system

Abstract

We provide the system. [Solution] Methods for obtaining manual information from the customer support database, A method for training a generative AI model with acquired manual information, A user interface means for receiving inquiries from users, A means of sending received inquiries to a generative AI model, A means of receiving and providing answers generated by a generative AI model to the user, A system that includes this.

Description

Technical Field

[0005]

[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, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In modern customer support operations, the increasing number of inquiries and the accompanying delay in responses have become problems. In particular, inquiries regarding technical manuals and FAQs of terminals are frequent, which has caused an increase in the labor and costs of sales support centers and care centers. Furthermore, customers are seeking quick and accurate answers, and it is difficult for conventional support systems to adequately respond to this. Against this background, there is a need to reduce the burden of customer support and improve customer satisfaction.

Means for Solving the Problems

[0005] This invention solves the aforementioned problems. Specifically, it provides a system that enables rapid responses by obtaining manual information from a customer support database and training a generative AI model with it. The above problems are solved by constructing a system that includes a user interface means for receiving inquiries from users, means for sending the received inquiries to a generative AI model, and means for receiving the answers generated by the generative AI model and providing them to the user. Furthermore, by including a rapid response system that provides answers generated by the generative AI model to the user in real time, the system improves the efficiency of inquiry handling and enhances customer satisfaction.

[0006] A "customer support database" is a database used to store and manage support information provided to customers.

[0007] "Manual information" refers to document data that includes information on how to use, configure, and troubleshoot products and services.

[0008] A "generative AI model" is an artificial intelligence model that learns from large amounts of text data and generates the optimal answer to a user's inquiry.

[0009] "User interface means" refers to interface means that include screens and input devices for users to interact with the system.

[0010] A "rapid response system" is a system that provides users with answers generated by generative AI models in real time. [Brief explanation of the drawing]

[0011] [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] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0012] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0013] First, let's explain the terminology used in the following explanation.

[0014] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0015] In the following embodiments, the labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0016] In the following embodiments, the labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0017] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark), and the like.

[0018] 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."

[0019] [First Embodiment]

[0020] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0021] 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.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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.

[0027] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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".

[0032] This invention provides a system that quickly responds to user inquiries by retrieving manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below.

[0033] Specific examples of program processing

[0034] Database Preparation

[0035] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions for new products in the database. The database contains data in various formats, such as PDFs, text files, and images.

[0036] Learning of generative AI

[0037] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, the server extracts "product initial setup instructions" as text and converts it into a format that the AI ​​model can understand.

[0038] Receiving inquiries from users

[0039] The device provides an interface for receiving inquiries from users. For example, the device can receive inquiries from users such as "How do I connect my product via Bluetooth?" through a web form or chatbot.

[0040] Handling inquiries

[0041] The device sends the user's inquiry to the server. The server receives this and forwards it to a generative AI model. For example, the device receives the question, "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0042] Generative AI models generate appropriate answers to queries based on information stored in a database. For example, an AI model might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0043] Providing a response

[0044] The server sends the response received from the generative AI model to the terminal. The terminal then displays this to the user. For example, the server sends the received steps to the terminal, which then displays them in a chat window for the user to see. In this way, the user can quickly resolve the problem.

[0045] Specific example

[0046] Example 1: How to deal with a power failure

[0047] The server registers "solutions for when the power won't turn on" in its database. The server then uses this information to train a generative AI model.

[0048] When a user enters the question "What should I do if the device won't turn on?" into the chatbot, the device receives this and sends it to the server. The generative AI model generates the response "Try restarting by pressing and holding the power button for 10 seconds." The server sends this response back to the device, which then displays it to the user.

[0049] This system enables faster and more efficient responses compared to traditional customer support, reducing the burden on sales support centers and care centers, and is expected to improve customer satisfaction.

[0050] The following describes the processing flow.

[0051] Step 1:

[0052] The server registers product and service manual information in the customer support database. This includes initial setup instructions and troubleshooting guides for Hianabi. The server stores this information in the database in PDF and text file formats.

[0053] Step 2:

[0054] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. For example, the server extracts text from PDF files and converts it to a format that can be used by generative AI models for training.

[0055] Step 3:

[0056] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This process prepares the AI ​​model to understand text data and generate appropriate responses. For example, the server trains the AI ​​model on "initial setup procedures."

[0057] Step 4:

[0058] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect Hianabi via Bluetooth?"

[0059] Step 5:

[0060] The device receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the device might send the question "How do I connect Hianabi via Bluetooth?" to the server, which then passes this to the AI.

[0061] Step 6:

[0062] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0063] Step 7:

[0064] The server sends the response received from the generative AI model to the terminal. The server holds the generated response and then sends it to the terminal. This process is rapid, so the user can receive a response immediately.

[0065] Step 8:

[0066] The device displays the user the response received from the server. For example, the device might display instructions in the chat window such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0067] In this way, specific processes are carried out at each step, resulting in a system that responds quickly and accurately to user inquiries.

[0068] (Example 1)

[0069] 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."

[0070] Traditional customer support systems often suffered from delays in responding to user inquiries and a failure to provide appropriate information. Furthermore, some inquiries required manual handling by operators, leading to decreased efficiency. These problems could negatively impact the quality and speed of customer support, potentially harming customer satisfaction.

[0071] 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.

[0072] In this invention, the server includes means for acquiring information from a customer support database, means for preprocessing and formatting the acquired information to train a generative AI model, and means for receiving inquiries from users. This makes it possible to provide immediate and appropriate answers to inquiries from users.

[0073] A "customer support database" is a database that stores support information for users and manual information about products and services.

[0074] "Information" refers to data such as detailed instructions, troubleshooting guides, and FAQs about products and services.

[0075] "Preprocessing" refers to a series of processes that convert data into a format that generative AI models can understand. Specifically, this includes tasks such as extracting text from PDFs and format conversion.

[0076] "Format conversion" refers to the process of converting acquired information into a format that is easily understood by generative AI models.

[0077] A "generative AI model" is an artificial intelligence model that learns from information obtained from a customer support database and generates appropriate answers to user inquiries.

[0078] "Interface means" refers to input methods such as web forms and chatbots that users use to enter inquiries.

[0079] A "server" refers to a device that retrieves information from a customer support database, performs preprocessing and format conversion, and supplies the data to a generative AI model.

[0080] "Inquiry" refers to the act of a user inputting specific questions or problems about a product or service through an interface.

[0081] "Answer" refers to the information that a generative AI model generates based on a user's inquiry.

[0082] A "rapid response system" refers to a system that provides users with answers generated by generative AI models in real time.

[0083] This invention provides a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below from the perspectives of the server, terminal, and user.

[0084] Database Preparation

[0085] The server registers manual information about products and services in the customer support database. Specifically, it stores information such as initial product setup procedures and troubleshooting guides in various formats, including PDFs, text files, and images. For example, it uploads the setup instructions for a new smartphone to the database.

[0086] Information extraction and preprocessing

[0087] The server extracts necessary manual information from the customer support database. During this process, it performs data preprocessing and format conversion. Specifically, it extracts text from PDFs and converts it into a format that generative AI models can understand. For example, it extracts the text "Connect the device to the power source and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[0088] Training of generative AI models

[0089] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. Specifically, it inputs text data into the AI ​​model and repeats a feedback loop with the generated response. For example, by training the AI ​​model with a manual on "how to turn on the power," it can generate the response "Press and hold the power button."

[0090] Receiving inquiries from users

[0091] The device provides an interface for receiving inquiries from users. Specifically, it receives user questions using web forms or chatbots. For example, the device might receive a user inquiry such as, "Please tell me how to set up my new smartphone."

[0092] Processing queries and sending them to the server

[0093] The device forwards the user's inquiry to the server. Specifically, it converts questions entered into web forms or chatbots into the appropriate format and sends them to the server. For example, it forwards the question "How do I connect my smartphone via Bluetooth?" to the server.

[0094] Answer generation using generative AI models

[0095] The server passes the received user inquiry to a generative AI model. The generative AI model generates an answer to the question based on the information stored in the database. Specifically, in response to the question, "How do I connect my smartphone via Bluetooth?", it generates the answer, "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[0096] Providing a response

[0097] The server sends the response received from the generative AI model to the terminal, which then displays it to the user. Specifically, the server-generated response is displayed to the user through a chatbot or web form. For example, the terminal might provide the user with the response, "Open the settings menu, select Bluetooth, and choose the device you want to connect to," in a chat window.

[0098] Specific example

[0099] Example 1: How to deal with a power failure

[0100] 1. The server registers "Troubleshooting steps for when the power does not turn on" as a PDF in its database.

[0101] 2. The server extracts the text "Please try restarting by pressing and holding the power button for 10 seconds" from the registered PDF and preprocesses it.

[0102] 3. The server uses the preprocessed text to train a generative AI model.

[0103] 4. The user enters "What should I do if the device won't turn on?" into the device's chatbot.

[0104] 5. The terminal sends the user's inquiry to the server.

[0105] 6. The generative AI model generates the response, "Please try restarting by pressing and holding the power button for 10 seconds."

[0106] 7. The server sends the generated response to the terminal, which then displays it in the chat window and provides it to the user.

[0107] In this way, users can quickly solve problems. As a specific example, the following prompt can be input into the generating AI model.

[0108] Example of a prompt

[0109] "Please tell me how to connect the product via Bluetooth."

[0110] "Please tell me what to do if the power won't turn on."

[0111] This system is expected to enable faster and more efficient responses compared to traditional customer support, reducing the workload on operators and improving customer satisfaction.

[0112] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0113] Step 1:

[0114] The server registers manual information about products and services in the customer support database. Inputs include initial setup instructions and troubleshooting guides for new products, provided in PDF, text file, and image formats. The server stores this information in the database and generates an organized database as output. Specifically, it uploads new product setup instructions as PDFs to the database.

[0115] Step 2:

[0116] The server extracts necessary manual information from the customer support database. PDFs and text files stored in the database are used as input. The server processes the data and converts it into a format suitable for generative AI models. The output is text data that the AI ​​model can understand. Specifically, it extracts the text "Connect the device to the power supply and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[0117] Step 3:

[0118] The server supplies pre-processed manual information to a generative AI model for training. Formatted text data is used as input. The server inputs this into the AI ​​model and improves the model's performance through a feedback loop. The output is a fully trained AI model. As a specific example, the AI ​​model is trained using text data about "how to turn on the power."

[0119] Step 4:

[0120] Users utilize an interface to submit inquiries from their devices. As input, the user's question is entered into the device via a web form or chatbot. The device receives this input and sends it to the server. As output, the inquiry content is sent to the server. For example, a user might enter the inquiry "Please tell me the initial setup procedure for my new smartphone" into the chatbot.

[0121] Step 5:

[0122] The terminal converts the received inquiry into the appropriate format and sends it to the server. The user's inquiry is provided to the terminal as input. The terminal converts this into the appropriate format and sends it to the server. The formatted inquiry is sent to the server as output. As a specific example, the question "How do I connect my smartphone via Bluetooth?" is converted into the appropriate format and sent to the server.

[0123] Step 6:

[0124] The server passes the user's inquiry received to a generative AI model. The server receives the inquiry content sent from the terminal as input. The server inputs this into the AI ​​model and generates an answer. The generated answer is obtained as output. For example, in response to the question "How do I connect my smartphone via Bluetooth?", it generates the answer "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[0125] Step 7:

[0126] The server sends the response received from the generative AI model to the terminal. The server receives the response generated by the AI ​​model as input. The server sends this to the terminal, which displays it to the user. The output is the response displayed to the user. Specifically, the server sends the generated response to the terminal via a chatbot or web form, and the terminal provides the user with the response "Open the settings menu, select Bluetooth, and choose the device you want to connect to" in the chat window.

[0127] (Application Example 1)

[0128] 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."

[0129] Current customer support systems struggle to respond quickly to user inquiries. This is especially true for e-commerce sites, where detailed product inquiries are frequent, requiring efficient and immediate responses. Furthermore, many users prefer to communicate via voice as well as text. However, few systems successfully balance the requirements for providing appropriate answers with user convenience, resulting in decreased user satisfaction and operational inefficiencies. These problems need to be addressed.

[0130] 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.

[0131] In this invention, the server includes means for obtaining manual information from a customer support database, means for training a generative AI model with the obtained manual information, means for a user interface for receiving inquiries from users, means for users to input questions about products in text or voice, means for sending received inquiries to the generative AI model, and means for receiving and providing answers generated by the generative AI model to the user. This makes it possible to respond quickly to user inquiries and improve user convenience.

[0132] A "customer support database" is a data storage system that stores information about products and services in order to respond to inquiries from users.

[0133] "Manual information" refers to documents and data that provide detailed information on how to use a product or service, setup procedures, troubleshooting, and more.

[0134] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers from given data.

[0135] A "user interface" is a software or hardware configuration that includes interactive elements that allow a user to directly interact with the system.

[0136] "Text or voice input" refers to the process by which users submit inquiries through text input or voice input.

[0137] A "rapid response system" is a mechanism that provides users with answers generated by generative AI models in real time.

[0138] This invention is a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. A specific embodiment of this system is described below.

[0139] System program

[0140] Database Preparation

[0141] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. The server also stores initial setup instructions for new products in the database.

[0142] Learning of generative AI

[0143] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, "product initial setup instructions" are extracted as text and converted into a format that the AI ​​model can understand.

[0144] User Interface

[0145] The user interface is an interface for users to enter questions about the product. This interface has the functionality to accept text input and voice input. It allows users to enter inquiries such as, "How do I connect the product via Bluetooth?"

[0146] Handling inquiries

[0147] The server receives the user's inquiry and forwards it to a generative AI model. The generative AI model generates an appropriate response based on the information stored in the database. For example, it might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0148] Providing a response

[0149] The server sends the response received from the generative AI model to the user interface. The user interface provides the response to the user in text or voice.

[0150] Hardware and software to be used

[0151] This system uses the following hardware and software.

[0152] Hardware:

[0153] Smartphone (the device on which the user uses the app)

[0154] Servers (data centers where databases and AI models run)

[0155] software:

[0156] OpenAI® API (utilizes generative AI models)

[0157] A temporary database module (a module that communicates with the database)

[0158] A temporary chatbot interface (a module that interacts with the user)

[0159] Examples of specific cases and prompt statements

[0160] Specific example

[0161] A user asks, "I'd like to know more about the shipping method for this product." The smartphone app sends the question to the server and retrieves the "shipping method" information from the database. The AI ​​generates a response saying, "Regarding shipping methods, it is usually shipped by courier service, but if you are in a hurry, an express service is available for an additional fee." The app then displays this response to the user.

[0162] Example of a prompt

[0163] The following are examples of prompts to input into the generated AI model.

[0164] User question: I'd like to know more about the shipping methods for this product.

[0165] Search results from our customer support database: Our standard shipping method is courier service, but for urgent orders, an express delivery service is available for an additional fee.

[0166] Answer the user's query based on the above information.

[0167] In this way, this invention makes it possible to realize a system that responds quickly and accurately to user inquiries.

[0168] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0169] Step 1: Prepare the database

[0170] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in the database. The input here is the manual information, and the output is the information registered in the database.

[0171] Step 2: Training a generative AI

[0172] The server extracts the necessary manual information from the database and uses it to train a generative AI model. Data preprocessing and format conversion are performed to transform the information into a format that the AI ​​model can understand. For example, the server extracts "product initial setup procedures" in text format and feeds them into the generative AI model. Here, the input is the manual information extracted from the database, and the output is the completed generative AI model.

[0173] Step 3: User Interface Operation

[0174] The user enters questions about the product through the terminal's user interface. This interface accepts both text and voice input. Specifically, the user might enter the question, "How do I connect the product via Bluetooth?" The input here is the user's question, and the output is the content of that question.

[0175] Step 4: Submitting the inquiry

[0176] The terminal sends the user's query to the server. The server forwards the received query to a generative AI model. At this stage, the terminal sends the received query to the server, and the server passes it to the AI. The input here is the user's query, and the output is the prompt message passed to the AI ​​model.

[0177] Step 5: AI-powered response generation

[0178] Generative AI models generate appropriate answers based on information stored in a database. For example, an AI model might generate the following procedure: "First, select Bluetooth from the settings menu, then search for the device you want to connect to." The input here is the prompt received from the server, and the output is the generated answer.

[0179] Step 6: Provide your response

[0180] The server sends the response received from the generative AI model to the terminal. The terminal provides the response to the user in text or voice. For example, the server sends the received procedure to the terminal, and the terminal displays it to the user. Here, the input is the response received from the AI ​​model, and the output is the response provided to the user through the user interface.

[0181] 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.

[0182] This invention provides a system that responds quickly and accurately to user inquiries by acquiring manual information from a customer support database and training a generative AI model with it. Furthermore, it is a system that combines this with an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response based on those emotions.

[0183] Specific examples of program processing

[0184] Database Preparation

[0185] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions and troubleshooting guides for new products in the database. The database includes PDFs, text files, images, and other formats.

