] "Real-time recognition" refers to the immediate recording and analysis of speech and actions.

[1072] "Text data" refers to a data format in which audio is converted into written information.

[1073] "Emotional state" refers to the user's emotional and psychological state.

[1074] "Appropriate response" refers to providing the most suitable answer based on the user's statements and emotions.

[1075] "Meeting minutes" refers to a record that summarizes the content of a meeting or discussion.

[1076] "Presentation materials" refer to explanatory documents created for purposes such as rebates or presentations.

[1077] "Brush-up" refers to the process of improving existing materials and information to create a higher-quality version.

[1078] "Business negotiation" refers to dialogue and negotiation with customers in a physical store.

[1079] "Device" refers to a device such as a smartphone, tablet, or smart glasses.

[1080] Modes for carrying out the invention

[1081] This invention is a system that recognizes the customer's statements and emotional state in real time during business negotiations with customers who visit the store, and generates appropriate responses. First, the server has a means of inputting a rebate topic and collects and summarizes information related to the topic from the internet or an internal database. The server sends this summarized information to a terminal, and the terminal displays the information.

[1082] Next, the device converts the customer's speech into text data in real time using speech recognition technology. The Python library `speech_recognition` is used for this speech recognition. Furthermore, the device analyzes the user's speech and uses an emotion engine to determine the customer's emotions. This emotion engine analyzes the user's emotional state based on voice tone analysis, facial expression analysis, and biometric data such as heart rate.

[1083] Based on the analyzed utterances and emotional state, the device generates an appropriate response. A generative AI model is used for response generation, ensuring the user receives the most suitable reply. These processes are seamlessly coordinated between the server and the device, utilizing devices such as smartphones, tablets, or smart glasses.

[1084] For example, if a customer visiting an electronics store asks, "Isn't this TV too expensive?", the terminal uses speech recognition technology to convert this statement into text data, and an emotion engine detects "anxiety" from the customer's tone of voice. Subsequently, a generative AI model determines that the customer's statement, "I feel the price of the new TV is too high," is anxiety-inducing and generates an appropriate response: "The initial price is on the higher side, but in the long run, you can save on electricity costs, making it a good deal overall." This response is immediately displayed on the terminal, allowing store employees to refer to it while assisting the customer.

[1085] A concrete example of a prompt message is: "I feel that the price of the new TV is too high." This prompt should be interpreted as a sign of concern, and an appropriate response should be generated to reassure the customer. Using such concrete examples enables the effective implementation of this system.

[1086] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1087] Step 1:

[1088] The server receives rebate topics from users (store clerks). The rebate topics entered by the user are sent to the server. The input data includes the topic name and a detailed description.

[1089] Step 2:

[1090] The server collects and summarizes relevant information from the internet and internal databases based on the received rebate topic. The collected data is summarized in text format, with key points extracted by a summarization algorithm. The summarized information is stored on the server. Input data is the raw data of the relevant information, while output data is the summarized text data.

[1091] Step 3:

[1092] The server sends summarized information to a device (smartphone, tablet, smart glasses, etc.). The device displays the received summarized information on its screen. The input data is summarized text data, and the output data is data in a format that can be displayed on the screen.

[1093] Step 4:

[1094] The device uses its microphone to capture what the user (customer) says aloud. The input data is raw audio data.

[1095] Step 5:

[1096] The device converts the acquired audio data into text data using the speech_recognition library. This speech recognition process captures the customer's spoken content in text format. The input data is audio data, and the output data is text data.

[1097] Step 6:

[1098] The device uses a generative AI model and an emotion engine to analyze acquired text data and determine the customer's emotional state. This analysis utilizes data such as voice tone, facial expression analysis, and heart rate. Input data consists of text data and biometric data, while output data is the analysis result (spoken content and emotional state).

[1099] Step 7:

[1100] The device generates an appropriate response using a generative AI model based on the analyzed utterance and emotional state. For example, if the utterance includes the keyword "too expensive" and the emotion of anxiety is detected, the response generation algorithm will generate an appropriate response such as "The initial price is a bit high, but it's a good deal in the long run." The input data is the analysis result, and the output data is the generated response.

[1101] Step 8:

[1102] The terminal displays the generated response on the user's (store clerk's) screen. This allows the clerk to immediately confirm and respond to the customer appropriately. The input data is the generated response, and the output data is the displayed text information.

[1103] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1104] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1105] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1106] [Fourth Embodiment]

[1107] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1108] As shown in Figure 7, the 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.

[1109] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1110] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1111] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1112] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1113] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1114] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1115] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1116] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1117] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1118] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1119] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1120] This invention relates to a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. This embodiment details how the system achieves these functions.

[1121] The server first receives a rebate topic entered by the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and its internal database. The collected information is summarized and stored in the server's database. Subsequently, the terminal (tablet or robot) displays this summarized information and provides it to the user (student).

[1122] When the rebate program begins, the server records each user's statements as text data in real time using speech recognition technology. The recorded statements are analyzed by a keyword extraction algorithm, and the server generates an appropriate response. For example, if a user (student) says, "Renewable energy is expensive," the server analyzes this statement and, based on the keyword "cost," generates a response such as, "While the initial investment in solar power generation is high, it has cost-saving effects in the long term."

[1123] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate. This summary is compiled into meeting minutes and displayed on the terminal. The meeting minutes can be reviewed and edited by the user (teacher).

[1124] Next, the server generates a presentation template based on the meeting minutes data. This template is provided in a format that includes charts and graphs and is displayed on the device (tablet or robot). The user (student) reviews the content of the material together with the rebate robot using this template and makes improvements.

[1125] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive." In response to this statement, the server generates an appropriate response, and the terminal (rebate robot) replies, "While the initial investment for solar power generation is high, it has cost-saving effects in the long term." Subsequently, the server analyzes the entire rebate, generates a summarized meeting minutes, and creates a presentation template. Based on this material, the user (student) reviews the final slides with the robot and makes revisions.

[1126] As described above, the present invention is a system that provides consistent support for rebates, from preparation and implementation to summarization and presentation preparation, thereby improving the quality of rebates and enabling efficient class management.