[0186] Learning of generative AI

[0187] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. The server then supplies the preprocessed text data to a generative AI model for training. For example, the server extracts "initial setup procedure" as text and uses it to train the AI ​​model.

[0188] Receiving inquiries from users

[0189] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[0190] Handling inquiries

[0191] The terminal receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the terminal receives the question "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0192] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0193] How the emotion recognition engine works

[0194] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. The emotion recognition engine detects the user's emotional state from the context of the input, voice, facial expressions, etc. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[0195] The server can adjust the content and expression of responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the user is irritated, the server will change the response to a more polite expression.

[0196] Escalation engine operation

[0197] The server activates the escalation engine if the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user's emotion is recognized as "very angry," the escalation engine will contact the support team leader.

[0198] Providing a response

[0199] The server sends the responses generated by the generative AI model and emotion recognition engine to the device. The device then displays this to the user. For example, the server sends a detailed instruction to the device, such as "Select Bluetooth from the settings menu, then search for the device you want to connect to," which the device then displays in the chat window for the user. The user then uses this information to configure the product.

[0200] This system recognizes user emotions and provides quick and appropriate responses, enabling more efficient support compared to traditional customer support. The introduction of an emotion recognition engine can further improve user satisfaction.

[0201] The following describes the processing flow.

[0202] Step 1:

[0203] The server registers manual information about products and services in the customer support database. This includes tasks such as storing initial product setup instructions and troubleshooting guides in PDF or text file format in the database.

[0204] Step 2:

[0205] The server extracts the necessary manual information from the database and preprocesses that information. Preprocessing includes cleaning and formatting text data. For example, the server extracts text from PDF files and converts it into a format that generative AI models can understand.

[0206] Step 3:

[0207] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This enables the AI ​​to generate appropriate answers to user inquiries. For example, the server trains the AI ​​model on "initial setup procedures."

[0208] Step 4:

[0209] The device provides an interface for receiving inquiries from users. Users enter their problems or questions through web forms or chatbots. For example, a user might type, "Please tell me how to connect the product via Bluetooth."

[0210] Step 5:

[0211] The terminal receives the user's inquiry and then sends that information to the server. The server then receives this inquiry and sends it to a generative AI model. For example, the terminal receives the inquiry "Please tell me how to connect the product via Bluetooth" and sends it to the server, which then passes it on to the AI ​​model.

[0212] Step 6:

[0213] Generative AI models analyze the content of inquiries and generate the most suitable answers based on information stored in a database. For example, they might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0214] Step 7:

[0215] When a terminal receives a user inquiry, it sends the input content to an emotion recognition engine. The emotion recognition engine analyzes the context of the input text, as well as voice and facial expressions, to detect the user's emotional state. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[0216] Step 8:

[0217] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the server detects that the user is angry, it will change the response to a more polite and calm expression.

[0218] Step 9:

[0219] The server activates the escalation engine if it determines that the user's emotional state is "urgent." The escalation engine follows a predefined procedure to escalate the issue to a higher-level support person. For example, if it is determined that the user is "very angry," the support team leader will be contacted.

[0220] Step 10:

[0221] The device displays responses generated by a generative AI model and an emotion recognition engine to the user. For example, it might display detailed instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0222] Through the steps described above, a system incorporating an emotion recognition engine can provide prompt and appropriate support that takes user emotions into account. This makes it possible to improve the efficiency of customer support and increase user satisfaction.

[0223] (Example 2)

[0224] 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".

[0225] Conventional customer support systems often suffer from delays in responding to user inquiries or fail to provide appropriate answers. Furthermore, they may fail to consider user emotions, leading to dissatisfaction. This results in decreased user satisfaction and customer churn. The present invention aims to solve these problems by providing a system that responds quickly and accurately to user inquiries, and further recognizes and appropriately addresses user emotions.

[0226] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0227] In this invention, the server includes means for acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, and means for an emotion recognition engine for analyzing the user's emotions. This enables a quick and appropriate response to user inquiries, and further, the provision of appropriate answers that take the user's emotions into consideration.

[0228] A "customer support database" is a database that stores various manuals and troubleshooting guides related to products and services, and is used to handle customer inquiries.

[0229] "Manual information" refers to information that includes detailed instructions and guides on how to use a product or service, setup procedures, troubleshooting, and more.

[0230] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate appropriate responses from input data.

[0231] A "user interface" is the interface through which a user makes inquiries to a system, and includes web forms, chatbots, and other similar elements.

[0232] An "emotion recognition engine" is an engine that analyzes and recognizes emotions from the user's input text, voice, and facial expressions.

[0233] "Methods for escalating the issue" refer to ways of transferring the problem to a higher-level person in charge, depending on the user's emotional state.

[0234] A "rapid response system for real-time delivery" is a system that provides generated answers to users immediately, with the aim of responding quickly to inquiries.

[0235] This invention provides a system that responds quickly and appropriately to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. Furthermore, this system incorporates an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response accordingly.

[0236] Database Preparation

[0237] The server registers manual information about products and services in the customer support database. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in PDF, text file, and image formats in the database. For example, the server saves "Printer Ink Replacement Procedure" as a PDF file in the database.

[0238] Learning of generative AI

[0239] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning the text data (removing unnecessary spaces and line breaks) and format conversion (from PDF to text). The server then feeds this preprocessed text to a generative AI model. By training the generative AI model (e.g., GPT-3®) with this data, the AI ​​learns how to configure and troubleshoot the product.

[0240] Receiving inquiries from users

[0241] The device provides an interface for receiving inquiries from users. Users enter their questions using web forms or chatbots. For example, a user might type "How do I connect the product via Bluetooth?" into the chatbot.

[0242] Handling inquiries

[0243] The terminal receives a user inquiry and sends its contents to the server. The server analyzes the inquiry and sends it to a generative AI model. For example, the terminal sends the question "How do I connect the product via Bluetooth?" to the server, which then passes this to the AI. The generative AI model analyzes the inquiry and generates an appropriate answer based on the information stored in the database. For example, the generative AI model might generate the procedure "Select Bluetooth from the settings menu, and then search for the device you want to connect to."

[0244] How the emotion recognition engine works

[0245] The device uses an emotion recognition engine to analyze the user's emotions when they make an inquiry. The emotion recognition engine detects the user's emotional state from the context of the input, voice, and facial expressions. For example, if a user sends a message that includes frustration, such as "Tell me quickly!", the emotion recognition engine will analyze this as "frustration." Based on the user's emotional state analyzed by the emotion recognition engine, the server adjusts the content and expression of the response generated by the generative AI model. For example, if the user is frustrated, the server adjusts the response to provide a more polite and prompt response.

[0246] Escalation engine operation

[0247] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user is identified as "very angry," the escalation engine will contact the support team leader and have them handle the issue directly.

[0248] Providing a response

[0249] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The device displays this to the user. The generated response, "Select Bluetooth from the settings menu and then search for the device you want to connect to," is politely rephrased according to the emotion to, "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to." This response is displayed in the device's chat window and provided to the user. The user uses this information to configure the product settings.

[0250] Examples of specific cases and prompt statements

[0251] Specific example

[0252] 1. User's question: "My printer is jammed. How can I fix it?"

[0253] 2. Server preprocessing: Extract troubleshooting guides for "paper jams" from the database and use them to train a generative AI model.

[0254] 3. Emotion Analysis: The emotion recognition engine detects "frustration" from the user's voice and facial expressions.

[0255] 4. Response generation and refinement: The generative AI model creates the procedure "turn off the power and carefully remove the paper," and the server refines the expression more carefully based on the emotion recognition results.

[0256] 5. Response Provision: The adjusted response is provided to the user by the device (for example, displayed in the chat window).

[0257] Example of a prompt

[0258] User's question: "My printer is jammed. How can I fix it?"

[0259] Sentiment Analysis: Detects irritation from input context, voice, and facial expressions.

[0260] Generative AI model's response: "Turn off the power and carefully remove the paper."

[0261] Emotion-based adjustment: "I'm sorry, but could you please turn off the printer and carefully remove the jammed paper?"

[0262] In this way, the system can respond to user inquiries efficiently and with consideration for their feelings.

[0263] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0264] Step 1:

[0265] The server registers manual information about products and services in the customer support database. Input includes manual information in PDF, text file, and image formats. The specific process for storing this information in the database involves the server receiving the manual information in various formats and saving it in the database in the appropriate format. For example, a PDF file titled "Printer Ink Replacement Procedure" might be saved in the database.

[0266] Step 2:

[0267] The server extracts the necessary manual information from the database and preprocesses it. The input is the manual information in the database. Specifically, the server cleans the text data (removing unnecessary spaces and line breaks) and converts the format (from PDF to text). The output is the preprocessed text data. For example, a PDF document titled "Initial Setup Procedure" is converted to text format.

[0268] Step 3:

[0269] The server supplies pre-processed text data to a generative AI model. The input is pre-processed text data. Specifically, the server feeds this data to a generative AI model (e.g., GPT-3) to train the model. The output is the trained generative AI model.

[0270] Step 4:

[0271] The terminal provides an interface for receiving inquiries from users. Input includes questions entered by the user via web forms or chatbots. Its specific operation is to retrieve the user's inquiry content through the user interface and send it to the server. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[0272] Step 5:

[0273] The server receives the user's inquiry and sends it to a generative AI model. The input is the inquiry sent from the terminal. Specifically, the server analyzes the inquiry and passes it to the generative AI model. The output is the answer generated by the generative AI model. For example, it might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0274] Step 6:

[0275] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. Input includes the user's written text, voice, and facial expressions. The emotion recognition engine analyzes this information to detect the user's emotional state. The output is the analyzed emotional state of the user (e.g., "irritated").

[0276] Step 7:

[0277] The server adjusts the response of the generative AI model based on the results analyzed by the emotion recognition engine. The inputs are the generated response and the emotion state from the emotion recognition engine. Specifically, the server adjusts the response generated by the generative AI model according to the emotion state, changing its expression. The output is the adjusted response. For example, the response might be changed to "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to."

[0278] Step 8:

[0279] The server escalates the problem to a higher-level person in charge considering the user's emotional state. As input, there is an emotional state judged as "urgent" from the emotion recognition engine. The specific operation is that the server starts the escalation engine and transfers the problem to a higher-level support person in charge. As output, the escalated problem is notified to the higher-level person in charge.

[0280] Step 9:

[0281] The server sends the answer generated by the generative AI model and the emotion recognition engine to the terminal. As input, there is the adjusted answer. The specific operation is that the server sends this answer to the terminal, and the terminal displays it to the user. As output, the answer displayed to the user is obtained. For example, a detailed procedure such as "Select Bluetooth from the settings menu and then search for the device you want to connect to" is displayed in the chat window of the terminal.

[0282] (Application Example 2)

[0283] Next, Application Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0284] In the conventional customer support system, due to providing a uniform response without considering the user's emotional state, user satisfaction often decreased. Also, since the functions for quickly providing appropriate answers to inquiries were limited, efficient support was difficult. Furthermore, since real-time response using the voice interface in the store was lacking, it was necessary to improve the user experience.

[0285] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for acquiring manual information from a customer support database, means for causing the acquired manual information to be learned by a generative AI model, an emotion recognition engine means for analyzing the emotional state of the user, means for adjusting the content and expression of the answer based on the emotional state, an escalation engine means for escalating the problem to a higher-level support person based on the emotional state of the user, and a voice interface means for receiving the user's inquiry by voice using voice recognition and providing the response by voice. As a result, it becomes possible to improve user satisfaction, provide quick and appropriate answers, and provide effective customer support through a real-time voice interface.

[0286] The "customer support database" is a database that stores information such as manual information regarding products and services, troubleshooting guides, and FAQs.

[0287] "Manual information" is information that includes detailed explanations such as how to use products and services, setting procedures, and troubleshooting.

[0288] The "generative AI model" is an artificial intelligence model that automatically generates appropriate answers to user inquiries using natural language processing technology.

[0289] The "user interface" is an interface for the user to input inquiries and interact with the system.

[0290] The "emotion recognition engine" is an engine for analyzing the emotional state from the user's input and voice and detecting emotions such as "anger" and "joy".

[0291] The "escalation engine" is an engine for escalating the problem to a higher-level support person as necessary based on the emotional state of the user analyzed by the emotion recognition engine.

[0292] A "rapid response system" is a system that provides users with answers generated by generative AI models in real time.

[0293] A "voice interface" is an interface that uses speech recognition technology to receive user inquiries in voice and provide responses in voice.

[0294] This invention is a system for streamlining customer support in physical stores. It retrieves manual information from a customer support database and uses it to train a generative AI model, enabling it to respond quickly and accurately to user inquiries. It also incorporates an emotion recognition engine that analyzes the user's emotional state at the time of inquiry and adjusts the content and expression of the response accordingly.

[0295] The server retrieves product and service manual information (e.g., initial setup procedures and troubleshooting guides) from the customer support database. This manual information is stored in various formats, including PDFs, text files, and images. The server preprocesses this data and trains a generative AI model. This process involves cleaning and formatting the text data. For example, it might extract the initial setup procedures for a new product as text and train the generative AI model.

[0296] The terminal provides an interface for receiving user inquiries. This is done via applications installed on smartphones or robots placed within the store. Users can enter questions through web forms, chatbots, or voice interfaces. For example, if a user asks, "Do you have this product in stock?", the terminal sends this inquiry to the server.

[0297] The server sends the received inquiry to a generative AI model, which analyzes the inquiry and generates an appropriate response. For example, it might generate a response such as, "We will check the stock status. Please wait a moment." The server also uses an emotion recognition engine to analyze the user's emotions. It detects the user's emotional state (e.g., "anger") from the context of the input, voice, and facial expressions, and adjusts the content and expression of the response generated by the generative AI model based on this. If the user is angry, the response is changed to a more polite expression.

[0298] Furthermore, if the emotion recognition engine determines that the user's emotional state is "urgent," the server activates the escalation engine to escalate the issue to a higher-level support person. For example, if the user's emotions are recognized as "very angry," the server will contact the support team leader.

[0299] In summary, this invention enables improved user satisfaction, the provision of quick and appropriate responses, and effective customer support through a real-time voice interface.

[0300] Specific example

[0301] When a user asks, "Do you have this item in stock?", the speech recognition engine converts the speech into text, and the generative AI model generates the response, "We will check the stock status. Please wait a moment."

[0302] Examples of prompts to input into a generative AI model:

[0303] User: Do you have this item in stock?

[0304] assistant:

[0305] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0306] Step 1: Prepare the database

[0307] The server registers manual information about products and services in the customer support database. The inputs are PDFs, text files, and images such as the initial setup procedures for new products and troubleshooting guides. The server stores this information in the database. As output, the stored information becomes available.

[0308] Step 2: Training of the generative AI

[0309] The server extracts the necessary manual information from the database and performs cleaning and format conversion of the text data. The input is the manual information obtained from the database. The server supplies the preprocessed text data to the generative AI model for training. As output, the generative AI model can generate appropriate answers to inquiries.

[0310] Step 3: Receiving inquiries from users

[0311] The terminal provides an interface for receiving inquiries from users. The input is the questions entered by the user through a web form, chatbot, or voice interface. The terminal receives this inquiry and generates data for transmission to the server as output.

[0312] Step 4: Processing of inquiries

[0313] The server sends the inquiry content received from the terminal to the generative AI model. The input is the user's question. The server processes the data for the generative AI model to analyze the inquiry content and generate an appropriate answer. As output, it receives the generated answer.

[0314] Step 5: Operation of the sentiment recognition engine

[0315] The device uses an emotion recognition engine to analyze the user's emotions when a user makes a request. Input includes the user's context, voice, and facial expressions. The device analyzes this information and sends the user's emotional state (e.g., "anger," "frustration") to the server as output.

[0316] Step 6: Adjusting the content and wording of your answer

[0317] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. The input consists of the user's emotional state and the generated responses. The server adjusts these and outputs the adjusted responses.

[0318] Step 7: Operation of the escalation engine

[0319] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The input is the analyzed user's emotional state. The server escalates the issue to a higher-level support person and generates an escalated issue as output.

[0320] Step 8: Provide your answer

[0321] The server sends the responses generated by the generative AI model and emotion recognition engine to the terminal. The input is the adjusted response. The terminal displays this to the user, and the output is the information provided to the user.

[0322] As a concrete example, if a user asks, "Do you have this item in stock?", the device uses a speech recognition engine to convert the speech into text and sends it to the server. The server uses a generative AI model to generate an answer and an emotion recognition engine to analyze the user's emotional state. After appropriate adjustments are made, the device provides the user with an answer such as, "I will check the stock status. Please wait a moment."

[0323] Examples of prompts to input into a generative AI model:

[0324] User: Do you have this item in stock?

[0325] assistant:

[0326] 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.

[0327] 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.

[0328] 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.

[0329] [Second Embodiment]

[0330] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0331] 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.

[0332] 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).

[0333] 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.

[0334] 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.

[0335] 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).

[0336] 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.

[0337] 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.

[0338] 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.

[0339] 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.

[0340] 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.

[0341] 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".

[0342] This invention provides a system that quickly responds to user inquiries by retrieving manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below.

[0343] Specific examples of program processing

[0344] Database Preparation

[0345] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions for new products in the database. The database contains data in various formats, such as PDFs, text files, and images.

[0346] Learning of generative AI

[0347] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, the server extracts "product initial setup instructions" as text and converts it into a format that the AI ​​model can understand.

[0348] Receiving inquiries from users

[0349] The device provides an interface for receiving inquiries from users. For example, the device can receive inquiries from users such as "How do I connect my product via Bluetooth?" through a web form or chatbot.

[0350] Handling inquiries

[0351] The device sends the user's inquiry to the server. The server receives this and forwards it to a generative AI model. For example, the device receives the question, "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0352] Generative AI models generate appropriate answers to queries based on information stored in a database. For example, an AI model might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0353] Providing a response

[0354] The server sends the response received from the generative AI model to the terminal. The terminal then displays this to the user. For example, the server sends the received steps to the terminal, which then displays them in a chat window for the user to see. In this way, the user can quickly resolve the problem.

[0355] Specific example

[0356] Example 1: How to deal with a power failure

[0357] The server registers "solutions for when the power won't turn on" in its database. The server then uses this information to train a generative AI model.

[0358] When a user enters the question "What should I do if the device won't turn on?" into the chatbot, the device receives this and sends it to the server. The generative AI model generates the response "Try restarting by pressing and holding the power button for 10 seconds." The server sends this response back to the device, which then displays it to the user.

[0359] This system enables faster and more efficient responses compared to traditional customer support, reducing the burden on sales support centers and care centers, and is expected to improve customer satisfaction.

[0360] The following describes the processing flow.

[0361] Step 1:

[0362] The server registers product and service manual information in the customer support database. This includes initial setup instructions and troubleshooting guides for Hianabi. The server stores this information in the database in PDF and text file formats.

[0363] Step 2:

[0364] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. For example, the server extracts text from PDF files and converts it to a format that can be used by generative AI models for training.

[0365] Step 3:

[0366] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This process prepares the AI ​​model to understand text data and generate appropriate responses. For example, the server trains the AI ​​model on "initial setup procedures."

[0367] Step 4:

[0368] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect Hianabi via Bluetooth?"

[0369] Step 5:

[0370] The device receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the device might send the question "How do I connect Hianabi via Bluetooth?" to the server, which then passes this to the AI.

[0371] Step 6:

[0372] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0373] Step 7:

[0374] The server sends the response received from the generative AI model to the terminal. The server holds the generated response and then sends it to the terminal. This process is rapid, so the user can receive a response immediately.

[0375] Step 8:

[0376] The device displays the user the response received from the server. For example, the device might display instructions in the chat window such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0377] In this way, specific processes are carried out at each step, resulting in a system that responds quickly and accurately to user inquiries.

[0378] (Example 1)

[0379] 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."

[0380] Traditional customer support systems often suffered from delays in responding to user inquiries and a failure to provide appropriate information. Furthermore, some inquiries required manual handling by operators, leading to decreased efficiency. These problems could negatively impact the quality and speed of customer support, potentially harming customer satisfaction.

[0381] 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.

[0382] In this invention, the server includes means for acquiring information from a customer support database, means for preprocessing and formatting the acquired information to train a generative AI model, and means for receiving inquiries from users. This makes it possible to provide immediate and appropriate answers to inquiries from users.

[0383] A "customer support database" is a database that stores support information for users and manual information about products and services.

[0384] "Information" refers to data such as detailed instructions, troubleshooting guides, and FAQs about products and services.

[0385] "Preprocessing" refers to a series of processes that convert data into a format that generative AI models can understand. Specifically, this includes tasks such as extracting text from PDFs and format conversion.

[0386] "Format conversion" refers to the process of converting acquired information into a format that is easily understood by generative AI models.

[0387] A "generative AI model" is an artificial intelligence model that learns from information obtained from a customer support database and generates appropriate answers to user inquiries.

[0388] "Interface means" refers to input methods such as web forms and chatbots that users use to enter inquiries.

[0389] A "server" refers to a device that retrieves information from a customer support database, performs preprocessing and format conversion, and supplies the data to a generative AI model.

[0390] "Inquiry" refers to the act of a user inputting specific questions or problems about a product or service through an interface.

[0391] "Answer" refers to the information that a generative AI model generates based on a user's inquiry.

[0392] A "rapid response system" refers to a system that provides users with answers generated by generative AI models in real time.

[0393] This invention provides a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below from the perspectives of the server, terminal, and user.

[0394] Database Preparation

[0395] The server registers manual information about products and services in the customer support database. Specifically, it stores information such as initial product setup procedures and troubleshooting guides in various formats, including PDFs, text files, and images. For example, it uploads the setup instructions for a new smartphone to the database.

[0396] Information extraction and preprocessing

[0397] The server extracts necessary manual information from the customer support database. During this process, it performs data preprocessing and format conversion. Specifically, it extracts text from PDFs and converts it into a format that generative AI models can understand. For example, it extracts the text "Connect the device to the power source and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[0398] Training of generative AI models

[0399] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. Specifically, it inputs text data into the AI ​​model and repeats a feedback loop with the generated response. For example, by training the AI ​​model with a manual on "how to turn on the power," it can generate the response "Press and hold the power button."

[0400] Receiving inquiries from users

[0401] The device provides an interface for receiving inquiries from users. Specifically, it receives user questions using web forms or chatbots. For example, the device might receive a user inquiry such as, "Please tell me how to set up my new smartphone."

[0402] Processing queries and sending them to the server

[0403] The device forwards the user's inquiry to the server. Specifically, it converts questions entered into web forms or chatbots into the appropriate format and sends them to the server. For example, it forwards the question "How do I connect my smartphone via Bluetooth?" to the server.

[0404] Answer generation using generative AI models

[0405] The server passes the received user inquiry to a generative AI model. The generative AI model generates an answer to the question based on the information stored in the database. Specifically, in response to the question, "How do I connect my smartphone via Bluetooth?", it generates the answer, "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[0406] Providing a response

[0407] The server sends the response received from the generative AI model to the terminal, which then displays it to the user. Specifically, the server-generated response is displayed to the user through a chatbot or web form. For example, the terminal might provide the user with the response, "Open the settings menu, select Bluetooth, and choose the device you want to connect to," in a chat window.

[0408] Specific example

[0409] Example 1: How to deal with a power failure

[0410] 1. The server registers "Troubleshooting steps for when the power does not turn on" as a PDF in its database.

[0411] 2. The server extracts the text "Please try restarting by pressing and holding the power button for 10 seconds" from the registered PDF and preprocesses it.

[0412] 3. The server uses the preprocessed text to train a generative AI model.

[0413] 4. The user enters "What should I do if the device won't turn on?" into the device's chatbot.

[0414] 5. The terminal sends the user's inquiry to the server.

[0415] 6. The generative AI model generates the response, "Please try restarting by pressing and holding the power button for 10 seconds."

[0416] 7. The server sends the generated response to the terminal, which then displays it in the chat window and provides it to the user.

[0417] In this way, users can quickly solve problems. As a specific example, the following prompt can be input into the generating AI model.

[0418] Example of a prompt

[0419] "Please tell me how to connect the product via Bluetooth."

[0420] "Please tell me what to do if the power won't turn on."

[0421] This system is expected to enable faster and more efficient responses compared to traditional customer support, reducing the workload on operators and improving customer satisfaction.

[0422] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0423] Step 1:

[0424] The server registers manual information about products and services in the customer support database. Inputs include initial setup instructions and troubleshooting guides for new products, provided in PDF, text file, and image formats. The server stores this information in the database and generates an organized database as output. Specifically, it uploads new product setup instructions as PDFs to the database.

[0425] Step 2:

[0426] The server extracts necessary manual information from the customer support database. PDFs and text files stored in the database are used as input. The server processes the data and converts it into a format suitable for generative AI models. The output is text data that the AI ​​model can understand. Specifically, it extracts the text "Connect the device to the power supply and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[0427] Step 3:

[0428] The server supplies pre-processed manual information to a generative AI model for training. Formatted text data is used as input. The server inputs this into the AI ​​model and improves the model's performance through a feedback loop. The output is a fully trained AI model. As a specific example, the AI ​​model is trained using text data about "how to turn on the power."

[0429] Step 4:

[0430] Users utilize an interface to submit inquiries from their devices. As input, the user's question is entered into the device via a web form or chatbot. The device receives this input and sends it to the server. As output, the inquiry content is sent to the server. For example, a user might enter the inquiry "Please tell me the initial setup procedure for my new smartphone" into the chatbot.

[0431] Step 5:

[0432] The terminal converts the received inquiry into the appropriate format and sends it to the server. The user's inquiry is provided to the terminal as input. The terminal converts this into the appropriate format and sends it to the server. The formatted inquiry is sent to the server as output. As a specific example, the question "How do I connect my smartphone via Bluetooth?" is converted into the appropriate format and sent to the server.

[0433] Step 6:

[0434] The server passes the user's inquiry received to a generative AI model. The server receives the inquiry content sent from the terminal as input. The server inputs this into the AI ​​model and generates an answer. The generated answer is obtained as output. For example, in response to the question "How do I connect my smartphone via Bluetooth?", it generates the answer "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[0435] Step 7:

[0436] The server sends the response received from the generative AI model to the terminal. The server receives the response generated by the AI ​​model as input. The server sends this to the terminal, which displays it to the user. The output is the response displayed to the user. Specifically, the server sends the generated response to the terminal via a chatbot or web form, and the terminal provides the user with the response "Open the settings menu, select Bluetooth, and choose the device you want to connect to" in the chat window.

[0437] (Application Example 1)

[0438] 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."

[0439] Current customer support systems struggle to respond quickly to user inquiries. This is especially true for e-commerce sites, where detailed product inquiries are frequent, requiring efficient and immediate responses. Furthermore, many users prefer to communicate via voice as well as text. However, few systems successfully balance the requirements for providing appropriate answers with user convenience, resulting in decreased user satisfaction and operational inefficiencies. These problems need to be addressed.

[0440] 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.

[0441] In this invention, the server includes means for obtaining manual information from a customer support database, means for training a generative AI model with the obtained manual information, means for a user interface for receiving inquiries from users, means for users to input questions about products in text or voice, means for sending received inquiries to the generative AI model, and means for receiving and providing answers generated by the generative AI model to the user. This makes it possible to respond quickly to user inquiries and improve user convenience.

[0442] A "customer support database" is a data storage system that stores information about products and services in order to respond to inquiries from users.

[0443] "Manual information" refers to documents and data that provide detailed information on how to use a product or service, setup procedures, troubleshooting, and more.

[0444] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers from given data.

[0445] A "user interface" is a software or hardware configuration that includes interactive elements that allow a user to directly interact with the system.

[0446] "Text or voice input" refers to the process by which users submit inquiries through text input or voice input.

[0447] A "rapid response system" is a mechanism that provides users with answers generated by generative AI models in real time.

[0448] This invention is a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. A specific embodiment of this system is described below.

[0449] System program

[0450] Database Preparation

[0451] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. The server also stores initial setup instructions for new products in the database.

[0452] Learning of generative AI

[0453] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, "product initial setup instructions" are extracted as text and converted into a format that the AI ​​model can understand.

[0454] User Interface

[0455] The user interface is an interface for users to enter questions about the product. This interface has the functionality to accept text input and voice input. It allows users to enter inquiries such as, "How do I connect the product via Bluetooth?"

[0456] Handling inquiries

[0457] The server receives the user's inquiry and forwards it to a generative AI model. The generative AI model generates an appropriate response based on the information stored in the database. For example, it might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0458] Providing a response

[0459] The server sends the response received from the generative AI model to the user interface. The user interface provides the response to the user in text or voice.

[0460] Hardware and software to be used

[0461] This system uses the following hardware and software.

[0462] Hardware:

[0463] Smartphone (the device on which the user uses the app)

[0464] Servers (data centers where databases and AI models run)

[0465] software:

[0466] OpenAI API (utilizing generative AI models)

[0467] A temporary database module (a module that communicates with the database)

[0468] A temporary chatbot interface (a module that interacts with the user)

[0469] Examples of specific cases and prompt statements

[0470] Specific example

[0471] A user asks, "I'd like to know more about the shipping method for this product." The smartphone app sends the question to the server and retrieves the "shipping method" information from the database. The AI ​​generates a response saying, "Regarding shipping methods, it is usually shipped by courier service, but if you are in a hurry, an express service is available for an additional fee." The app then displays this response to the user.

[0472] Example of a prompt

[0473] The following are examples of prompts to input into the generated AI model.

[0474] User question: I'd like to know more about the shipping methods for this product.

[0475] Search results from our customer support database: Our standard shipping method is courier service, but for urgent orders, an express delivery service is available for an additional fee.

[0476] Answer the user's query based on the above information.

[0477] In this way, this invention makes it possible to realize a system that responds quickly and accurately to user inquiries.

[0478] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0479] Step 1: Prepare the database

[0480] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in the database. The input here is the manual information, and the output is the information registered in the database.

[0481] Step 2: Training a generative AI

[0482] The server extracts the necessary manual information from the database and uses it to train a generative AI model. Data preprocessing and format conversion are performed to transform the information into a format that the AI ​​model can understand. For example, the server extracts "product initial setup procedures" in text format and feeds them into the generative AI model. Here, the input is the manual information extracted from the database, and the output is the completed generative AI model.

[0483] Step 3: User Interface Operation

[0484] The user enters questions about the product through the terminal's user interface. This interface accepts both text and voice input. Specifically, the user might enter the question, "How do I connect the product via Bluetooth?" The input here is the user's question, and the output is the content of that question.

[0485] Step 4: Submitting the inquiry

[0486] The terminal sends the user's query to the server. The server forwards the received query to a generative AI model. At this stage, the terminal sends the received query to the server, and the server passes it to the AI. The input here is the user's query, and the output is the prompt message passed to the AI ​​model.

[0487] Step 5: AI-powered response generation

[0488] Generative AI models generate appropriate answers based on information stored in a database. For example, an AI model might generate the following procedure: "First, select Bluetooth from the settings menu, then search for the device you want to connect to." The input here is the prompt received from the server, and the output is the generated answer.

[0489] Step 6: Provide your response

[0490] The server sends the response received from the generative AI model to the terminal. The terminal provides the response to the user in text or voice. For example, the server sends the received procedure to the terminal, and the terminal displays it to the user. Here, the input is the response received from the AI ​​model, and the output is the response provided to the user through the user interface.

[0491] 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.

[0492] This invention provides a system that responds quickly and accurately to user inquiries by acquiring manual information from a customer support database and training a generative AI model with it. Furthermore, it is a system that combines this with an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response based on those emotions.

[0493] Specific examples of program processing

[0494] Database Preparation

[0495] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions and troubleshooting guides for new products in the database. The database includes PDFs, text files, images, and other formats.

[0496] Learning of generative AI

[0497] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. The server then supplies the preprocessed text data to a generative AI model for training. For example, the server extracts "initial setup procedure" as text and uses it to train the AI ​​model.

[0498] Receiving inquiries from users

[0499] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[0500] Handling inquiries

[0501] The terminal receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the terminal receives the question "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0502] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0503] How the emotion recognition engine works

[0504] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. The emotion recognition engine detects the user's emotional state from the context of the input, voice, facial expressions, etc. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[0505] The server can adjust the content and expression of responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the user is irritated, the server will change the response to a more polite expression.

[0506] Escalation engine operation

[0507] The server activates the escalation engine if the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user's emotion is recognized as "very angry," the escalation engine will contact the support team leader.

[0508] Providing a response

[0509] The server sends the responses generated by the generative AI model and emotion recognition engine to the device. The device then displays this to the user. For example, the server sends a detailed instruction to the device, such as "Select Bluetooth from the settings menu, then search for the device you want to connect to," which the device then displays in the chat window for the user. The user then uses this information to configure the product.

[0510] This system recognizes user emotions and provides quick and appropriate responses, enabling more efficient support compared to traditional customer support. The introduction of an emotion recognition engine can further improve user satisfaction.

[0511] The following describes the processing flow.

[0512] Step 1:

[0513] The server registers manual information about products and services in the customer support database. This includes tasks such as storing initial product setup instructions and troubleshooting guides in PDF or text file format in the database.

[0514] Step 2:

[0515] The server extracts the necessary manual information from the database and preprocesses that information. Preprocessing includes cleaning and formatting text data. For example, the server extracts text from PDF files and converts it into a format that generative AI models can understand.

[0516] Step 3:

[0517] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This enables the AI ​​to generate appropriate answers to user inquiries. For example, the server trains the AI ​​model on "initial setup procedures."

[0518] Step 4:

[0519] The device provides an interface for receiving inquiries from users. Users enter their problems or questions through web forms or chatbots. For example, a user might type, "Please tell me how to connect the product via Bluetooth."