[1127] The following describes the processing flow.

[1128] Step 1:

[1129] The server receives a rebate topic from the user (teacher). The user (teacher) enters the rebate topic "The pros and cons of promoting renewable energy" into the server.

[1130] Step 2:

[1131] The server collects relevant information based on the received rebate topic. This collection involves accessing APIs to retrieve data from the internet and accessing internal databases. However, the collected raw information is not usable as is, so natural language processing (NLP) algorithms are used to summarize the key points and convert it into an easily understandable format.

[1132] Step 3:

[1133] The device (tablet or rebate robot) prepares to provide summarized information to the user (student). At this point, the information is displayed on the tablet or robot's screen.

[1134] Step 4:

[1135] The user (student) begins the rebate. The server uses speech recognition technology to convert each student's statements into text data in real time and records them. During this process, a time stamp of the spoken content is also saved.

[1136] Step 5:

[1137] The server analyzes the user's statements. Using a keyword extraction algorithm, it identifies important keywords from the statements. For example, from the statement "Renewable energy is expensive," it extracts "cost" as an important keyword.

[1138] Step 6:

[1139] The terminal (rebate robot) generates appropriate responses based on analysis information sent from the server. It provides opinions from various perspectives and deepens the discussion. For example, it might generate a response such as, "The initial investment for solar power generation is high, but it has cost-saving effects in the long term."

[1140] Step 7:

[1141] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. Using natural language processing algorithms, it extracts the main points of the discussion and summarizes the key takeaways.

[1142] Step 8:

[1143] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make corrections as needed.

[1144] Step 9:

[1145] The server generates a presentation template based on the meeting minutes. This template includes key points, data, charts, and graphs discussed during the discussion.

[1146] Step 10:

[1147] The terminal displays the generated document template to the user (student). The user (student) reviews the document content together with the rebate robot and makes improvements. For example, they can add data or charts and change the format to make it easier to understand.

[1148] Step 11:

[1149] The user (student) reviews the completed materials and saves them as the final presentation materials. The device provides the completed materials for the presentation day.

[1150] The above is a detailed explanation of the processing steps of the rebate robot system. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates.

[1151] (Example 1)

[1152] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1153] Conventional rebate support systems do not consistently handle the entire process, from collecting and summarizing information related to rebate topics, to recording and analyzing statements, generating appropriate responses, creating meeting minutes, and preparing presentation materials. As a result, the process from rebate preparation to implementation and summarization is inefficient, leading to a decline in the quality of rebates.

[1154] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1155] In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the generated presentation materials with the user, means for collecting relevant information from the internet or an internal database based on the rebate topic entered by the user, means for summarizing the collected information and storing it in a database, means for using natural language processing technology in the collection and summarization process, means for accessing external data sources via an internet connection, means for recording each user's statements as text data in real time using speech recognition technology, means for analyzing the recorded statements with a keyword extraction algorithm, means for using a generative AI model to generate responses based on keywords, and means for inputting prompt sentences into the generative AI model to generate appropriate responses. This makes it possible to consistently and efficiently carry out everything from rebate preparation and implementation to summarization and presentation preparation.

[1156] A "rebate topic" is the subject or theme discussed during a rebate (debate).

[1157] "Users" refer to the people who use the system, specifically teachers and students.

[1158] "Means of information gathering" refers to methods and technologies for obtaining information related to rebate topics from the internet or internal databases.

[1159] "Means of summarization" refer to methods and techniques for concisely organizing collected information.

[1160] "Means of display" refers to methods and techniques for presenting collected and summarized information to the user.

[1161] "Means of recognition and recording" refers to methods and technologies for real-time speech recognition of a user's speech and saving it as text data.

[1162] "Means of analysis" refer to methods and techniques for analyzing recorded text data and extracting meaning and keywords.

[1163] "Means for generating appropriate responses" refer to methods and technologies for responding to users based on analysis results.

[1164] "Means of analyzing audio recordings" refers to methods and techniques for analyzing audio recordings of rebates and extracting important statements and key points.

[1165] "Means of displaying meeting minutes" refers to methods and technologies for displaying summarized meeting minutes to users.

[1166] "Means for generating presentation materials" refers to methods and techniques for creating presentation materials based on meeting minutes data.

[1167] "Methods for refinement" refer to methods and techniques for reviewing generated presentation materials together with the user and making corrections and improvements as needed.

[1168] An "external data source" refers to an information source located outside the system, such as publicly available information on the internet or external databases.

[1169] "Natural language processing technology" refers to artificial intelligence technology used to analyze and summarize collected information and user statements.

[1170] "Speech recognition technology" refers to the technology used to convert speech data into text data.

[1171] A "keyword extraction algorithm" is a computational method or technique for selecting important words and phrases from text data.

[1172] A "generative AI model" is an artificial intelligence model designed to generate appropriate responses based on user statements.

[1173] A "prompt statement" is a sentence input to a generative AI model that provides instructions or context for generating a specific response.

[1174] The present invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user statements during the rebate process in real time. Specific embodiments of the present invention are described below.

[1175] 1. Basic System Configuration

[1176] This system consists of a server, terminals (tablets and robots), and users (teachers and students). The server plays a central role in information collection and analysis, while the terminals provide information to users and accept user input.

[1177] Hardware and software to be used

[1178] Server: A server computer with a high-performance processor and large memory capacity is recommended. The server accesses external data sources via an internet connection and manages the internal database.

[1179] Internet connection: A high-speed internet connection is required to facilitate data collection and access to external data sources.

[1180] Speech recognition technology: The Google Speech-to-Text API is used to convert user speech into text data in real time.

[1181] Natural Language Processing (NLP) technologies: Google Cloud's Natural Language API and Python libraries (e.g., BeautifulSoup, spaCy) are used. This is used to summarize relevant information and analyze user utterances.

[1182] Keyword extraction algorithm: TF-IDF and the spaCy library are used to extract important keywords from user utterances.

[1183] Generative AI Model: Generative AI models such as OpenAI's GPT-3 are used. This is to generate appropriate responses based on user statements.