[0520] Step 5:

[0521] The terminal receives the user's inquiry and then sends that information to the server. The server then receives this inquiry and sends it to a generative AI model. For example, the terminal receives the inquiry "Please tell me how to connect the product via Bluetooth" and sends it to the server, which then passes it on to the AI ​​model.

[0522] Step 6:

[0523] Generative AI models analyze the content of inquiries and generate the most suitable answers based on information stored in a database. For example, they might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0524] Step 7:

[0525] When a terminal receives a user inquiry, it sends the input content to an emotion recognition engine. The emotion recognition engine analyzes the context of the input text, as well as voice and facial expressions, to detect the user's emotional state. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[0526] Step 8:

[0527] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the server detects that the user is angry, it will change the response to a more polite and calm expression.

[0528] Step 9:

[0529] The server activates the escalation engine if it determines that the user's emotional state is "urgent." The escalation engine follows a predefined procedure to escalate the issue to a higher-level support person. For example, if it is determined that the user is "very angry," the support team leader will be contacted.

[0530] Step 10:

[0531] The device displays responses generated by a generative AI model and an emotion recognition engine to the user. For example, it might display detailed instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0532] Through the steps described above, a system incorporating an emotion recognition engine can provide prompt and appropriate support that takes user emotions into account. This makes it possible to improve the efficiency of customer support and increase user satisfaction.

[0533] (Example 2)

[0534] 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".

[0535] Conventional customer support systems often suffer from delays in responding to user inquiries or fail to provide appropriate answers. Furthermore, they may fail to consider user emotions, leading to dissatisfaction. This results in decreased user satisfaction and customer churn. The present invention aims to solve these problems by providing a system that responds quickly and accurately to user inquiries, and further recognizes and appropriately addresses user emotions.

[0536] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0537] In this invention, the server includes means for acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, and means for an emotion recognition engine for analyzing the user's emotions. This enables a quick and appropriate response to user inquiries, and further, the provision of appropriate answers that take the user's emotions into consideration.

[0538] A "customer support database" is a database that stores various manuals and troubleshooting guides related to products and services, and is used to handle customer inquiries.

[0539] "Manual information" refers to information that includes detailed instructions and guides on how to use a product or service, setup procedures, troubleshooting, and more.

[0540] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate appropriate responses from input data.

[0541] A "user interface" is the interface through which a user makes inquiries to a system, and includes web forms, chatbots, and other similar elements.

[0542] An "emotion recognition engine" is an engine that analyzes and recognizes emotions from the user's input text, voice, and facial expressions.

[0543] "Methods for escalating the issue" refer to ways of transferring the problem to a higher-level person in charge, depending on the user's emotional state.

[0544] A "rapid response system for real-time delivery" is a system that provides generated answers to users immediately, with the aim of responding quickly to inquiries.

[0545] This invention provides a system that responds quickly and appropriately to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. Furthermore, this system incorporates an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response accordingly.

[0546] Database Preparation

[0547] The server registers manual information about products and services in the customer support database. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in PDF, text file, and image formats in the database. For example, the server saves "Printer Ink Replacement Procedure" as a PDF file in the database.

[0548] Learning of generative AI

[0549] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning the text data (removing unnecessary spaces and line breaks) and format conversion (from PDF to text). The server then feeds this preprocessed text to a generative AI model. By training the generative AI model (e.g., GPT-3) with this data, the AI ​​learns how to configure and troubleshoot the product.

[0550] Receiving inquiries from users

[0551] The device provides an interface for receiving inquiries from users. Users enter their questions using web forms or chatbots. For example, a user might type "How do I connect the product via Bluetooth?" into the chatbot.

[0552] Handling inquiries

[0553] The terminal receives a user inquiry and sends its contents to the server. The server analyzes the inquiry and sends it to a generative AI model. For example, the terminal sends the question "How do I connect the product via Bluetooth?" to the server, which then passes this to the AI. The generative AI model analyzes the inquiry and generates an appropriate answer based on the information stored in the database. For example, the generative AI model might generate the procedure "Select Bluetooth from the settings menu, and then search for the device you want to connect to."

[0554] How the emotion recognition engine works

[0555] The device uses an emotion recognition engine to analyze the user's emotions when they make an inquiry. The emotion recognition engine detects the user's emotional state from the context of the input, voice, and facial expressions. For example, if a user sends a message that includes frustration, such as "Tell me quickly!", the emotion recognition engine will analyze this as "frustration." Based on the user's emotional state analyzed by the emotion recognition engine, the server adjusts the content and expression of the response generated by the generative AI model. For example, if the user is frustrated, the server adjusts the response to provide a more polite and prompt response.

[0556] Escalation engine operation

[0557] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user is identified as "very angry," the escalation engine will contact the support team leader and have them handle the issue directly.

[0558] Providing a response

[0559] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The device displays this to the user. The generated response, "Select Bluetooth from the settings menu and then search for the device you want to connect to," is politely rephrased according to the emotion to, "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to." This response is displayed in the device's chat window and provided to the user. The user uses this information to configure the product settings.

[0560] Examples of specific cases and prompt statements

[0561] Specific example

[0562] 1. User's question: "My printer is jammed. How can I fix it?"

[0563] 2. Server preprocessing: Extract troubleshooting guides for "paper jams" from the database and use them to train a generative AI model.

[0564] 3. Emotion Analysis: The emotion recognition engine detects "frustration" from the user's voice and facial expressions.

[0565] 4. Response generation and refinement: The generative AI model creates the procedure "turn off the power and carefully remove the paper," and the server refines the expression more carefully based on the emotion recognition results.

[0566] 5. Response Provision: The adjusted response is provided to the user by the device (for example, displayed in the chat window).

[0567] Example of a prompt

[0568] User's question: "My printer is jammed. How can I fix it?"

[0569] Sentiment Analysis: Detects irritation from input context, voice, and facial expressions.

[0570] Generative AI model's response: "Turn off the power and carefully remove the paper."

[0571] Emotion-based adjustment: "I'm sorry, but could you please turn off the printer and carefully remove the jammed paper?"

[0572] In this way, the system can respond to user inquiries efficiently and with consideration for their feelings.

[0573] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0574] Step 1:

[0575] The server registers manual information about products and services in the customer support database. Input includes manual information in PDF, text file, and image formats. The specific process for storing this information in the database involves the server receiving the manual information in various formats and saving it in the database in the appropriate format. For example, a PDF file titled "Printer Ink Replacement Procedure" might be saved in the database.

[0576] Step 2:

[0577] The server extracts the necessary manual information from the database and preprocesses it. The input is the manual information in the database. Specifically, the server cleans the text data (removing unnecessary spaces and line breaks) and converts the format (from PDF to text). The output is the preprocessed text data. For example, a PDF document titled "Initial Setup Procedure" is converted to text format.

[0578] Step 3:

[0579] The server supplies pre-processed text data to a generative AI model. The input is pre-processed text data. Specifically, the server feeds this data to a generative AI model (e.g., GPT-3) to train the model. The output is the trained generative AI model.

[0580] Step 4:

[0581] The terminal provides an interface for receiving inquiries from users. Input includes questions entered by the user via web forms or chatbots. Its specific operation is to retrieve the user's inquiry content through the user interface and send it to the server. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[0582] Step 5:

[0583] The server receives the user's inquiry and sends it to a generative AI model. The input is the inquiry sent from the terminal. Specifically, the server analyzes the inquiry and passes it to the generative AI model. The output is the answer generated by the generative AI model. For example, it might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0584] Step 6:

[0585] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. Input includes the user's written text, voice, and facial expressions. The emotion recognition engine analyzes this information to detect the user's emotional state. The output is the analyzed emotional state of the user (e.g., "irritated").

[0586] Step 7:

[0587] The server adjusts the response of the generative AI model based on the results analyzed by the emotion recognition engine. The inputs are the generated response and the emotion state from the emotion recognition engine. Specifically, the server adjusts the response generated by the generative AI model according to the emotion state, changing its expression. The output is the adjusted response. For example, the response might be changed to "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to."

[0588] Step 8:

[0589] The server considers the user's emotional state and escalates the issue to a higher-level person in charge. The input is the emotional state, which the emotion recognition engine has determined to be "urgent." The specific action is for the server to activate the escalation engine and transfer the issue to a higher-level support person. The output is a notification to the higher-level person in charge that the issue has been escalated.

[0590] Step 9:

[0591] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The input is a refined response. Specifically, the server sends this response to the device, and the device displays it to the user. The output is the response displayed to the user. For example, the device's chat window might display helpful instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0592] (Application Example 2)

[0593] 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."

[0594] Traditional customer support systems often resulted in low user satisfaction because they provided uniform responses without considering the user's emotional state. Furthermore, their limited functionality for providing timely and appropriate answers to inquiries made efficient support difficult. Additionally, the lack of real-time support utilizing in-store voice interfaces created a need to improve the user experience.

[0595] 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 acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, emotion recognition engine means for analyzing the user's emotional state, means for adjusting the content and expression of the response based on the emotional state, escalation engine means for escalating the problem to a higher-level support staff member based on the user's emotional state, and voice interface means for receiving user inquiries by voice using speech recognition and providing responses by voice. This enables improved user satisfaction, the provision of quick and appropriate answers, and effective customer support through a real-time voice interface.

[0596] A "customer support database" is a database that stores information such as manuals, troubleshooting guides, and FAQs related to products and services.

[0597] "Manual information" refers to information that includes detailed explanations of how to use a product or service, setup procedures, troubleshooting, and more.

[0598] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to automatically generate appropriate answers to user inquiries.

[0599] A "user interface" is an interface through which a user inputs inquiries and interacts with the system.

[0600] An "emotion recognition engine" is an engine that analyzes the user's emotional state from their input or voice to detect emotions such as "anger" or "joy."

[0601] The "escalation engine" is an engine that, based on the user's emotional state analyzed by the emotion recognition engine, escalates the issue to a higher-level support representative as needed.

[0602] A "rapid response system" is a system that provides users with answers generated by generative AI models in real time.

[0603] A "voice interface" is an interface that uses speech recognition technology to receive user inquiries in voice and provide responses in voice.

[0604] This invention is a system for streamlining customer support in physical stores. It retrieves manual information from a customer support database and uses it to train a generative AI model, enabling it to respond quickly and accurately to user inquiries. It also incorporates an emotion recognition engine that analyzes the user's emotional state at the time of inquiry and adjusts the content and expression of the response accordingly.

[0605] The server retrieves product and service manual information (e.g., initial setup procedures and troubleshooting guides) from the customer support database. This manual information is stored in various formats, including PDFs, text files, and images. The server preprocesses this data and trains a generative AI model. This process involves cleaning and formatting the text data. For example, it might extract the initial setup procedures for a new product as text and train the generative AI model.

[0606] The terminal provides an interface for receiving user inquiries. This is done via applications installed on smartphones or robots placed within the store. Users can enter questions through web forms, chatbots, or voice interfaces. For example, if a user asks, "Do you have this product in stock?", the terminal sends this inquiry to the server.

[0607] The server sends the received inquiry to a generative AI model, which analyzes the inquiry and generates an appropriate response. For example, it might generate a response such as, "We will check the stock status. Please wait a moment." The server also uses an emotion recognition engine to analyze the user's emotions. It detects the user's emotional state (e.g., "anger") from the context of the input, voice, and facial expressions, and adjusts the content and expression of the response generated by the generative AI model based on this. If the user is angry, the response is changed to a more polite expression.

[0608] Furthermore, if the emotion recognition engine determines that the user's emotional state is "urgent," the server activates the escalation engine to escalate the issue to a higher-level support person. For example, if the user's emotions are recognized as "very angry," the server will contact the support team leader.

[0609] In summary, this invention enables improved user satisfaction, the provision of quick and appropriate responses, and effective customer support through a real-time voice interface.

[0610] Specific example

[0611] When a user asks, "Do you have this item in stock?", the speech recognition engine converts the speech into text, and the generative AI model generates the response, "We will check the stock status. Please wait a moment."

[0612] Examples of prompts to input into a generative AI model:

[0613] User: Do you have this item in stock?

[0614] assistant:

[0615] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0616] Step 1: Prepare the database

[0617] The server registers product and service manual information in the customer support database. Inputs include PDFs, text files, and images, such as initial setup instructions and troubleshooting guides for new products. The server stores this information in the database. The output is that the stored information becomes available for use.

[0618] Step 2: Training a generative AI

[0619] The server extracts necessary manual information from the database and performs text data cleaning and format conversion. The input is the manual information obtained from the database. The server supplies the pre-processed text data to a generative AI model for training. The output is that the generative AI model can generate appropriate answers to queries.

[0620] Step 3: Receiving inquiries from users

[0621] The terminal provides an interface for receiving inquiries from users. Inputs include questions entered by the user via web forms, chatbots, or voice interfaces. The terminal receives these inquiries and generates data to send to the server as output.

[0622] Step 4: Processing the inquiry

[0623] The server sends the query received from the terminal to a generative AI model. The input is the user's question. The server processes the data necessary for the generative AI model to analyze the query and generate an appropriate answer. The server receives the generated answer as output.

[0624] Step 5: Operation of the emotion recognition engine

[0625] The device uses an emotion recognition engine to analyze the user's emotions when a user makes a request. Input includes the user's context, voice, and facial expressions. The device analyzes this information and sends the user's emotional state (e.g., "anger," "frustration") to the server as output.

[0626] Step 6: Adjusting the content and wording of your answer

[0627] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. The input consists of the user's emotional state and the generated responses. The server adjusts these and outputs the adjusted responses.

[0628] Step 7: Operation of the escalation engine

[0629] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The input is the analyzed user's emotional state. The server escalates the issue to a higher-level support person and generates an escalated issue as output.

[0630] Step 8: Provide your answer

[0631] The server sends the responses generated by the generative AI model and emotion recognition engine to the terminal. The input is the adjusted response. The terminal displays this to the user, and the output is the information provided to the user.

[0632] As a concrete example, if a user asks, "Do you have this item in stock?", the device uses a speech recognition engine to convert the speech into text and sends it to the server. The server uses a generative AI model to generate an answer and an emotion recognition engine to analyze the user's emotional state. After appropriate adjustments are made, the device provides the user with an answer such as, "I will check the stock status. Please wait a moment."

[0633] Examples of prompts to input into a generative AI model:

[0634] User: Do you have this item in stock?

[0635] assistant:

[0636] 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.

[0637] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">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.

[0638] 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.

[0639] [Third Embodiment]

[0640] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0641] 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.

[0642] 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).

[0643] 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.

[0644] 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.

[0645] 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).

[0646] 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.

[0647] 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.

[0648] 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.

[0649] 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.

[0650] 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.

[0651] 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".

[0652] This invention provides a system that quickly responds to user inquiries by retrieving manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below.

[0653] Specific examples of program processing

[0654] Database Preparation

[0655] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions for new products in the database. The database contains data in various formats, such as PDFs, text files, and images.

[0656] Learning of generative AI

[0657] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, the server extracts "product initial setup instructions" as text and converts it into a format that the AI ​​model can understand.

[0658] Receiving inquiries from users

[0659] The device provides an interface for receiving inquiries from users. For example, the device can receive inquiries from users such as "How do I connect my product via Bluetooth?" through a web form or chatbot.

[0660] Handling inquiries

[0661] The device sends the user's inquiry to the server. The server receives this and forwards it to a generative AI model. For example, the device receives the question, "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0662] Generative AI models generate appropriate answers to queries based on information stored in a database. For example, an AI model might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0663] Providing a response

[0664] The server sends the response received from the generative AI model to the terminal. The terminal then displays this to the user. For example, the server sends the received steps to the terminal, which then displays them in a chat window for the user to see. In this way, the user can quickly resolve the problem.

[0665] Specific example

[0666] Example 1: How to deal with a power failure

[0667] The server registers "solutions for when the power won't turn on" in its database. The server then uses this information to train a generative AI model.

[0668] When a user enters the question "What should I do if the device won't turn on?" into the chatbot, the device receives this and sends it to the server. The generative AI model generates the response "Try restarting by pressing and holding the power button for 10 seconds." The server sends this response back to the device, which then displays it to the user.

[0669] This system enables faster and more efficient responses compared to traditional customer support, reducing the burden on sales support centers and care centers, and is expected to improve customer satisfaction.

[0670] The following describes the processing flow.

[0671] Step 1:

[0672] The server registers product and service manual information in the customer support database. This includes initial setup instructions and troubleshooting guides for Hianabi. The server stores this information in the database in PDF and text file formats.

[0673] Step 2:

[0674] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. For example, the server extracts text from PDF files and converts it to a format that can be used by generative AI models for training.

[0675] Step 3:

[0676] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This process prepares the AI ​​model to understand text data and generate appropriate responses. For example, the server trains the AI ​​model on "initial setup procedures."

[0677] Step 4:

[0678] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect Hianabi via Bluetooth?"

[0679] Step 5:

[0680] The device receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the device might send the question "How do I connect Hianabi via Bluetooth?" to the server, which then passes this to the AI.

[0681] Step 6:

[0682] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0683] Step 7:

[0684] The server sends the response received from the generative AI model to the terminal. The server holds the generated response and then sends it to the terminal. This process is rapid, so the user can receive a response immediately.

[0685] Step 8:

[0686] The device displays the user the response received from the server. For example, the device might display instructions in the chat window such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0687] In this way, specific processes are carried out at each step, resulting in a system that responds quickly and accurately to user inquiries.

[0688] (Example 1)

[0689] 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."

[0690] Traditional customer support systems often suffered from delays in responding to user inquiries and a failure to provide appropriate information. Furthermore, some inquiries required manual handling by operators, leading to decreased efficiency. These problems could negatively impact the quality and speed of customer support, potentially harming customer satisfaction.

[0691] 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.

[0692] In this invention, the server includes means for acquiring information from a customer support database, means for preprocessing and formatting the acquired information to train a generative AI model, and means for receiving inquiries from users. This makes it possible to provide immediate and appropriate answers to inquiries from users.

[0693] A "customer support database" is a database that stores support information for users and manual information about products and services.

[0694] "Information" refers to data such as detailed instructions, troubleshooting guides, and FAQs about products and services.

[0695] "Preprocessing" refers to a series of processes that convert data into a format that generative AI models can understand. Specifically, this includes tasks such as extracting text from PDFs and format conversion.

[0696] "Format conversion" refers to the process of converting acquired information into a format that is easily understood by generative AI models.

[0697] A "generative AI model" is an artificial intelligence model that learns from information obtained from a customer support database and generates appropriate answers to user inquiries.

[0698] "Interface means" refers to input methods such as web forms and chatbots that users use to enter inquiries.

[0699] A "server" refers to a device that retrieves information from a customer support database, performs preprocessing and format conversion, and supplies the data to a generative AI model.

[0700] "Inquiry" refers to the act of a user inputting specific questions or problems about a product or service through an interface.

[0701] "Answer" refers to the information that a generative AI model generates based on a user's inquiry.

[0702] A "rapid response system" refers to a system that provides users with answers generated by generative AI models in real time.

[0703] This invention provides a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below from the perspectives of the server, terminal, and user.

[0704] Database Preparation

[0705] The server registers manual information about products and services in the customer support database. Specifically, it stores information such as initial product setup procedures and troubleshooting guides in various formats, including PDFs, text files, and images. For example, it uploads the setup instructions for a new smartphone to the database.

[0706] Information extraction and preprocessing

[0707] The server extracts necessary manual information from the customer support database. During this process, it performs data preprocessing and format conversion. Specifically, it extracts text from PDFs and converts it into a format that generative AI models can understand. For example, it extracts the text "Connect the device to the power source and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[0708] Training of generative AI models

[0709] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. Specifically, it inputs text data into the AI ​​model and repeats a feedback loop with the generated response. For example, by training the AI ​​model with a manual on "how to turn on the power," it can generate the response "Press and hold the power button."

[0710] Receiving inquiries from users

[0711] The device provides an interface for receiving inquiries from users. Specifically, it receives user questions using web forms or chatbots. For example, the device might receive a user inquiry such as, "Please tell me how to set up my new smartphone."

[0712] Processing queries and sending them to the server

[0713] The device forwards the user's inquiry to the server. Specifically, it converts questions entered into web forms or chatbots into the appropriate format and sends them to the server. For example, it forwards the question "How do I connect my smartphone via Bluetooth?" to the server.

[0714] Answer generation using generative AI models

[0715] The server passes the received user inquiry to a generative AI model. The generative AI model generates an answer to the question based on the information stored in the database. Specifically, in response to the question, "How do I connect my smartphone via Bluetooth?", it generates the answer, "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[0716] Providing a response

[0717] The server sends the response received from the generative AI model to the terminal, which then displays it to the user. Specifically, the server-generated response is displayed to the user through a chatbot or web form. For example, the terminal might provide the user with the response, "Open the settings menu, select Bluetooth, and choose the device you want to connect to," in a chat window.

[0718] Specific example

[0719] Example 1: How to deal with a power failure

[0720] 1. The server registers "Troubleshooting steps for when the power does not turn on" as a PDF in its database.

[0721] 2. The server extracts the text "Please try restarting by pressing and holding the power button for 10 seconds" from the registered PDF and preprocesses it.

[0722] 3. The server uses the preprocessed text to train a generative AI model.

[0723] 4. The user enters "What should I do if the device won't turn on?" into the device's chatbot.

[0724] 5. The terminal sends the user's inquiry to the server.

[0725] 6. The generative AI model generates the response, "Please try restarting by pressing and holding the power button for 10 seconds."

[0726] 7. The server sends the generated response to the terminal, which then displays it in the chat window and provides it to the user.

[0727] In this way, users can quickly solve problems. As a specific example, the following prompt can be input into the generating AI model.

[0728] Example of a prompt

[0729] "Please tell me how to connect the product via Bluetooth."

[0730] "Please tell me what to do if the power won't turn on."

[0731] This system is expected to enable faster and more efficient responses compared to traditional customer support, reducing the workload on operators and improving customer satisfaction.

[0732] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0733] Step 1:

[0734] The server registers manual information about products and services in the customer support database. Inputs include initial setup instructions and troubleshooting guides for new products, provided in PDF, text file, and image formats. The server stores this information in the database and generates an organized database as output. Specifically, it uploads new product setup instructions as PDFs to the database.

[0735] Step 2:

[0736] The server extracts necessary manual information from the customer support database. PDFs and text files stored in the database are used as input. The server processes the data and converts it into a format suitable for generative AI models. The output is text data that the AI ​​model can understand. Specifically, it extracts the text "Connect the device to the power supply and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[0737] Step 3:

[0738] The server supplies pre-processed manual information to a generative AI model for training. Formatted text data is used as input. The server inputs this into the AI ​​model and improves the model's performance through a feedback loop. The output is a fully trained AI model. As a specific example, the AI ​​model is trained using text data about "how to turn on the power."

[0739] Step 4:

[0740] Users utilize an interface to submit inquiries from their devices. As input, the user's question is entered into the device via a web form or chatbot. The device receives this input and sends it to the server. As output, the inquiry content is sent to the server. For example, a user might enter the inquiry "Please tell me the initial setup procedure for my new smartphone" into the chatbot.

[0741] Step 5:

[0742] The terminal converts the received inquiry into the appropriate format and sends it to the server. The user's inquiry is provided to the terminal as input. The terminal converts this into the appropriate format and sends it to the server. The formatted inquiry is sent to the server as output. As a specific example, the question "How do I connect my smartphone via Bluetooth?" is converted into the appropriate format and sent to the server.

[0743] Step 6:

[0744] The server passes the user's inquiry received to a generative AI model. The server receives the inquiry content sent from the terminal as input. The server inputs this into the AI ​​model and generates an answer. The generated answer is obtained as output. For example, in response to the question "How do I connect my smartphone via Bluetooth?", it generates the answer "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[0745] Step 7:

[0746] The server sends the response received from the generative AI model to the terminal. The server receives the response generated by the AI ​​model as input. The server sends this to the terminal, which displays it to the user. The output is the response displayed to the user. Specifically, the server sends the generated response to the terminal via a chatbot or web form, and the terminal provides the user with the response "Open the settings menu, select Bluetooth, and choose the device you want to connect to" in the chat window.

[0747] (Application Example 1)

[0748] 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."

[0749] Current customer support systems struggle to respond quickly to user inquiries. This is especially true for e-commerce sites, where detailed product inquiries are frequent, requiring efficient and immediate responses. Furthermore, many users prefer to communicate via voice as well as text. However, few systems successfully balance the requirements for providing appropriate answers with user convenience, resulting in decreased user satisfaction and operational inefficiencies. These problems need to be addressed.

[0750] 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.

[0751] In this invention, the server includes means for obtaining manual information from a customer support database, means for training a generative AI model with the obtained manual information, means for a user interface for receiving inquiries from users, means for users to input questions about products in text or voice, means for sending received inquiries to the generative AI model, and means for receiving and providing answers generated by the generative AI model to the user. This makes it possible to respond quickly to user inquiries and improve user convenience.

[0752] A "customer support database" is a data storage system that stores information about products and services in order to respond to inquiries from users.

[0753] "Manual information" refers to documents and data that provide detailed information on how to use a product or service, setup procedures, troubleshooting, and more.

[0754] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers from given data.

[0755] A "user interface" is a software or hardware configuration that includes interactive elements that allow a user to directly interact with the system.

[0756] "Text or voice input" refers to the process by which users submit inquiries through text input or voice input.

[0757] A "rapid response system" is a mechanism that provides users with answers generated by generative AI models in real time.

[0758] This invention is a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. A specific embodiment of this system is described below.

[0759] System program

[0760] Database Preparation

[0761] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. The server also stores initial setup instructions for new products in the database.

[0762] Learning of generative AI

[0763] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, "product initial setup instructions" are extracted as text and converted into a format that the AI ​​model can understand.

[0764] User Interface

[0765] The user interface is an interface for users to enter questions about the product. This interface has the functionality to accept text input and voice input. It allows users to enter inquiries such as, "How do I connect the product via Bluetooth?"

[0766] Handling inquiries

[0767] The server receives the user's inquiry and forwards it to a generative AI model. The generative AI model generates an appropriate response based on the information stored in the database. For example, it might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0768] Providing a response

[0769] The server sends the response received from the generative AI model to the user interface. The user interface provides the response to the user in text or voice.

[0770] Hardware and software to be used

[0771] This system uses the following hardware and software.

[0772] Hardware:

[0773] Smartphone (the device on which the user uses the app)

[0774] Servers (data centers where databases and AI models run)

[0775] software:

[0776] OpenAI API (utilizing generative AI models)

[0777] A temporary database module (a module that communicates with the database)

[0778] A temporary chatbot interface (a module that interacts with the user)

[0779] Examples of specific cases and prompt statements

[0780] Specific example

[0781] A user asks, "I'd like to know more about the shipping method for this product." The smartphone app sends the question to the server and retrieves the "shipping method" information from the database. The AI ​​generates a response saying, "Regarding shipping methods, it is usually shipped by courier service, but if you are in a hurry, an express service is available for an additional fee." The app then displays this response to the user.

[0782] Example of a prompt

[0783] The following are examples of prompts to input into the generated AI model.

[0784] User question: I'd like to know more about the shipping methods for this product.

[0785] Search results from our customer support database: Our standard shipping method is courier service, but for urgent orders, an express delivery service is available for an additional fee.

[0786] Answer the user's query based on the above information.

[0787] In this way, this invention makes it possible to realize a system that responds quickly and accurately to user inquiries.

[0788] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0789] Step 1: Prepare the database

[0790] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in the database. The input here is the manual information, and the output is the information registered in the database.

[0791] Step 2: Training a generative AI

[0792] The server extracts the necessary manual information from the database and uses it to train a generative AI model. Data preprocessing and format conversion are performed to transform the information into a format that the AI ​​model can understand. For example, the server extracts "product initial setup procedures" in text format and feeds them into the generative AI model. Here, the input is the manual information extracted from the database, and the output is the completed generative AI model.

[0793] Step 3: User Interface Operation

[0794] The user enters questions about the product through the terminal's user interface. This interface accepts both text and voice input. Specifically, the user might enter the question, "How do I connect the product via Bluetooth?" The input here is the user's question, and the output is the content of that question.

[0795] Step 4: Submitting the inquiry

[0796] The terminal sends the user's query to the server. The server forwards the received query to a generative AI model. At this stage, the terminal sends the received query to the server, and the server passes it to the AI. The input here is the user's query, and the output is the prompt message passed to the AI ​​model.

[0797] Step 5: AI-powered response generation

[0798] Generative AI models generate appropriate answers based on information stored in a database. For example, an AI model might generate the following procedure: "First, select Bluetooth from the settings menu, then search for the device you want to connect to." The input here is the prompt received from the server, and the output is the generated answer.

[0799] Step 6: Provide your response

[0800] The server sends the response received from the generative AI model to the terminal. The terminal provides the response to the user in text or voice. For example, the server sends the received procedure to the terminal, and the terminal displays it to the user. Here, the input is the response received from the AI ​​model, and the output is the response provided to the user through the user interface.

[0801] 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.

[0802] This invention provides a system that responds quickly and accurately to user inquiries by acquiring manual information from a customer support database and training a generative AI model with it. Furthermore, it is a system that combines this with an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response based on those emotions.

[0803] Specific examples of program processing

[0804] Database Preparation

[0805] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions and troubleshooting guides for new products in the database. The database includes PDFs, text files, images, and other formats.

[0806] Learning of generative AI

[0807] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. The server then supplies the preprocessed text data to a generative AI model for training. For example, the server extracts "initial setup procedure" as text and uses it to train the AI ​​model.

[0808] Receiving inquiries from users

[0809] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[0810] Handling inquiries

[0811] The terminal receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the terminal receives the question "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0812] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0813] How the emotion recognition engine works

[0814] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. The emotion recognition engine detects the user's emotional state from the context of the input, voice, facial expressions, etc. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[0815] The server can adjust the content and expression of responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the user is irritated, the server will change the response to a more polite expression.

[0816] Escalation engine operation

[0817] The server activates the escalation engine if the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user's emotion is recognized as "very angry," the escalation engine will contact the support team leader.

[0818] Providing a response

[0819] The server sends the responses generated by the generative AI model and emotion recognition engine to the device. The device then displays this to the user. For example, the server sends a detailed instruction to the device, such as "Select Bluetooth from the settings menu, then search for the device you want to connect to," which the device then displays in the chat window for the user. The user then uses this information to configure the product.

[0820] This system recognizes user emotions and provides quick and appropriate responses, enabling more efficient support compared to traditional customer support. The introduction of an emotion recognition engine can further improve user satisfaction.

[0821] The following describes the processing flow.

[0822] Step 1:

[0823] The server registers manual information about products and services in the customer support database. This includes tasks such as storing initial product setup instructions and troubleshooting guides in PDF or text file format in the database.

[0824] Step 2:

[0825] The server extracts the necessary manual information from the database and preprocesses that information. Preprocessing includes cleaning and formatting text data. For example, the server extracts text from PDF files and converts it into a format that generative AI models can understand.

[0826] Step 3:

[0827] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This enables the AI ​​to generate appropriate answers to user inquiries. For example, the server trains the AI ​​model on "initial setup procedures."

[0828] Step 4:

[0829] The device provides an interface for receiving inquiries from users. Users enter their problems or questions through web forms or chatbots. For example, a user might type, "Please tell me how to connect the product via Bluetooth."

[0830] Step 5:

[0831] The terminal receives the user's inquiry and then sends that information to the server. The server then receives this inquiry and sends it to a generative AI model. For example, the terminal receives the inquiry "Please tell me how to connect the product via Bluetooth" and sends it to the server, which then passes it on to the AI ​​model.

[0832] Step 6:

[0833] Generative AI models analyze the content of inquiries and generate the most suitable answers based on information stored in a database. For example, they might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0834] Step 7:

[0835] When a terminal receives a user inquiry, it sends the input content to an emotion recognition engine. The emotion recognition engine analyzes the context of the input text, as well as voice and facial expressions, to detect the user's emotional state. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[0836] Step 8:

[0837] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the server detects that the user is angry, it will change the response to a more polite and calm expression.

[0838] Step 9:

[0839] The server activates the escalation engine if it determines that the user's emotional state is "urgent." The escalation engine follows a predefined procedure to escalate the issue to a higher-level support person. For example, if it is determined that the user is "very angry," the support team leader will be contacted.

[0840] Step 10:

[0841] The device displays responses generated by a generative AI model and an emotion recognition engine to the user. For example, it might display detailed instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0842] Through the steps described above, a system incorporating an emotion recognition engine can provide prompt and appropriate support that takes user emotions into account. This makes it possible to improve the efficiency of customer support and increase user satisfaction.

[0843] (Example 2)

[0844] 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."

[0845] Conventional customer support systems often suffer from delays in responding to user inquiries or fail to provide appropriate answers. Furthermore, they may fail to consider user emotions, leading to dissatisfaction. This results in decreased user satisfaction and customer churn. The present invention aims to solve these problems by providing a system that responds quickly and accurately to user inquiries, and further recognizes and appropriately addresses user emotions.

[0846] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0847] In this invention, the server includes means for acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, and means for an emotion recognition engine for analyzing the user's emotions. This enables a quick and appropriate response to user inquiries, and further, the provision of appropriate answers that take the user's emotions into consideration.

[0848] A "customer support database" is a database that stores various manuals and troubleshooting guides related to products and services, and is used to handle customer inquiries.

[0849] "Manual information" refers to information that includes detailed instructions and guides on how to use a product or service, setup procedures, troubleshooting, and more.

[0850] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate appropriate responses from input data.

[0851] A "user interface" is the interface through which a user makes inquiries to a system, and includes web forms, chatbots, and other similar elements.

[0852] An "emotion recognition engine" is an engine that analyzes and recognizes emotions from the user's input text, voice, and facial expressions.