[1184] Database: Relational databases such as MySQL and PostgreSQL are used to manage collected information, summary data, and meeting minutes.

[1185] 2. System Operation

[1186] The server first receives a rebate topic as input from the user (teacher). Based on this input, the server collects relevant information from the internet and its internal database. Google's search API and queries from the internal database are used for information collection.

[1187] The collected information is summarized using the Google Cloud Natural Language API. The summarized information is stored in the server's database.

[1188] Next, the device (tablet or robot) displays this summary information to the user (student). This allows the student to prepare for their rebate in advance.

[1189] Rebate start:

[1190] During the rebate process, the server records each user's speech in real time as text data using the Google Speech-to-Text API. This recorded data is then analyzed using TF-IDF and the spaCy library to extract important keywords.

[1191] Based on the analyzed data, the server inputs prompts into a generative AI model (e.g., OpenAI's GPT-3) to generate an appropriate response. For example, if a user says, "Renewable energy is expensive," the server inputs the prompt, "Generate a response regarding the cost of renewable energy," into the generative AI model, which then generates an appropriate response.

[1192] After the rebate session ends, the server analyzes the audio recording and summarizes the entire session. This summary is compiled into meeting minutes and displayed on the terminal. The user (teacher) reviews the minutes and makes corrections as needed.

[1193] Finally, the server generates a presentation template based on the meeting minutes. This template includes charts and graphs, and uses libraries such as Matplotlib and Pandas. This material is displayed on the terminal, and the user (student) reviews the content together with the rebate robot and makes final adjustments.

[1194] 3. Specific Examples

[1195] If the rebate topic is "The pros and cons of promoting renewable energy," the user (teacher) enters this topic into the server. The server uses the internet and its internal database to collect relevant information, summarizes data such as "data on the costs and efficiency of renewable energy implementation," and stores it in the database. The terminal displays this information to the students to help them prepare for their rebates.

[1196] During the rebate process, the user's (student's) statement, "Renewable energy is expensive," is recorded in real time as text data by the server, and the keyword "cost" is analyzed by a keyword extraction algorithm. The prompt "Generate a response about the cost of renewable energy" is input to the generative AI model (GPT-3), and the response "The initial investment for solar power generation is high, but it has cost-saving effects in the long term" is generated.

[1197] After the rebate process is complete, the server re-analyzes the audio data and creates meeting minutes. It then generates a presentation template, which the user reviews and revises together with the rebate robot.

[1198] Thus, this system makes it possible to efficiently handle the entire process of rebate preparation, from preparation and execution to summarization and presentation preparation.

[1199] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1200] Step 1:

[1201] The server receives a rebate topic as input from the user (teacher).

[1202] Input: Users enter rebate topics such as "The pros and cons of promoting renewable energy."

[1203] Data processing: The server saves this topic and begins searching for related information.

[1204] Output: Ready to explore

[1205] Step 2:

[1206] The server collects relevant information from the internet and internal databases based on the rebate topic.

[1207] Input: Rebate Topic

[1208] Data Processing: We collect the latest information from the internet using Google's search API and other tools, and also retrieve relevant information from our internal database.

[1209] Output: List of collected related information

[1210] Step 3:

[1211] The server summarizes the collected information and stores it in a database.

[1212] Input: List of collected information

[1213] Data processing: Text summarization is performed using the Google Cloud Natural Language API.

[1214] Output: Summarized information is saved to the database.

[1215] Step 4:

[1216] The device (tablet or robot) provides summary information to the user (student).

[1217] Input: Summary information stored in the database

[1218] Data processing: None

[1219] Output: Summary information will be displayed on the terminal screen.

[1220] Step 5:

[1221] The server recognizes and records each user's statements in real time after the rebate program begins.

[1222] Input: User's voice utterance

[1223] Data processing: Convert speech to text using the Google Speech-to-Text API.

[1224] Output: Real-time recorded text data

[1225] Step 6:

[1226] The server analyzes the recorded statements and generates an appropriate response.

[1227] Input: Text data recorded in real time

[1228] Data processing: Extract keywords using TF-IDF or the spaCy library.

[1229] Output: Keyword extraction results

[1230] Step 7:

[1231] The server uses a generative AI model to generate an appropriate response.

[1232] Input: Keyword extraction results

[1233] Data processing: Input prompt text into a generative AI model (e.g., OpenAI GPT-3) and generate a response.

[1234] Output: Response generated by the generative AI model

[1235] Specific example: If a student says "Renewable energy is expensive" during a rebate session, and the prompt is "Generate a response about the cost of renewable energy," the response will be "The initial investment for solar power is high, but it has cost-saving effects in the long term."

[1236] Step 8:

[1237] After the rebate is completed, the server analyzes the recorded data, summarizes it, and creates meeting minutes.

[1238] Input: Rebate recording data

[1239] Data processing: Summarize the audio data using a text analysis algorithm.

[1240] Output: Summarized meeting minutes

[1241] Step 9:

[1242] The terminal displays summarized meeting minutes to the user (teacher) and assists with review and correction.

[1243] Input: Summarized meeting minutes

[1244] Data processing: None

[1245] Output: The meeting minutes will be displayed on the terminal screen.

[1246] Step 10:

[1247] The server generates presentation material templates based on the meeting minutes.

[1248] Input: Summarized meeting minutes

[1249] Data processing: Create charts and graphs using Matplotlib and Pandas.

[1250] Output: Presentation material template

[1251] Step 11:

[1252] The device provides the generated presentation materials to the user (student) and supports them in refining their work.

[1253] Input: Presentation material template

[1254] Data processing: Provides an interface to reflect necessary modifications.

[1255] Output: Finalized presentation materials

[1256] Through the above processing steps, this system can efficiently handle everything from preparing and executing rebates to summarizing and preparing presentations.

[1257] (Application Example 1)

[1258] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1259] In modern factory settings, there is a need to quickly identify problems on the production line and propose improvements. However, there is a lack of means to identify problems in real time and provide appropriate information, which can lead to decreased production efficiency. In particular, there is a need for a system that provides quick and concrete solutions to production problems and troubles faced by operators.

[1260] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1261] In this invention, the server includes means for inputting rebate topics, means for collecting and summarizing information related to rebate topics, means for displaying information related to rebate topics, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on meeting minutes, means for refining presentation materials with the user, means for collecting problems and improvement suggestions at the production site in real time using speech recognition technology, means for extracting keywords from speech-recognized text and collecting related information, and means for summarizing and displaying the collected information. This enables the rapid and specific provision of information at the production site.

[1262] A "rebuttal topic" is a theme or issue that is discussed in a debate.

[1263] "Means of collecting and summarizing information" refers to methods and devices for obtaining relevant information from sources such as the internet or internal databases and compiling it into a concise format.

[1264] "Means of displaying information" refers to devices and software that visually provide collected and summarized information to users.

[1265] "Means for recognizing and recording user speech in real time" refers to a device or method that uses speech recognition technology to instantly save user speech as text data.

[1266] "Means for analyzing user utterances and generating appropriate responses" refers to methods or devices that extract keywords from speech-recognized text, analyze their content, and automatically create appropriate responses.

[1267] "Means for analyzing and summarizing debate recordings" refers to devices or methods for analyzing debate session recordings and concisely summarizing the main points of the discussion.

[1268] "Means of displaying meeting minutes" refers to devices or software that visually provide users with a summarized version of the discussion.

[1269] "Means for generating presentation materials" refers to devices or methods that automatically create presentation-appropriate forms and materials based on meeting minutes data.

[1270] "Means of refinement" refer to devices or methods that allow users to modify or improve the generated presentation materials.

[1271] "Methods for collecting problems and improvement suggestions in the production field in real time using voice recognition technology" refers to devices and technologies that recognize voice communication on the production line and collect that information immediately.

[1272] "Means for extracting keywords and collecting related information" refers to methods or devices that extract important words from speech-recognized text and search for and obtain related information based on those words.

[1273] "Means for summarizing and displaying collected information" refers to devices or software that condense acquired information into a concise format and provide it to the user visually.

[1274] This invention is a system that takes a rebate topic as input, collects and summarizes relevant information, and records and analyzes user comments in real time. Applying this to factory robots, the following system is realized to collect problems and improvement suggestions in real time on the production floor.

[1275] 1. System Program

[1276] The system consists of the following main processing steps:

[1277] 1. Input a rebate topic (problems or improvement suggestions in the production environment), and collect, summarize, and display related information.

[1278] 2. The system performs speech recognition on the user's speech in real time and records the resulting text data.

[1279] 3. Extract keywords from the speech-recognized text, collect and analyze related information, and generate an appropriate response.

[1280] 4. Summarize the collected information and present it to the user visually.

[1281] 2. Explain the program's processing in natural language.

[1282] Input, gather, summarize, and display information on rebate topics.

[1283] The server collects relevant information from the internet and internal databases based on the rebate topic entered by the user. This process utilizes major search engine APIs and internal database querying techniques. The collected information is summarized using a transformer model and stored in the database. Subsequently, the terminal (factory robot) displays this summarized information and provides it to the on-site operator.

[1284] Speech recognition and text conversion of user speech.

[1285] The user's (operator's) speech is collected through the robot's microphone and converted into text data using the speech_recognition library. By utilizing Google's speech recognition service, highly accurate speech-to-text conversion is achieved.

[1286] Keyword extraction, collection and analysis of related information

[1287] Key keywords are extracted from the speech-recognized text. This is done using natural language processing techniques and simple text segmentation algorithms. Based on the extracted keywords, relevant information is collected from the Google Search API and other databases. The collected information is then summarized again using a transformer model to generate an appropriate response.

[1288] Display summary information

[1289] Ultimately, the server generates summarized information and provides it to the user through the terminal's display. This process allows the user to obtain specific and immediate solutions.

[1290] 3. Add specific examples

[1291] As a concrete example, let's consider the problem of frequent defective products on a production line. When an operator asks the robot, "The defect rate of our products has been increasing recently. What could be the cause?", the robot analyzes the statement and gathers relevant information. Finally, it responds, "It may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening the inspection process."

[1292] Example of a prompt

[1293] "We've encountered a problem where the number of defective products has suddenly increased during the manufacturing process. Please gather information related to this issue and propose appropriate solutions."

[1294] This will enable the rapid and specific provision of information at the production site.

[1295] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1296] Step 1:

[1297] The server receives a rebate topic as input from the user. This rebate topic serves as the basis for collecting relevant information from the internet and internal databases. For example, if a user enters "causes of defective products" on a production line, information collection will begin based on this topic.

[1298] Step 2:

[1299] The server collects relevant information based on rebate topics. This information collection utilizes the Google Search API and internal database querying techniques. The collected data is summarized using natural language processing techniques. Specifically, key points are extracted using a transformer model.

[1300] Step 3:

[1301] The server stores the summarized information in a database and then sends it to the terminal (factory robot). This summarized information is displayed on the terminal's screen, allowing the user (operator) to visually confirm it.

[1302] Step 4:

[1303] The user speaks about problems and suggestions for improvement. This speech is collected by the robot's microphone. For example, they might ask, "The product defect rate has been increasing recently. What is the reason?"

[1304] Step 5:

[1305] The server converts the voice data transmitted from the robot into text data using speech recognition technology (e.g., the speech_recognition library). This conversion process uses Google's speech recognition service to achieve high accuracy in text transcription.

[1306] Step 6:

[1307] The server extracts important keywords from the speech-recognized text. This keyword extraction uses natural language processing techniques and simple text segmentation algorithms. For example, words like "defect rate" and "cause" might be extracted.

[1308] Step 7:

[1309] The server then collects information again based on the extracted keywords. This information gathering again uses the Google Search API and internal database queries. The collected information is similarly summarized using a transformer model.

[1310] Step 8:

[1311] The server analyzes the summarized information and generates an appropriate response. The generated response is sent to the terminal in text format. For example, it might offer specific improvement suggestions such as, "This may be due to a decline in the quality of the material supply. Please consider changing suppliers or strengthening your inspection process."