[0853] "Methods for escalating the issue" refer to ways of transferring the problem to a higher-level person in charge, depending on the user's emotional state.

[0854] A "rapid response system for real-time delivery" is a system that provides generated answers to users immediately, with the aim of responding quickly to inquiries.

[0855] This invention provides a system that responds quickly and appropriately to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. Furthermore, this system incorporates an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response accordingly.

[0856] Database Preparation

[0857] The server registers manual information about products and services in the customer support database. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in PDF, text file, and image formats in the database. For example, the server saves "Printer Ink Replacement Procedure" as a PDF file in the database.

[0858] Learning of generative AI

[0859] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning the text data (removing unnecessary spaces and line breaks) and format conversion (from PDF to text). The server then feeds this preprocessed text to a generative AI model. By training the generative AI model (e.g., GPT-3) with this data, the AI ​​learns how to configure and troubleshoot the product.

[0860] Receiving inquiries from users

[0861] The device provides an interface for receiving inquiries from users. Users enter their questions using web forms or chatbots. For example, a user might type "How do I connect the product via Bluetooth?" into the chatbot.

[0862] Handling inquiries

[0863] The terminal receives a user inquiry and sends its contents to the server. The server analyzes the inquiry and sends it to a generative AI model. For example, the terminal sends the question "How do I connect the product via Bluetooth?" to the server, which then passes this to the AI. The generative AI model analyzes the inquiry and generates an appropriate answer based on the information stored in the database. For example, the generative AI model might generate the procedure "Select Bluetooth from the settings menu, and then search for the device you want to connect to."

[0864] How the emotion recognition engine works

[0865] The device uses an emotion recognition engine to analyze the user's emotions when they make an inquiry. The emotion recognition engine detects the user's emotional state from the context of the input, voice, and facial expressions. For example, if a user sends a message that includes frustration, such as "Tell me quickly!", the emotion recognition engine will analyze this as "frustration." Based on the user's emotional state analyzed by the emotion recognition engine, the server adjusts the content and expression of the response generated by the generative AI model. For example, if the user is frustrated, the server adjusts the response to provide a more polite and prompt response.

[0866] Escalation engine operation

[0867] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user is identified as "very angry," the escalation engine will contact the support team leader and have them handle the issue directly.

[0868] Providing a response

[0869] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The device displays this to the user. The generated response, "Select Bluetooth from the settings menu and then search for the device you want to connect to," is politely rephrased according to the emotion to, "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to." This response is displayed in the device's chat window and provided to the user. The user uses this information to configure the product settings.

[0870] Examples of specific cases and prompt statements

[0871] Specific example

[0872] 1. User's question: "My printer is jammed. How can I fix it?"

[0873] 2. Server preprocessing: Extract troubleshooting guides for "paper jams" from the database and use them to train a generative AI model.

[0874] 3. Emotion Analysis: The emotion recognition engine detects "frustration" from the user's voice and facial expressions.

[0875] 4. Response generation and refinement: The generative AI model creates the procedure "turn off the power and carefully remove the paper," and the server refines the expression more carefully based on the emotion recognition results.

[0876] 5. Response Provision: The adjusted response is provided to the user by the device (for example, displayed in the chat window).

[0877] Example of a prompt

[0878] User's question: "My printer is jammed. How can I fix it?"

[0879] Sentiment Analysis: Detects irritation from input context, voice, and facial expressions.

[0880] Generative AI model's response: "Turn off the power and carefully remove the paper."

[0881] Emotion-based adjustment: "I'm sorry, but could you please turn off the printer and carefully remove the jammed paper?"

[0882] In this way, the system can respond to user inquiries efficiently and with consideration for their feelings.

[0883] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0884] Step 1:

[0885] The server registers manual information about products and services in the customer support database. Input includes manual information in PDF, text file, and image formats. The specific process for storing this information in the database involves the server receiving the manual information in various formats and saving it in the database in the appropriate format. For example, a PDF file titled "Printer Ink Replacement Procedure" might be saved in the database.

[0886] Step 2:

[0887] The server extracts the necessary manual information from the database and preprocesses it. The input is the manual information in the database. Specifically, the server cleans the text data (removing unnecessary spaces and line breaks) and converts the format (from PDF to text). The output is the preprocessed text data. For example, a PDF document titled "Initial Setup Procedure" is converted to text format.

[0888] Step 3:

[0889] The server supplies pre-processed text data to a generative AI model. The input is pre-processed text data. Specifically, the server feeds this data to a generative AI model (e.g., GPT-3) to train the model. The output is the trained generative AI model.

[0890] Step 4:

[0891] The terminal provides an interface for receiving inquiries from users. Input includes questions entered by the user via web forms or chatbots. Its specific operation is to retrieve the user's inquiry content through the user interface and send it to the server. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[0892] Step 5:

[0893] The server receives the user's inquiry and sends it to a generative AI model. The input is the inquiry sent from the terminal. Specifically, the server analyzes the inquiry and passes it to the generative AI model. The output is the answer generated by the generative AI model. For example, it might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0894] Step 6:

[0895] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. Input includes the user's written text, voice, and facial expressions. The emotion recognition engine analyzes this information to detect the user's emotional state. The output is the analyzed emotional state of the user (e.g., "irritated").

[0896] Step 7:

[0897] The server adjusts the response of the generative AI model based on the results analyzed by the emotion recognition engine. The inputs are the generated response and the emotion state from the emotion recognition engine. Specifically, the server adjusts the response generated by the generative AI model according to the emotion state, changing its expression. The output is the adjusted response. For example, the response might be changed to "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to."

[0898] Step 8:

[0899] The server considers the user's emotional state and escalates the issue to a higher-level person in charge. The input is the emotional state, which the emotion recognition engine has determined to be "urgent." The specific action is for the server to activate the escalation engine and transfer the issue to a higher-level support person. The output is a notification to the higher-level person in charge that the issue has been escalated.

[0900] Step 9:

[0901] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The input is a refined response. Specifically, the server sends this response to the device, and the device displays it to the user. The output is the response displayed to the user. For example, the device's chat window might display helpful instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0902] (Application Example 2)

[0903] 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."

[0904] Traditional customer support systems often resulted in low user satisfaction because they provided uniform responses without considering the user's emotional state. Furthermore, their limited functionality for providing timely and appropriate answers to inquiries made efficient support difficult. Additionally, the lack of real-time support utilizing in-store voice interfaces created a need to improve the user experience.

[0905] 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 acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, emotion recognition engine means for analyzing the user's emotional state, means for adjusting the content and expression of the response based on the emotional state, escalation engine means for escalating the problem to a higher-level support staff member based on the user's emotional state, and voice interface means for receiving user inquiries by voice using speech recognition and providing responses by voice. This enables improved user satisfaction, the provision of quick and appropriate answers, and effective customer support through a real-time voice interface.

[0906] A "customer support database" is a database that stores information such as manuals, troubleshooting guides, and FAQs related to products and services.

[0907] "Manual information" refers to information that includes detailed explanations of how to use a product or service, setup procedures, troubleshooting, and more.

[0908] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to automatically generate appropriate answers to user inquiries.

[0909] A "user interface" is an interface through which a user inputs inquiries and interacts with the system.

[0910] An "emotion recognition engine" is an engine that analyzes the user's emotional state from their input or voice to detect emotions such as "anger" or "joy."

[0911] The "escalation engine" is an engine that, based on the user's emotional state analyzed by the emotion recognition engine, escalates the issue to a higher-level support representative as needed.

[0912] A "rapid response system" is a system that provides users with answers generated by generative AI models in real time.

[0913] A "voice interface" is an interface that uses speech recognition technology to receive user inquiries in voice and provide responses in voice.

[0914] This invention is a system for streamlining customer support in physical stores. It retrieves manual information from a customer support database and uses it to train a generative AI model, enabling it to respond quickly and accurately to user inquiries. It also incorporates an emotion recognition engine that analyzes the user's emotional state at the time of inquiry and adjusts the content and expression of the response accordingly.

[0915] The server retrieves product and service manual information (e.g., initial setup procedures and troubleshooting guides) from the customer support database. This manual information is stored in various formats, including PDFs, text files, and images. The server preprocesses this data and trains a generative AI model. This process involves cleaning and formatting the text data. For example, it might extract the initial setup procedures for a new product as text and train the generative AI model.

[0916] The terminal provides an interface for receiving user inquiries. This is done via applications installed on smartphones or robots placed within the store. Users can enter questions through web forms, chatbots, or voice interfaces. For example, if a user asks, "Do you have this product in stock?", the terminal sends this inquiry to the server.

[0917] The server sends the received inquiry to a generative AI model, which analyzes the inquiry and generates an appropriate response. For example, it might generate a response such as, "We will check the stock status. Please wait a moment." The server also uses an emotion recognition engine to analyze the user's emotions. It detects the user's emotional state (e.g., "anger") from the context of the input, voice, and facial expressions, and adjusts the content and expression of the response generated by the generative AI model based on this. If the user is angry, the response is changed to a more polite expression.

[0918] Furthermore, if the emotion recognition engine determines that the user's emotional state is "urgent," the server activates the escalation engine to escalate the issue to a higher-level support person. For example, if the user's emotions are recognized as "very angry," the server will contact the support team leader.

[0919] In summary, this invention enables improved user satisfaction, the provision of quick and appropriate responses, and effective customer support through a real-time voice interface.

[0920] Specific example

[0921] When a user asks, "Do you have this item in stock?", the speech recognition engine converts the speech into text, and the generative AI model generates the response, "We will check the stock status. Please wait a moment."

[0922] Examples of prompts to input into a generative AI model:

[0923] User: Do you have this item in stock?

[0924] assistant:

[0925] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0926] Step 1: Prepare the database

[0927] The server registers product and service manual information in the customer support database. Inputs include PDFs, text files, and images, such as initial setup instructions and troubleshooting guides for new products. The server stores this information in the database. The output is that the stored information becomes available for use.

[0928] Step 2: Training a generative AI

[0929] The server extracts necessary manual information from the database and performs text data cleaning and format conversion. The input is the manual information obtained from the database. The server supplies the pre-processed text data to a generative AI model for training. The output is that the generative AI model can generate appropriate answers to queries.

[0930] Step 3: Receiving inquiries from users

[0931] The terminal provides an interface for receiving inquiries from users. Inputs include questions entered by the user via web forms, chatbots, or voice interfaces. The terminal receives these inquiries and generates data to send to the server as output.

[0932] Step 4: Processing the inquiry

[0933] The server sends the query received from the terminal to a generative AI model. The input is the user's question. The server processes the data necessary for the generative AI model to analyze the query and generate an appropriate answer. The server receives the generated answer as output.

[0934] Step 5: Operation of the emotion recognition engine

[0935] The device uses an emotion recognition engine to analyze the user's emotions when a user makes a request. Input includes the user's context, voice, and facial expressions. The device analyzes this information and sends the user's emotional state (e.g., "anger," "frustration") to the server as output.

[0936] Step 6: Adjusting the content and wording of your answer

[0937] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. The input consists of the user's emotional state and the generated responses. The server adjusts these and outputs the adjusted responses.

[0938] Step 7: Operation of the escalation engine

[0939] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The input is the analyzed user's emotional state. The server escalates the issue to a higher-level support person and generates an escalated issue as output.

[0940] Step 8: Provide your answer

[0941] The server sends the responses generated by the generative AI model and emotion recognition engine to the terminal. The input is the adjusted response. The terminal displays this to the user, and the output is the information provided to the user.

[0942] As a concrete example, if a user asks, "Do you have this item in stock?", the device uses a speech recognition engine to convert the speech into text and sends it to the server. The server uses a generative AI model to generate an answer and an emotion recognition engine to analyze the user's emotional state. After appropriate adjustments are made, the device provides the user with an answer such as, "I will check the stock status. Please wait a moment."

[0943] Examples of prompts to input into a generative AI model:

[0944] User: Do you have this item in stock?

[0945] assistant:

[0946] 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.

[0947] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">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.

[0948] 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.

[0949] [Fourth Embodiment]

[0950] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0951] 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.

[0952] 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).

[0953] 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.

[0954] 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.

[0955] 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).

[0956] 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.

[0957] 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.

[0958] 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.

[0959] 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.

[0960] 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.

[0961] 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.

[0962] 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".

[0963] This invention provides a system that quickly responds to user inquiries by retrieving manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below.

[0964] Specific examples of program processing

[0965] Database Preparation

[0966] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions for new products in the database. The database contains data in various formats, such as PDFs, text files, and images.

[0967] Learning of generative AI

[0968] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, the server extracts "product initial setup instructions" as text and converts it into a format that the AI ​​model can understand.

[0969] Receiving inquiries from users

[0970] The device provides an interface for receiving inquiries from users. For example, the device can receive inquiries from users such as "How do I connect my product via Bluetooth?" through a web form or chatbot.

[0971] Handling inquiries

[0972] The device sends the user's inquiry to the server. The server receives this and forwards it to a generative AI model. For example, the device receives the question, "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[0973] Generative AI models generate appropriate answers to queries based on information stored in a database. For example, an AI model might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[0974] Providing a response

[0975] The server sends the response received from the generative AI model to the terminal. The terminal then displays this to the user. For example, the server sends the received steps to the terminal, which then displays them in a chat window for the user to see. In this way, the user can quickly resolve the problem.

[0976] Specific example

[0977] Example 1: How to deal with a power failure

[0978] The server registers "solutions for when the power won't turn on" in its database. The server then uses this information to train a generative AI model.

[0979] When a user enters the question "What should I do if the device won't turn on?" into the chatbot, the device receives this and sends it to the server. The generative AI model generates the response "Try restarting by pressing and holding the power button for 10 seconds." The server sends this response back to the device, which then displays it to the user.

[0980] This system enables faster and more efficient responses compared to traditional customer support, reducing the burden on sales support centers and care centers, and is expected to improve customer satisfaction.

[0981] The following describes the processing flow.

[0982] Step 1:

[0983] The server registers product and service manual information in the customer support database. This includes initial setup instructions and troubleshooting guides for Hianabi. The server stores this information in the database in PDF and text file formats.

[0984] Step 2:

[0985] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. For example, the server extracts text from PDF files and converts it to a format that can be used by generative AI models for training.

[0986] Step 3:

[0987] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This process prepares the AI ​​model to understand text data and generate appropriate responses. For example, the server trains the AI ​​model on "initial setup procedures."

[0988] Step 4:

[0989] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect Hianabi via Bluetooth?"

[0990] Step 5:

[0991] The device receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the device might send the question "How do I connect Hianabi via Bluetooth?" to the server, which then passes this to the AI.

[0992] Step 6:

[0993] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[0994] Step 7:

[0995] The server sends the response received from the generative AI model to the terminal. The server holds the generated response and then sends it to the terminal. This process is rapid, so the user can receive a response immediately.

[0996] Step 8:

[0997] The device displays the user the response received from the server. For example, the device might display instructions in the chat window such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[0998] In this way, specific processes are carried out at each step, resulting in a system that responds quickly and accurately to user inquiries.

[0999] (Example 1)

[1000] 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".

[1001] Traditional customer support systems often suffered from delays in responding to user inquiries and a failure to provide appropriate information. Furthermore, some inquiries required manual handling by operators, leading to decreased efficiency. These problems could negatively impact the quality and speed of customer support, potentially harming customer satisfaction.

[1002] 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.

[1003] In this invention, the server includes means for acquiring information from a customer support database, means for preprocessing and formatting the acquired information to train a generative AI model, and means for receiving inquiries from users. This makes it possible to provide immediate and appropriate answers to inquiries from users.

[1004] A "customer support database" is a database that stores support information for users and manual information about products and services.

[1005] "Information" refers to data such as detailed instructions, troubleshooting guides, and FAQs about products and services.

[1006] "Preprocessing" refers to a series of processes that convert data into a format that generative AI models can understand. Specifically, this includes tasks such as extracting text from PDFs and format conversion.

[1007] "Format conversion" refers to the process of converting acquired information into a format that is easily understood by generative AI models.

[1008] A "generative AI model" is an artificial intelligence model that learns from information obtained from a customer support database and generates appropriate answers to user inquiries.

[1009] "Interface means" refers to input methods such as web forms and chatbots that users use to enter inquiries.

[1010] A "server" refers to a device that retrieves information from a customer support database, performs preprocessing and format conversion, and supplies the data to a generative AI model.

[1011] "Inquiry" refers to the act of a user inputting specific questions or problems about a product or service through an interface.

[1012] "Answer" refers to the information that a generative AI model generates based on a user's inquiry.

[1013] A "rapid response system" refers to a system that provides users with answers generated by generative AI models in real time.

[1014] This invention provides a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. The specific operation of the system is described below from the perspectives of the server, terminal, and user.

[1015] Database Preparation

[1016] The server registers manual information about products and services in the customer support database. Specifically, it stores information such as initial product setup procedures and troubleshooting guides in various formats, including PDFs, text files, and images. For example, it uploads the setup instructions for a new smartphone to the database.