[1312] Step 9:

[1313] The terminal displays the generated response on its screen for the user to review. Based on the displayed information, the user considers and implements specific countermeasures.

[1314] Step 10:

[1315] The server analyzes the entire rebate process from the audio recording and generates a summary as meeting minutes. This summary is then saved back to the database and can be viewed and modified by the user on their terminal.

[1316] Step 11:

[1317] The server generates presentation materials based on the meeting minutes. The generated materials are provided in a format that includes charts and graphs, and users can make final reviews and revisions.

[1318] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1319] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement with an emotion engine that recognizes the user's emotions in real time. This makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[1320] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects and summarizes relevant information from the internet and internal databases. The collected information is summarized and stored in the database. Subsequently, the device (tablet or robot) prepares to provide the summarized information to the user (student).

[1321] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. The user's speech is then analyzed using a keyword extraction algorithm to identify important keywords.

[1322] In addition, the emotion engine, along with the user's statements,

[1323] The system also recognizes and records user emotions in real time. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis, and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[1324] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, if a user says, "Renewable energy is expensive," and the emotion engine detects "anxiety," the robot will provide a response such as, "While the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[1325] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[1326] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final presentation material.

[1327] As a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers long-term cost savings and is therefore reassuring," which the terminal (rebate robot) responds to. Subsequently, the entire rebate is analyzed, and a summarized meeting transcript and presentation materials are generated. Finally, the user (student) and the rebate robot refine the materials to complete them.

[1328] The above is a specific embodiment of a rebate robot system that incorporates an emotion engine. This system allows for efficient preparation, execution, summarization, and presentation preparation of rebates, and because it also takes the user's emotional state into consideration, it can achieve a deeper understanding and empathy.

[1329] The following describes the processing flow.

[1330] Step 1:

[1331] The server receives the rebate topic entered by the user (teacher). For example, the user (teacher) enters the topic "The pros and cons of promoting renewable energy."

[1332] Step 2:

[1333] The server collects relevant information from the internet and internal databases based on the received rebate topic. This process utilizes APIs and web scraping techniques. The collected information is summarized using natural language processing (NLP) algorithms.

[1334] Step 3:

[1335] The device (tablet or rebate robot) displays summarized information to the user (student), allowing the student to prepare for the rebate.

[1336] Step 4:

[1337] At the start of the rebate, the server uses speech recognition technology in real time to convert the user's (student's) statements into text data and begins recording.

[1338] Step 5:

[1339] The emotion engine analyzes the user's (student's) speech, recognizing and recording their emotional state (e.g., joy, anxiety, anger) in real time. The analysis uses data such as voice tone, facial expression analysis, and heart rate.

[1340] Step 6:

[1341] The server analyzes the text and sentiment data of the statements. It uses a keyword extraction algorithm to identify important keywords from the statements.

[1342] Step 7:

[1343] The terminal (rebate robot) generates an appropriate response based on the analysis results sent from the server. For example, in response to the statement "Renewable energy is expensive" and the emotional state "anxiety," it generates a response such as "Although the initial investment in solar power generation is high, it has cost-saving effects in the long term, so you can rest assured."

[1344] Step 8:

[1345] After the rebate session ends, the server analyzes the entire audio recording and summarizes the key points and the user's emotional state. The summary is then compiled into meeting minutes.

[1346] Step 9:

[1347] The terminal displays summarized meeting minutes to the user (teacher). The user (teacher) can review the minutes and make any necessary corrections.

[1348] Step 10:

[1349] The server generates a presentation template based on the reviewed and revised meeting minutes. This template includes key points of the discussion, data, charts, graphs, and information on sentiment.

[1350] Step 11:

[1351] The terminal displays a generated document template to the user (student). The user (student) then collaborates with the rebate robot to refine the document's content. For example, they might input additional data or change the format of graphs.

[1352] Step 12:

[1353] The user (student) reviews and saves the final presentation materials. The device then provides these final materials for presentation.

[1354] The above outlines the specific processing steps of the rebate robot system that incorporates an emotion engine. This system enables efficient preparation, execution, summarization, and presentation preparation of rebates while taking into account the user's emotional state.

[1355] (Example 2)

[1356] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1357] In rebate processes, it is difficult to analyze users' statements and emotional states in real time and provide appropriate responses and feedback based on that analysis. Furthermore, tasks such as summarizing the content after the rebate, generating meeting minutes, creating presentation materials, and refining them are time-consuming and labor-intensive, making them difficult to perform efficiently.

[1358] In Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying the summarized information, means for recognizing user statements in real time and converting them into text data, means for analyzing user statements and extracting keywords, means for recognizing and recording emotional states, means for generating appropriate responses based on user statements and emotional data, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation material templates based on meeting minutes, and means for refining presentation material templates. This makes it possible to provide efficient and appropriate responses while considering the user's statements and emotional states during the rebate process, and to efficiently carry out post-rebate work.

[1359] A "rebuttal topic" refers to the subject or theme of a rebuttal, and the content that will be discussed in that rebuttal.

[1360] "Means for collecting and summarizing information" refers to a system or method for searching for relevant information from the internet or internal databases and summarizing that information concisely using natural language processing technology.

[1361] "Means of displaying information" refers to an interface that visually provides users with access to collected and summarized information.

[1362] "Speech recognition means" refers to a system or method that receives speech input and converts it into text data.

[1363] A "keyword extraction method" is a system or method that automatically identifies and extracts important words and phrases from text data.

[1364] "Emotion recognition means" refers to a system or method that identifies and records a user's emotional state from multiple information sources such as voice, facial expressions, and biometric data.

[1365] A "response generation means" is a system or method that automatically generates an appropriate response or feedback based on the user's statements and emotional state.

[1366] "Means for analyzing and summarizing recorded data" refers to a system or method for analyzing audio data and concisely summarizing its contents.

[1367] "Means for displaying meeting minutes" refers to an interface for presenting summarized rebate details in a format that users can view.

[1368] "Means for generating presentation material templates" refers to a system or method that automatically creates a template containing the necessary points, data, graphs, and charts for a presentation, based on summarized meeting minutes.

[1369] "Methods for refining presentation material templates" refers to support systems or methods that allow users to edit generated templates and complete them as final presentation materials.

[1370] This invention is a system that combines input of rebate topics, collection and summarization of related information, real-time recognition and recording of user statements, analysis of statements and generation of responses, analysis and summarization of recorded data, display of meeting minutes, generation of presentation materials, and material refinement, with an emotion engine that recognizes the user's emotions in real time. This system makes it possible to provide responses and feedback that take into account the emotional state of the user during the rebate process.

[1371] The server first receives a rebate topic from the user (teacher). Based on this rebate topic, the server collects relevant information from the internet and internal databases and summarizes it using natural language processing techniques. This technique uses common natural language processing models, such as BERT or GPT-4. The collected information is summarized and stored in the database. The terminal (tablet or robot) then prepares to provide the summarized information to the user (student).

[1372] At the start of the rebate process, the server converts the user's speech into text data in real time using speech recognition technology and records it. For speech recognition, a service such as Google Speech-to-Text is used. The user's speech is analyzed, and key keywords are identified using keyword extraction algorithms (e.g., TF-IDF or word cloud).

[1373] In addition, the emotion engine recognizes and records the user's emotional state in real time along with their statements. The emotion engine determines the user's emotional state based on biometric data such as voice tone, facial expression analysis (e.g., Azure Face API), and heart rate. For example, if a user says, "Renewable energy is expensive," and the emotion engine simultaneously detects "anxiety" from the user's voice tone, the server adjusts its response based on this information.

[1374] The terminal (rebate robot) generates appropriate responses based on the user's statements and emotions. For example, a response generated by the server might include something like, "Although the initial investment in solar power generation is high, it offers cost savings in the long term, so you can rest assured."

[1375] After the rebate session ends, the server analyzes the recorded data and summarizes the details of the rebate and the user's emotional state. The summarized content is compiled into meeting minutes and displayed on the terminal. The meeting minutes are reviewed by the user (teacher) and revised as needed.

[1376] Next, the server generates a presentation template based on the meeting minutes. This template includes key points and data, charts, graphs, and information about the user's emotional state. The terminal displays this presentation template to the user (student). The user (student) refines the material with the rebate robot and saves it as the final version.

[1377] To give a concrete example, in response to the rebate topic "The pros and cons of spreading renewable energy," student A states, "Renewable energy is expensive," and the emotion engine detects "anxiety." Based on this information, the server generates an appropriate response, "While the initial investment in solar power generation is high, it offers cost-saving benefits in the long term, so it's reassuring," and the terminal (rebate robot) responds. Subsequently, the entire rebate is analyzed, a summarized meeting minute and presentation materials are generated, and finally, the user (student) and the rebate robot refine the materials to complete them.

[1378] Examples of prompts for a generative AI model include:

[1379] "Regarding the rebate topic 'The pros and cons of widespread renewable energy,' generate an appropriate response using student statements and sentiment data. A student stated, 'Renewable energy is expensive,' and the sentiment engine detected 'anxiety.' Provide the optimal response."

[1380] It can be used as such.

[1381] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1382] Step 1: Enter the rebate topic.

[1383] Input: The user (teacher) enters the rebate topic using a terminal.

[1384] Specific operation: The user (teacher) enters a topic such as "the pros and cons of promoting renewable energy" into the terminal interface. This information is sent from the terminal to the server.

[1385] Output: The input topic information is received by the server.

[1386] Step 2: Gather and summarize relevant information

[1387] Input: Rebate topic information received by the server.

[1388] Specific operation: The server searches for relevant information from the internet and internal databases, and summarizes the information using natural language processing techniques (e.g., BERT or GPT-4).

[1389] Data processing: Analyze data from the sources used, extract necessary information, and summarize it.

[1390] Output: The summarized information is saved to the database.

[1391] Step 3: Prepare to provide summary information

[1392] Input: Summarized information.

[1393] Specific operation: The terminal receives summary information from the server and prepares it to display the information in a format that is easy for the user (student) to use.

[1394] Output: Summary information in a format viewable by the user (student).

[1395] Step 4: Initiating the rebate and recognizing the statement

[1396] Input: The rebate process begins, and the user's (student's) comments are collected via the microphone.

[1397] Specific operation: The server uses speech recognition technology (e.g., Google Speech-to-Text) to convert and record speech data into text data in real time.

[1398] Data processing: Real-time text conversion of audio data.

[1399] Output: Text data of the user's (student's) statements.

[1400] Step 5: Recognizing the user's emotions

[1401] Input: User (student) speech data and biometric data (voice tone, facial expressions, heart rate).

[1402] Specific operation: The emotion engine analyzes this data to determine the user's (student's) emotional state. Sentiment Analysis technology is used for voice tone analysis, facial expression analysis APIs for facial expression analysis, and biosensors for biometric data analysis.

[1403] Output: User (student) emotion recognition data.

[1404] Step 6: Generating the response

[1405] Input: Transcribed speech data and sentiment recognition data.

[1406] Specific operation: The terminal (rebate robot) uses an NLP model to generate an appropriate response based on statements and sentiment data from the server.

[1407] Data processing: A response generation process that takes emotional states into consideration.

[1408] Output: The generated response.

[1409] Step 7: Analysis and summary after the rebate is completed.

[1410] Input: Voice and emotion data collected during rebates.

[1411] Specific operation: The server analyzes the recorded data and summarizes the rebate details and emotional state. The recorded data is then subjected to speech recognition again, converted to text, and analyzed.

[1412] Data processing: The process of summarizing from multiple data sources.

[1413] Output: Summarized meeting minutes data.

[1414] Step 8: View and review the meeting minutes.

[1415] Input: Summarized meeting minutes data.

[1416] Specific operation: The terminal displays the meeting minutes, and the user (teacher) reviews the content and makes corrections as needed.

[1417] Output: Reviewed and revised meeting minutes.

[1418] Step 9: Generating presentation materials

[1419] Input: Reviewed and revised meeting minutes.

[1420] Specific operation: The server automatically generates a presentation template based on the meeting minutes, including key points, data, graphs, and charts.