[1017] Information extraction and preprocessing

[1018] The server extracts necessary manual information from the customer support database. During this process, it performs data preprocessing and format conversion. Specifically, it extracts text from PDFs and converts it into a format that generative AI models can understand. For example, it extracts the text "Connect the device to the power source and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[1019] Training of generative AI models

[1020] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. Specifically, it inputs text data into the AI ​​model and repeats a feedback loop with the generated response. For example, by training the AI ​​model with a manual on "how to turn on the power," it can generate the response "Press and hold the power button."

[1021] Receiving inquiries from users

[1022] The device provides an interface for receiving inquiries from users. Specifically, it receives user questions using web forms or chatbots. For example, the device might receive a user inquiry such as, "Please tell me how to set up my new smartphone."

[1023] Processing queries and sending them to the server

[1024] The device forwards the user's inquiry to the server. Specifically, it converts questions entered into web forms or chatbots into the appropriate format and sends them to the server. For example, it forwards the question "How do I connect my smartphone via Bluetooth?" to the server.

[1025] Answer generation using generative AI models

[1026] The server passes the received user inquiry to a generative AI model. The generative AI model generates an answer to the question based on the information stored in the database. Specifically, in response to the question, "How do I connect my smartphone via Bluetooth?", it generates the answer, "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[1027] Providing a response

[1028] The server sends the response received from the generative AI model to the terminal, which then displays it to the user. Specifically, the server-generated response is displayed to the user through a chatbot or web form. For example, the terminal might provide the user with the response, "Open the settings menu, select Bluetooth, and choose the device you want to connect to," in a chat window.

[1029] Specific example

[1030] Example 1: How to deal with a power failure

[1031] 1. The server registers "Troubleshooting steps for when the power does not turn on" as a PDF in its database.

[1032] 2. The server extracts the text "Please try restarting by pressing and holding the power button for 10 seconds" from the registered PDF and preprocesses it.

[1033] 3. The server uses the preprocessed text to train a generative AI model.

[1034] 4. The user enters "What should I do if the device won't turn on?" into the device's chatbot.

[1035] 5. The terminal sends the user's inquiry to the server.

[1036] 6. The generative AI model generates the response, "Please try restarting by pressing and holding the power button for 10 seconds."

[1037] 7. The server sends the generated response to the terminal, which then displays it in the chat window and provides it to the user.

[1038] In this way, users can quickly solve problems. As a specific example, the following prompt can be input into the generating AI model.

[1039] Example of a prompt

[1040] "Please tell me how to connect the product via Bluetooth."

[1041] "Please tell me what to do if the power won't turn on."

[1042] This system is expected to enable faster and more efficient responses compared to traditional customer support, reducing the workload on operators and improving customer satisfaction.

[1043] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1044] Step 1:

[1045] The server registers manual information about products and services in the customer support database. Inputs include initial setup instructions and troubleshooting guides for new products, provided in PDF, text file, and image formats. The server stores this information in the database and generates an organized database as output. Specifically, it uploads new product setup instructions as PDFs to the database.

[1046] Step 2:

[1047] The server extracts necessary manual information from the customer support database. PDFs and text files stored in the database are used as input. The server processes the data and converts it into a format suitable for generative AI models. The output is text data that the AI ​​model can understand. Specifically, it extracts the text "Connect the device to the power supply and press the power button" from a PDF and converts it into text data that the AI ​​model can understand.

[1048] Step 3:

[1049] The server supplies pre-processed manual information to a generative AI model for training. Formatted text data is used as input. The server inputs this into the AI ​​model and improves the model's performance through a feedback loop. The output is a fully trained AI model. As a specific example, the AI ​​model is trained using text data about "how to turn on the power."

[1050] Step 4:

[1051] Users utilize an interface to submit inquiries from their devices. As input, the user's question is entered into the device via a web form or chatbot. The device receives this input and sends it to the server. As output, the inquiry content is sent to the server. For example, a user might enter the inquiry "Please tell me the initial setup procedure for my new smartphone" into the chatbot.

[1052] Step 5:

[1053] The terminal converts the received inquiry into the appropriate format and sends it to the server. The user's inquiry is provided to the terminal as input. The terminal converts this into the appropriate format and sends it to the server. The formatted inquiry is sent to the server as output. As a specific example, the question "How do I connect my smartphone via Bluetooth?" is converted into the appropriate format and sent to the server.

[1054] Step 6:

[1055] The server passes the user's inquiry received to a generative AI model. The server receives the inquiry content sent from the terminal as input. The server inputs this into the AI ​​model and generates an answer. The generated answer is obtained as output. For example, in response to the question "How do I connect my smartphone via Bluetooth?", it generates the answer "Open the settings menu, select Bluetooth, and choose the device you want to connect to."

[1056] Step 7:

[1057] The server sends the response received from the generative AI model to the terminal. The server receives the response generated by the AI ​​model as input. The server sends this to the terminal, which displays it to the user. The output is the response displayed to the user. Specifically, the server sends the generated response to the terminal via a chatbot or web form, and the terminal provides the user with the response "Open the settings menu, select Bluetooth, and choose the device you want to connect to" in the chat window.

[1058] (Application Example 1)

[1059] 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".

[1060] Current customer support systems struggle to respond quickly to user inquiries. This is especially true for e-commerce sites, where detailed product inquiries are frequent, requiring efficient and immediate responses. Furthermore, many users prefer to communicate via voice as well as text. However, few systems successfully balance the requirements for providing appropriate answers with user convenience, resulting in decreased user satisfaction and operational inefficiencies. These problems need to be addressed.

[1061] 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.

[1062] In this invention, the server includes means for obtaining manual information from a customer support database, means for training a generative AI model with the obtained manual information, means for a user interface for receiving inquiries from users, means for users to input questions about products in text or voice, means for sending received inquiries to the generative AI model, and means for receiving and providing answers generated by the generative AI model to the user. This makes it possible to respond quickly to user inquiries and improve user convenience.

[1063] A "customer support database" is a data storage system that stores information about products and services in order to respond to inquiries from users.

[1064] "Manual information" refers to documents and data that provide detailed information on how to use a product or service, setup procedures, troubleshooting, and more.

[1065] A "generative AI model" is an artificial intelligence algorithm that generates appropriate answers from given data.

[1066] A "user interface" is a software or hardware configuration that includes interactive elements that allow a user to directly interact with the system.

[1067] "Text or voice input" refers to the process by which users submit inquiries through text input or voice input.

[1068] A "rapid response system" is a mechanism that provides users with answers generated by generative AI models in real time.

[1069] This invention is a system that quickly responds to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. A specific embodiment of this system is described below.

[1070] System program

[1071] Database Preparation

[1072] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. The server also stores initial setup instructions for new products in the database.

[1073] Learning of generative AI

[1074] The server extracts the necessary manual information from the database and uses it to train a generative AI model. This process involves data preprocessing and format conversion. For example, "product initial setup instructions" are extracted as text and converted into a format that the AI ​​model can understand.

[1075] User Interface

[1076] The user interface is an interface for users to enter questions about the product. This interface has the functionality to accept text input and voice input. It allows users to enter inquiries such as, "How do I connect the product via Bluetooth?"

[1077] Handling inquiries

[1078] The server receives the user's inquiry and forwards it to a generative AI model. The generative AI model generates an appropriate response based on the information stored in the database. For example, it might generate instructions such as, "First, select Bluetooth from the settings menu, then search for the device you want to connect to."

[1079] Providing a response

[1080] The server sends the response received from the generative AI model to the user interface. The user interface provides the response to the user in text or voice.

[1081] Hardware and software to be used

[1082] This system uses the following hardware and software.

[1083] Hardware:

[1084] Smartphone (the device on which the user uses the app)

[1085] Servers (data centers where databases and AI models run)

[1086] software:

[1087] OpenAI API (utilizing generative AI models)

[1088] A temporary database module (a module that communicates with the database)

[1089] A temporary chatbot interface (a module that interacts with the user)

[1090] Examples of specific cases and prompt statements

[1091] Specific example

[1092] A user asks, "I'd like to know more about the shipping method for this product." The smartphone app sends the question to the server and retrieves the "shipping method" information from the database. The AI ​​generates a response saying, "Regarding shipping methods, it is usually shipped by courier service, but if you are in a hurry, an express service is available for an additional fee." The app then displays this response to the user.

[1093] Example of a prompt

[1094] The following are examples of prompts to input into the generated AI model.

[1095] User question: I'd like to know more about the shipping methods for this product.

[1096] Search results from our customer support database: Our standard shipping method is courier service, but for urgent orders, an express delivery service is available for an additional fee.

[1097] Answer the user's query based on the above information.

[1098] In this way, this invention makes it possible to realize a system that responds quickly and accurately to user inquiries.

[1099] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1100] Step 1: Prepare the database

[1101] The server registers manual information related to products and services in the customer support database. The database contains data in various formats, including PDFs, text files, and images. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in the database. The input here is the manual information, and the output is the information registered in the database.

[1102] Step 2: Training a generative AI

[1103] The server extracts the necessary manual information from the database and uses it to train a generative AI model. Data preprocessing and format conversion are performed to transform the information into a format that the AI ​​model can understand. For example, the server extracts "product initial setup procedures" in text format and feeds them into the generative AI model. Here, the input is the manual information extracted from the database, and the output is the completed generative AI model.

[1104] Step 3: User Interface Operation

[1105] The user enters questions about the product through the terminal's user interface. This interface accepts both text and voice input. Specifically, the user might enter the question, "How do I connect the product via Bluetooth?" The input here is the user's question, and the output is the content of that question.

[1106] Step 4: Submitting the inquiry

[1107] The terminal sends the user's query to the server. The server forwards the received query to a generative AI model. At this stage, the terminal sends the received query to the server, and the server passes it to the AI. The input here is the user's query, and the output is the prompt message passed to the AI ​​model.

[1108] Step 5: AI-powered response generation

[1109] Generative AI models generate appropriate answers based on information stored in a database. For example, an AI model might generate the following procedure: "First, select Bluetooth from the settings menu, then search for the device you want to connect to." The input here is the prompt received from the server, and the output is the generated answer.

[1110] Step 6: Provide your response

[1111] The server sends the response received from the generative AI model to the terminal. The terminal provides the response to the user in text or voice. For example, the server sends the received procedure to the terminal, and the terminal displays it to the user. Here, the input is the response received from the AI ​​model, and the output is the response provided to the user through the user interface.

[1112] 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.

[1113] This invention provides a system that responds quickly and accurately to user inquiries by acquiring manual information from a customer support database and training a generative AI model with it. Furthermore, it is a system that combines this with an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response based on those emotions.

[1114] Specific examples of program processing

[1115] Database Preparation

[1116] The server registers manual information about products and services in the customer support database. For example, the server stores initial setup instructions and troubleshooting guides for new products in the database. The database includes PDFs, text files, images, and other formats.

[1117] Learning of generative AI

[1118] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning and formatting the text data. The server then supplies the preprocessed text data to a generative AI model for training. For example, the server extracts "initial setup procedure" as text and uses it to train the AI ​​model.

[1119] Receiving inquiries from users

[1120] The device provides an interface for receiving inquiries from users. Users enter their questions through web forms or chatbots. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[1121] Handling inquiries

[1122] The terminal receives a user inquiry and then sends the content to the server. The server receives the inquiry and sends it to a generative AI model. For example, the terminal receives the question "How do I connect the product via Bluetooth?" and sends it to the server, which then passes it on to the AI.

[1123] Generative AI models analyze the content of a query and generate an appropriate answer based on information stored in a database. For example, an AI model might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[1124] How the emotion recognition engine works

[1125] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. The emotion recognition engine detects the user's emotional state from the context of the input, voice, facial expressions, etc. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[1126] The server can adjust the content and expression of responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the user is irritated, the server will change the response to a more polite expression.

[1127] Escalation engine operation

[1128] The server activates the escalation engine if the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user's emotion is recognized as "very angry," the escalation engine will contact the support team leader.

[1129] Providing a response

[1130] The server sends the responses generated by the generative AI model and emotion recognition engine to the device. The device then displays this to the user. For example, the server sends a detailed instruction to the device, such as "Select Bluetooth from the settings menu, then search for the device you want to connect to," which the device then displays in the chat window for the user. The user then uses this information to configure the product.

[1131] This system recognizes user emotions and provides quick and appropriate responses, enabling more efficient support compared to traditional customer support. The introduction of an emotion recognition engine can further improve user satisfaction.

[1132] The following describes the processing flow.

[1133] Step 1:

[1134] The server registers manual information about products and services in the customer support database. This includes tasks such as storing initial product setup instructions and troubleshooting guides in PDF or text file format in the database.

[1135] Step 2:

[1136] The server extracts the necessary manual information from the database and preprocesses that information. Preprocessing includes cleaning and formatting text data. For example, the server extracts text from PDF files and converts it into a format that generative AI models can understand.

[1137] Step 3:

[1138] The server supplies pre-processed manual information to a generative AI model, allowing the model to learn. This enables the AI ​​to generate appropriate answers to user inquiries. For example, the server trains the AI ​​model on "initial setup procedures."

[1139] Step 4:

[1140] The device provides an interface for receiving inquiries from users. Users enter their problems or questions through web forms or chatbots. For example, a user might type, "Please tell me how to connect the product via Bluetooth."

[1141] Step 5:

[1142] The terminal receives the user's inquiry and then sends that information to the server. The server then receives this inquiry and sends it to a generative AI model. For example, the terminal receives the inquiry "Please tell me how to connect the product via Bluetooth" and sends it to the server, which then passes it on to the AI ​​model.

[1143] Step 6:

[1144] Generative AI models analyze the content of inquiries and generate the most suitable answers based on information stored in a database. For example, they might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[1145] Step 7:

[1146] When a terminal receives a user inquiry, it sends the input content to an emotion recognition engine. The emotion recognition engine analyzes the context of the input text, as well as voice and facial expressions, to detect the user's emotional state. For example, if the user is irritated, the emotion recognition engine will analyze this as "anger."

[1147] Step 8:

[1148] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. For example, if the server detects that the user is angry, it will change the response to a more polite and calm expression.

[1149] Step 9:

[1150] The server activates the escalation engine if it determines that the user's emotional state is "urgent." The escalation engine follows a predefined procedure to escalate the issue to a higher-level support person. For example, if it is determined that the user is "very angry," the support team leader will be contacted.

[1151] Step 10:

[1152] The device displays responses generated by a generative AI model and an emotion recognition engine to the user. For example, it might display detailed instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to." The user then uses this information to configure the product.

[1153] Through the steps described above, a system incorporating an emotion recognition engine can provide prompt and appropriate support that takes user emotions into account. This makes it possible to improve the efficiency of customer support and increase user satisfaction.

[1154] (Example 2)

[1155] 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".

[1156] Conventional customer support systems often suffer from delays in responding to user inquiries or fail to provide appropriate answers. Furthermore, they may fail to consider user emotions, leading to dissatisfaction. This results in decreased user satisfaction and customer churn. The present invention aims to solve these problems by providing a system that responds quickly and accurately to user inquiries, and further recognizes and appropriately addresses user emotions.

[1157] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1158] In this invention, the server includes means for acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, and means for an emotion recognition engine for analyzing the user's emotions. This enables a quick and appropriate response to user inquiries, and further, the provision of appropriate answers that take the user's emotions into consideration.

[1159] A "customer support database" is a database that stores various manuals and troubleshooting guides related to products and services, and is used to handle customer inquiries.

[1160] "Manual information" refers to information that includes detailed instructions and guides on how to use a product or service, setup procedures, troubleshooting, and more.

[1161] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate appropriate responses from input data.

[1162] A "user interface" is the interface through which a user makes inquiries to a system, and includes web forms, chatbots, and other similar elements.

[1163] An "emotion recognition engine" is an engine that analyzes and recognizes emotions from the user's input text, voice, and facial expressions.

[1164] "Methods for escalating the issue" refer to ways of transferring the problem to a higher-level person in charge, depending on the user's emotional state.

[1165] A "rapid response system for real-time delivery" is a system that provides generated answers to users immediately, with the aim of responding quickly to inquiries.

[1166] This invention provides a system that responds quickly and appropriately to user inquiries by obtaining manual information from a customer support database and training a generative AI model with it. Furthermore, this system incorporates an emotion recognition engine that recognizes the user's emotions and adjusts the content and expression of the response accordingly.

[1167] Database Preparation

[1168] The server registers manual information about products and services in the customer support database. Specifically, the server stores initial setup procedures and troubleshooting guides for new products in PDF, text file, and image formats in the database. For example, the server saves "Printer Ink Replacement Procedure" as a PDF file in the database.

[1169] Learning of generative AI

[1170] The server extracts the necessary manual information from the database and preprocesses it. Preprocessing includes cleaning the text data (removing unnecessary spaces and line breaks) and format conversion (from PDF to text). The server then feeds this preprocessed text to a generative AI model. By training the generative AI model (e.g., GPT-3) with this data, the AI ​​learns how to configure and troubleshoot the product.