[1421] Data processing: The process of visualizing data and creating templates.

[1422] Output: Generated presentation template.

[1423] Step 10: Refining the presentation materials

[1424] Input: Presentation material template.

[1425] Specific operation: The user (student) edits materials together with the rebate robot and refines them into final presentation materials.

[1426] Output: Final presentation materials.

[1427] (Application Example 2)

[1428] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1429] Traditional rebate systems and in-store customer service systems have difficulty recognizing and analyzing user statements in real time and generating responses that take into account the user's emotional state. This has resulted in challenges in providing services that accurately capture user needs and emotions, hindering improvements in customer satisfaction and efficient business negotiations.

[1430] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for inputting a rebate topic, means for collecting and summarizing information related to the rebate topic, means for displaying information related to the rebate topic, means for recognizing and recording user statements in real time, means for analyzing user statements and generating appropriate responses, means for analyzing and summarizing rebate recording data, means for displaying summarized meeting minutes, means for generating presentation materials based on the meeting minutes, means for refining the presentation materials with the user, means for analyzing conversations with customers in real time during business negotiations and converting customer statements into text data, means for determining the content of customer statements and emotional state and generating appropriate responses based thereon, and means for displaying the generated responses on the store clerk's terminal. This makes it possible to analyze user statements and emotions in real time and provide accurate responses in rebate and in-store customer service.

[1431] A "rebate topic" refers to the subject matter that becomes the theme of a discussion or debate.

[1432] "Real-time recognition" refers to the immediate recording and analysis of speech and actions.

[1433] "Text data" refers to a data format in which audio is converted into written information.

[1434] "Emotional state" refers to the user's emotional and psychological state.

[1435] "Appropriate response" refers to providing the most suitable answer based on the user's statements and emotions.

[1436] "Meeting minutes" refers to a record that summarizes the content of a meeting or discussion.

[1437] "Presentation materials" refer to explanatory documents created for purposes such as rebates or presentations.

[1438] "Brush-up" refers to the process of improving existing materials and information to create a higher-quality version.

[1439] "Business negotiation" refers to dialogue and negotiation with customers in a physical store.

[1440] "Device" refers to a device such as a smartphone, tablet, or smart glasses.

[1441] Modes for carrying out the invention

[1442] This invention is a system that recognizes the customer's statements and emotional state in real time during business negotiations with customers who visit the store, and generates appropriate responses. First, the server has a means of inputting a rebate topic and collects and summarizes information related to the topic from the internet or an internal database. The server sends this summarized information to a terminal, and the terminal displays the information.

[1443] Next, the device converts the customer's speech into text data in real time using speech recognition technology. The Python library `speech_recognition` is used for this speech recognition. Furthermore, the device analyzes the user's speech and uses an emotion engine to determine the customer's emotions. This emotion engine analyzes the user's emotional state based on voice tone analysis, facial expression analysis, and biometric data such as heart rate.

[1444] Based on the analyzed utterances and emotional state, the device generates an appropriate response. A generative AI model is used for response generation, ensuring the user receives the most suitable reply. These processes are seamlessly coordinated between the server and the device, utilizing devices such as smartphones, tablets, or smart glasses.

[1445] For example, if a customer visiting an electronics store asks, "Isn't this TV too expensive?", the terminal uses speech recognition technology to convert this statement into text data, and an emotion engine detects "anxiety" from the customer's tone of voice. Subsequently, a generative AI model determines that the customer's statement, "I feel the price of the new TV is too high," is anxiety-inducing and generates an appropriate response: "The initial price is on the higher side, but in the long run, you can save on electricity costs, making it a good deal overall." This response is immediately displayed on the terminal, allowing store employees to refer to it while assisting the customer.

[1446] A concrete example of a prompt message is: "I feel that the price of the new TV is too high." This prompt should be interpreted as a sign of concern, and an appropriate response should be generated to reassure the customer. Using such concrete examples enables the effective implementation of this system.

[1447] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1448] Step 1:

[1449] The server receives rebate topics from users (store clerks). The rebate topics entered by the user are sent to the server. The input data includes the topic name and a detailed description.

[1450] Step 2:

[1451] The server collects and summarizes relevant information from the internet and internal databases based on the received rebate topic. The collected data is summarized in text format, with key points extracted by a summarization algorithm. The summarized information is stored on the server. Input data is the raw data of the relevant information, while output data is the summarized text data.

[1452] Step 3:

[1453] The server sends summarized information to a device (smartphone, tablet, smart glasses, etc.). The device displays the received summarized information on its screen. The input data is summarized text data, and the output data is data in a format that can be displayed on the screen.

[1454] Step 4:

[1455] The device uses its microphone to capture what the user (customer) says aloud. The input data is raw audio data.

[1456] Step 5:

[1457] The device converts the acquired audio data into text data using the speech_recognition library. This speech recognition process captures the customer's spoken content in text format. The input data is audio data, and the output data is text data.

[1458] Step 6:

[1459] The device uses a generative AI model and an emotion engine to analyze acquired text data and determine the customer's emotional state. This analysis utilizes data such as voice tone, facial expression analysis, and heart rate. Input data consists of text data and biometric data, while output data is the analysis result (spoken content and emotional state).

[1460] Step 7:

[1461] The device generates an appropriate response using a generative AI model based on the analyzed utterance and emotional state. For example, if the utterance includes the keyword "too expensive" and the emotion of anxiety is detected, the response generation algorithm will generate an appropriate response such as "The initial price is a bit high, but it's a good deal in the long run." The input data is the analysis result, and the output data is the generated response.

[1462] Step 8:

[1463] The terminal displays the generated response on the user's (store clerk's) screen. This allows the clerk to immediately confirm and respond to the customer appropriately. The input data is the generated response, and the output data is the displayed text information.

[1464] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1465] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1466] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1467] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1468] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1469] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1470] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1471] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1472] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1473] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1474] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1475] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1476] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1477] 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.

[1478] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1479] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1480] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1481] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1482] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1483] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1484] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1485] The following is further disclosed regarding the embodiments described above.