[1171] Receiving inquiries from users

[1172] The device provides an interface for receiving inquiries from users. Users enter their questions using web forms or chatbots. For example, a user might type "How do I connect the product via Bluetooth?" into the chatbot.

[1173] Handling inquiries

[1174] The terminal receives a user inquiry and sends its contents to the server. The server analyzes the inquiry and sends it to a generative AI model. For example, the terminal sends the question "How do I connect the product via Bluetooth?" to the server, which then passes this to the AI. The generative AI model analyzes the inquiry and generates an appropriate answer based on the information stored in the database. For example, the generative AI model might generate the procedure "Select Bluetooth from the settings menu, and then search for the device you want to connect to."

[1175] How the emotion recognition engine works

[1176] The device uses an emotion recognition engine to analyze the user's emotions when they make an inquiry. The emotion recognition engine detects the user's emotional state from the context of the input, voice, and facial expressions. For example, if a user sends a message that includes frustration, such as "Tell me quickly!", the emotion recognition engine will analyze this as "frustration." Based on the user's emotional state analyzed by the emotion recognition engine, the server adjusts the content and expression of the response generated by the generative AI model. For example, if the user is frustrated, the server adjusts the response to provide a more polite and prompt response.

[1177] Escalation engine operation

[1178] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The escalation engine follows predefined procedures to escalate the issue to a higher-level support person. For example, if the user is identified as "very angry," the escalation engine will contact the support team leader and have them handle the issue directly.

[1179] Providing a response

[1180] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The device displays this to the user. The generated response, "Select Bluetooth from the settings menu and then search for the device you want to connect to," is politely rephrased according to the emotion to, "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to." This response is displayed in the device's chat window and provided to the user. The user uses this information to configure the product settings.

[1181] Examples of specific cases and prompt statements

[1182] Specific example

[1183] 1. User's question: "My printer is jammed. How can I fix it?"

[1184] 2. Server preprocessing: Extract troubleshooting guides for "paper jams" from the database and use them to train a generative AI model.

[1185] 3. Emotion Analysis: The emotion recognition engine detects "frustration" from the user's voice and facial expressions.

[1186] 4. Response generation and refinement: The generative AI model creates the procedure "turn off the power and carefully remove the paper," and the server refines the expression more carefully based on the emotion recognition results.

[1187] 5. Response Provision: The adjusted response is provided to the user by the device (for example, displayed in the chat window).

[1188] Example of a prompt

[1189] User's question: "My printer is jammed. How can I fix it?"

[1190] Sentiment Analysis: Detects irritation from input context, voice, and facial expressions.

[1191] Generative AI model's response: "Turn off the power and carefully remove the paper."

[1192] Emotion-based adjustment: "I'm sorry, but could you please turn off the printer and carefully remove the jammed paper?"

[1193] In this way, the system can respond to user inquiries efficiently and with consideration for their feelings.

[1194] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1195] Step 1:

[1196] The server registers manual information about products and services in the customer support database. Input includes manual information in PDF, text file, and image formats. The specific process for storing this information in the database involves the server receiving the manual information in various formats and saving it in the database in the appropriate format. For example, a PDF file titled "Printer Ink Replacement Procedure" might be saved in the database.

[1197] Step 2:

[1198] The server extracts the necessary manual information from the database and preprocesses it. The input is the manual information in the database. Specifically, the server cleans the text data (removing unnecessary spaces and line breaks) and converts the format (from PDF to text). The output is the preprocessed text data. For example, a PDF document titled "Initial Setup Procedure" is converted to text format.

[1199] Step 3:

[1200] The server supplies pre-processed text data to a generative AI model. The input is pre-processed text data. Specifically, the server feeds this data to a generative AI model (e.g., GPT-3) to train the model. The output is the trained generative AI model.

[1201] Step 4:

[1202] The terminal provides an interface for receiving inquiries from users. Input includes questions entered by the user via web forms or chatbots. Its specific operation is to retrieve the user's inquiry content through the user interface and send it to the server. For example, a user might enter the question, "How do I connect the product via Bluetooth?"

[1203] Step 5:

[1204] The server receives the user's inquiry and sends it to a generative AI model. The input is the inquiry sent from the terminal. Specifically, the server analyzes the inquiry and passes it to the generative AI model. The output is the answer generated by the generative AI model. For example, it might generate instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[1205] Step 6:

[1206] The device uses an emotion recognition engine to analyze the user's emotions when they make a request. Input includes the user's written text, voice, and facial expressions. The emotion recognition engine analyzes this information to detect the user's emotional state. The output is the analyzed emotional state of the user (e.g., "irritated").

[1207] Step 7:

[1208] The server adjusts the response of the generative AI model based on the results analyzed by the emotion recognition engine. The inputs are the generated response and the emotion state from the emotion recognition engine. Specifically, the server adjusts the response generated by the generative AI model according to the emotion state, changing its expression. The output is the adjusted response. For example, the response might be changed to "Sorry for the wait, but please select Bluetooth from the settings menu and then search for the device you want to connect to."

[1209] Step 8:

[1210] The server considers the user's emotional state and escalates the issue to a higher-level person in charge. The input is the emotional state, which the emotion recognition engine has determined to be "urgent." The specific action is for the server to activate the escalation engine and transfer the issue to a higher-level support person. The output is a notification to the higher-level person in charge that the issue has been escalated.

[1211] Step 9:

[1212] The server sends the response generated by the generative AI model and emotion recognition engine to the device. The input is a refined response. Specifically, the server sends this response to the device, and the device displays it to the user. The output is the response displayed to the user. For example, the device's chat window might display helpful instructions such as, "Select Bluetooth from the settings menu, then search for the device you want to connect to."

[1213] (Application Example 2)

[1214] 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".

[1215] Traditional customer support systems often resulted in low user satisfaction because they provided uniform responses without considering the user's emotional state. Furthermore, their limited functionality for providing timely and appropriate answers to inquiries made efficient support difficult. Additionally, the lack of real-time support utilizing in-store voice interfaces created a need to improve the user experience.

[1216] 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 acquiring manual information from a customer support database, means for training a generative AI model with the acquired manual information, emotion recognition engine means for analyzing the user's emotional state, means for adjusting the content and expression of the response based on the emotional state, escalation engine means for escalating the problem to a higher-level support staff member based on the user's emotional state, and voice interface means for receiving user inquiries by voice using speech recognition and providing responses by voice. This enables improved user satisfaction, the provision of quick and appropriate answers, and effective customer support through a real-time voice interface.

[1217] A "customer support database" is a database that stores information such as manuals, troubleshooting guides, and FAQs related to products and services.

[1218] "Manual information" refers to information that includes detailed explanations of how to use a product or service, setup procedures, troubleshooting, and more.

[1219] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to automatically generate appropriate answers to user inquiries.

[1220] A "user interface" is an interface through which a user inputs inquiries and interacts with the system.

[1221] An "emotion recognition engine" is an engine that analyzes the user's emotional state from their input or voice to detect emotions such as "anger" or "joy."

[1222] The "escalation engine" is an engine that, based on the user's emotional state analyzed by the emotion recognition engine, escalates the issue to a higher-level support representative as needed.

[1223] A "rapid response system" is a system that provides users with answers generated by generative AI models in real time.

[1224] A "voice interface" is an interface that uses speech recognition technology to receive user inquiries in voice and provide responses in voice.

[1225] This invention is a system for streamlining customer support in physical stores. It retrieves manual information from a customer support database and uses it to train a generative AI model, enabling it to respond quickly and accurately to user inquiries. It also incorporates an emotion recognition engine that analyzes the user's emotional state at the time of inquiry and adjusts the content and expression of the response accordingly.

[1226] The server retrieves product and service manual information (e.g., initial setup procedures and troubleshooting guides) from the customer support database. This manual information is stored in various formats, including PDFs, text files, and images. The server preprocesses this data and trains a generative AI model. This process involves cleaning and formatting the text data. For example, it might extract the initial setup procedures for a new product as text and train the generative AI model.

[1227] The terminal provides an interface for receiving user inquiries. This is done via applications installed on smartphones or robots placed within the store. Users can enter questions through web forms, chatbots, or voice interfaces. For example, if a user asks, "Do you have this product in stock?", the terminal sends this inquiry to the server.

[1228] The server sends the received inquiry to a generative AI model, which analyzes the inquiry and generates an appropriate response. For example, it might generate a response such as, "We will check the stock status. Please wait a moment." The server also uses an emotion recognition engine to analyze the user's emotions. It detects the user's emotional state (e.g., "anger") from the context of the input, voice, and facial expressions, and adjusts the content and expression of the response generated by the generative AI model based on this. If the user is angry, the response is changed to a more polite expression.

[1229] Furthermore, if the emotion recognition engine determines that the user's emotional state is "urgent," the server activates the escalation engine to escalate the issue to a higher-level support person. For example, if the user's emotions are recognized as "very angry," the server will contact the support team leader.

[1230] In summary, this invention enables improved user satisfaction, the provision of quick and appropriate responses, and effective customer support through a real-time voice interface.

[1231] Specific example

[1232] When a user asks, "Do you have this item in stock?", the speech recognition engine converts the speech into text, and the generative AI model generates the response, "We will check the stock status. Please wait a moment."

[1233] Examples of prompts to input into a generative AI model:

[1234] User: Do you have this item in stock?

[1235] assistant:

[1236] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1237] Step 1: Prepare the database

[1238] The server registers product and service manual information in the customer support database. Inputs include PDFs, text files, and images, such as initial setup instructions and troubleshooting guides for new products. The server stores this information in the database. The output is that the stored information becomes available for use.

[1239] Step 2: Training a generative AI

[1240] The server extracts necessary manual information from the database and performs text data cleaning and format conversion. The input is the manual information obtained from the database. The server supplies the pre-processed text data to a generative AI model for training. The output is that the generative AI model can generate appropriate answers to queries.

[1241] Step 3: Receiving inquiries from users

[1242] The terminal provides an interface for receiving inquiries from users. Inputs include questions entered by the user via web forms, chatbots, or voice interfaces. The terminal receives these inquiries and generates data to send to the server as output.

[1243] Step 4: Processing the inquiry

[1244] The server sends the query received from the terminal to a generative AI model. The input is the user's question. The server processes the data necessary for the generative AI model to analyze the query and generate an appropriate answer. The server receives the generated answer as output.

[1245] Step 5: Operation of the emotion recognition engine

[1246] The device uses an emotion recognition engine to analyze the user's emotions when a user makes a request. Input includes the user's context, voice, and facial expressions. The device analyzes this information and sends the user's emotional state (e.g., "anger," "frustration") to the server as output.

[1247] Step 6: Adjusting the content and wording of your answer

[1248] The server adjusts the content and expression of the responses generated by the generative AI model based on the user's emotional state analyzed by the emotion recognition engine. The input consists of the user's emotional state and the generated responses. The server adjusts these and outputs the adjusted responses.

[1249] Step 7: Operation of the escalation engine

[1250] The server activates the escalation engine when the emotion recognition engine determines that the user's emotional state is "urgent." The input is the analyzed user's emotional state. The server escalates the issue to a higher-level support person and generates an escalated issue as output.

[1251] Step 8: Provide your answer

[1252] The server sends the responses generated by the generative AI model and emotion recognition engine to the terminal. The input is the adjusted response. The terminal displays this to the user, and the output is the information provided to the user.

[1253] As a concrete example, if a user asks, "Do you have this item in stock?", the device uses a speech recognition engine to convert the speech into text and sends it to the server. The server uses a generative AI model to generate an answer and an emotion recognition engine to analyze the user's emotional state. After appropriate adjustments are made, the device provides the user with an answer such as, "I will check the stock status. Please wait a moment."

[1254] Examples of prompts to input into a generative AI model:

[1255] User: Do you have this item in stock?

[1256] assistant:

[1257] 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.

[1258] The data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">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.

[1259] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1260] 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.

[1261] 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.

[1262] 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.

[1263] 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.

[1264] 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, for example, based 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.

[1265] 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."

[1266] 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.

[1267] 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.

[1268] 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.

[1269] 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.

[1270] 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.

[1271] 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.

[1272] 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.

[1273] 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.

[1274] 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.

[1275] 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.

[1276] 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.

[1277] 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 to be incorporated by reference.

[1278] The following is further disclosed regarding the embodiments described above.

[1279] (Claim 1)

[1280] Methods for obtaining manual information from the customer support database,

[1281] A method for training a generative AI model with acquired manual information,

[1282] A user interface means for receiving inquiries from users,

[1283] A means of sending received inquiries to a generative AI model,

[1284] A means of receiving and providing answers generated by a generative AI model to the user,

[1285] A system that includes this.

[1286] (Claim 2)

[1287] The system according to claim 1, which generates an answer to a user by searching for the most suitable information from a database in response to a user inquiry and supplying it to a generative AI model.

[1288] (Claim 3)

[1289] The system according to claim 1, comprising a rapid response system for providing a user with an answer generated by a generative AI model in real time.

[1290] "Example 1"

[1291] (Claim 1)

[1292] Means of obtaining information from the customer support database,

[1293] A means of preprocessing and formatting acquired information to train a generative AI model,

[1294] An interface for receiving inquiries from users,

[1295] A means of sending received inquiries to a generative AI model via a server,

[1296] A means of receiving and providing answers generated by a generative AI model to the user,

[1297] A system that includes this.

[1298] (Claim 2)

[1299] The system according to claim 1, comprising means for searching for the most suitable information from a customer support database in response to a user inquiry, and performing preprocessing and format conversion for supplying it to a generative AI model.

[1300] (Claim 3)

[1301] The system according to claim 1, comprising a rapid response system for providing a user with a response generated by a generative AI model in real time.

[1302] "Application Example 1"

[1303] (Claim 1)

[1304] Methods for obtaining manual information from the customer support database,

[1305] A method for training a generative AI model with acquired manual information,

[1306] A user interface means for receiving inquiries from users,

[1307] A means of sending received inquiries to a generative AI model,

[1308] A means of receiving and providing answers generated by a generative AI model to the user,

[1309] A means for users to ask questions about products (by text or voice input),

[1310] A system that includes this.

[1311] (Claim 2)

[1312] The system according to claim 1, which generates an answer to a user by searching for the most suitable information from a database in response to a user inquiry and supplying it to a generative AI model.

[1313] (Claim 3)

[1314] The system according to claim 1, comprising a rapid response system for providing a user with an answer generated by a generative AI model in real time.

[1315] "Example 2 of combining an emotion engine"

[1316] (Claim 1)

[1317] Methods for obtaining manual information from the customer support database,

[1318] A method for training a generative AI model with acquired manual information,

[1319] A user interface means for receiving inquiries from users,

[1320] A means of sending received inquiries to a generative AI model,

[1321] A means of receiving and providing answers generated by a generative AI model to the user,

[1322] An emotion recognition engine means for analyzing user emotions,

[1323] A means for adjusting the response of a generative AI model based on the results analyzed by the emotion recognition engine,

[1324] Taking into account the user's emotional state, a means to escalate the issue to a higher-level person when necessary,

[1325] A system that includes this.

[1326] (Claim 2)

[1327] The system according to claim 1, which generates an answer to a user by searching for the most suitable information from a database in response to a user inquiry and supplying it to a generative AI model.

[1328] (Claim 3)

[1329] The system according to claim 1, comprising a rapid response system for providing a user with an answer generated by a generative AI model in real time.

[1330] "Application example 2 when combining with an emotional engine"

[1331] (Claim 1)

[1332] Methods for obtaining manual information from the customer support database,

[1333] A method for training a generative AI model with acquired manual information,

[1334] A user interface means for receiving inquiries from users,

[1335] A means of sending received inquiries to a generative AI model,

[1336] A means of receiving and providing answers generated by a generative AI model to the user,

[1337] A means (emotion recognition engine) for analyzing the user's emotional state,

[1338] A means of adjusting the content and expression of responses based on emotional state,

[1339] A means (escalation engine) for escalating issues to higher-level support personnel based on the user's emotional state,

[1340] A system that includes this.

[1341] (Claim 2)

[1342] The system according to claim 1, which generates an answer to a user by searching for the most suitable information from a database in response to a user inquiry and supplying it to a generative AI model.

[1343] (Claim 3)

[1344] A rapid response system for providing users with answers generated by generative AI models in real time,

[1345] The system according to claim 1, comprising a voice interface means for receiving user inquiries by voice using speech recognition and providing responses by voice. [Explanation of symbols]

[1346] 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. Methods for obtaining manual information from the customer support database, A method for training a generative AI model with acquired manual information, A user interface means for receiving inquiries from users, A means of sending received inquiries to a generative AI model, A means of receiving and providing answers generated by a generative AI model to the user, A system that includes this.

2. The system according to claim 1, which generates an answer to a user by searching for the most suitable information from a database in response to a user inquiry and supplying it to a generative AI model.

3. The system according to claim 1, comprising a rapid response system for providing a user with an answer generated by a generative AI model in real time.

Images