[1486] (Claim 1)

[1487] A means of entering a rebate topic,

[1488] A means of collecting and summarizing information related to rebate topics,

[1489] A means of displaying information related to the rebate topic,

[1490] A means of recognizing and recording user statements in real time,

[1491] A means for analyzing user statements and generating appropriate responses,

[1492] A means of analyzing and summarizing rebate recordings,

[1493] A means of displaying summarized meeting minutes,

[1494] A method for generating presentation materials based on meeting minutes,

[1495] A means of refining presentation materials with users,

[1496] A system that includes this.

[1497] (Claim 2)

[1498] The system according to claim 1, wherein the means for analyzing a user's statements is to extract and analyze keywords from the statements.

[1499] (Claim 3)

[1500] The system according to claim 1, comprising means for providing the generated presentation materials in a format including charts and graphs.

[1501] "Example 1"

[1502] (Claim 1)

[1503] A means of entering a rebate topic,

[1504] A means of collecting and summarizing information related to rebate topics,

[1505] A means of displaying information related to the rebate topic,

[1506] A means of recognizing and recording user statements in real time,

[1507] A means for analyzing user statements and generating appropriate responses,

[1508] A means of analyzing and summarizing rebate recordings,

[1509] A means of displaying summarized meeting minutes,

[1510] A method for generating presentation materials based on meeting minutes,

[1511] A means of refining the generated presentation materials with the user,

[1512] A means of collecting relevant information from the internet and internal databases based on rebate topics entered by users,

[1513] A means of summarizing the collected information and storing it in a database,

[1514] Means for using natural language processing techniques in the collection and summarization process,

[1515] Means of accessing external data sources via an internet connection,

[1516] A means of recording each user's speech as text data in real time using speech recognition technology,

[1517] A method for analyzing recorded statements using a keyword extraction algorithm,

[1518] A means of using a generative AI model to generate responses based on keywords,

[1519] A means of inputting a prompt sentence into a generative AI model to generate an appropriate response,

[1520] A system that includes this.

[1521] (Claim 2)

[1522] The system according to claim 1, wherein the means for analyzing a user's statements is to extract and analyze keywords from the statements.

[1523] (Claim 3)

[1524] The system according to claim 1, comprising means for providing the generated presentation materials in a format including charts and graphs.

[1525] "Application Example 1"

[1526] (Claim 1)

[1527] A means of entering a rebate topic,

[1528] A means of collecting and summarizing information related to rebate topics,

[1529] A means of displaying information related to the rebate topic,

[1530] A means of recognizing and recording user statements in real time,

[1531] A means for analyzing user statements and generating appropriate responses,

[1532] A means of analyzing and summarizing rebate recordings,

[1533] A means of displaying summarized meeting minutes,

[1534] A method for generating presentation materials based on meeting minutes,

[1535] A means of refining presentation materials with users,

[1536] A method for collecting problems and improvement suggestions in the production site in real time using speech recognition technology,

[1537] A means of extracting keywords from speech-recognized text and collecting related information,

[1538] A means of summarizing and displaying the collected information,

[1539] A system that includes this.

[1540] (Claim 2)

[1541] The system according to claim 1, wherein the means for analyzing a user's statements is to extract and analyze keywords from the statements.

[1542] (Claim 3)

[1543] The system according to claim 1, comprising means for providing the generated presentation materials in a format including charts and graphs.

[1544] "Example 2 of combining an emotion engine"

[1545] (Claim 1)

[1546] A means of entering a rebate topic,

[1547] A means of collecting and summarizing information related to rebate topics,

[1548] Means for displaying summarized information,

[1549] A means of recognizing user speech in real time and converting it into text data,

[1550] A method for analyzing user statements and extracting keywords,

[1551] A means of recognizing and recording emotional states,

[1552] A means of generating an appropriate response based on the user's statements and sentiment data,

[1553] A means of analyzing and summarizing rebate recordings,

[1554] A means of displaying summarized meeting minutes,

[1555] A method for generating presentation material templates based on meeting minutes,

[1556] Methods for refining presentation material templates,

[1557] A system that includes this.

[1558] (Claim 2)

[1559] The system according to claim 1, wherein the means for analyzing user statements analyzes and records the user's emotional state.

[1560] (Claim 3)

[1561] The system according to claim 1, comprising means for providing the generated presentation materials in a format including charts and graphs.

[1562] "Application example 2 when combining with an emotional engine"

[1563] (Claim 1)

[1564] A means of entering a rebate topic,

[1565] A means of collecting and summarizing information related to rebate topics,

[1566] A means of displaying information related to the rebate topic,

[1567] A means of recognizing and recording user statements in real time,

[1568] A means for analyzing user statements and generating appropriate responses,

[1569] A means of analyzing and summarizing rebate recordings,

[1570] A means of displaying summarized meeting minutes,

[1571] A method for generating presentation materials based on meeting minutes,

[1572] A means of refining presentation materials with users,

[1573] A method for analyzing conversations with customers during business negotiations in real time and converting customer statements into text data,

[1574] A means for determining the content of a customer's statements and emotional state, and for generating an appropriate response based on this,

[1575] A means of displaying the generated response on the store clerk's terminal,

[1576] A system that includes this.

[1577] (Claim 2)

[1578] The system according to claim 1, wherein the means for analyzing a user's statements is to extract and analyze keywords from the statements.

[1579] (Claim 3)

[1580] The system according to claim 1, comprising means for providing the generated presentation materials in a format including charts and graphs. [Explanation of Symbols]

[1581] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of entering a rebate topic, A means of collecting and summarizing information related to rebate topics, A means of displaying information related to the rebate topic, A means of recognizing and recording user statements in real time, A means for analyzing user statements and generating appropriate responses, A means of analyzing and summarizing rebate recordings, A means of displaying summarized meeting minutes, A method for generating presentation materials based on meeting minutes, A means of refining presentation materials with users, A system that includes this.

2. The system according to claim 1, wherein the means for analyzing a user's statements is to extract and analyze keywords from the statements.

3. The system according to claim 1, comprising means for providing the generated presentation materials in a format including charts and graphs.

Images