System

Abstract

A system is provided.SOLUTION: A system comprising: means for using generative AI to automate customer instruction and basic query responses; means for initializing and managing a generative AI engine; means for collecting and displaying multimedia materials; means for receiving and analyzing input from a user; and means for displaying or playing back responses to a user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In addition to signing contracts, mobile phone shops also spend a lot of time explaining how to use and operate mobile phones. Detailed explanations are especially necessary for customers who are unfamiliar with digital technology, placing a significant burden on store staff. However, with the worsening labor shortage, it is difficult to provide sufficient explanations, and time is often wasted on tasks that are not directly related to sales. Therefore, there is a need for a method to efficiently provide instructions to customers and reduce the burden on store staff. [Means for solving the problem]

[0005] To address this issue, the present invention provides a system that uses generative AI to automate customer operation instructions and responses to basic questions. Specifically, the system includes a means for initializing and managing the generative AI engine, a means for collecting and displaying multimedia materials, a means for receiving and analyzing input data from users, and a means for displaying or playing responses to users. The system also has a self-learning function and is equipped with a means for updating the generative AI engine model based on past questions and feedback, thereby continuously improving response quality. Furthermore, the system provides a chat support function that responds to customer questions in real time, enabling fast and efficient support. This solves problems such as staff shortages and reduced efficiency, and improves customer satisfaction.

[0006] "Generative AI" is an artificial intelligence technology that analyzes input data from users and automatically generates appropriate responses and operating instructions.

[0007] A "generative AI engine" is a software and hardware component that executes generative AI processing.

[0008] "Means to initialize and manage" refers to the functionality for bringing the generative AI engine online and performing the required settings and configuration.

[0009] "Multimedia material" is data that contains multiple media formats, such as images, video, audio, and text.

[0010] "Means for receiving and analyzing input data from a user" refers to a function for receiving voice or text data input by a user and analyzing the intent of the data.

[0011] "Means for displaying or playing back a response" refers to a function for providing the generated response to the user visually or audibly.

[0012] The "self-learning function" is a feature that improves the quality of responses by learning from past questions and feedback and updating the generative AI model.

[0013] The "chat support function" is a function that responds to user questions in real time in a text-based interactive format. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0022] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0035] Below, we will describe a specific embodiment of a customer support system that utilizes generative AI.

[0036] First, the overall system configuration consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and is responsible for processing requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[0037] Basic operation of the system

[0038] 1. System startup and initialization

[0039] Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine, and performs the necessary settings and configuration.

[0040] 2. Accepting user input

[0041] Terminal: Displays an interface that allows speech recognition and text input. Users can enter operation instructions and questions by voice or text.

[0042] Device: If the input is voice, it is converted into text and the input data is sent to the server.

[0043] 3. Question Analysis and Answer Generation

[0044] Server: Analyzes the text data received from the device. The generative AI engine identifies the intent of the question and generates appropriate responses and operating instructions.

[0045] Server: Optionally, add multimedia material such as images or videos to the response.

[0046] 4. Providing a Response

[0047] Server: Sends the generated response and multimedia material to the terminal.

[0048] Terminal: Displays the received response to the user and plays the audio prompt, if available.

[0049] 5. Gather feedback and learn

[0050] User: Can provide feedback on the response.

[0051] Device: Sends user feedback to the server.

[0052] Server: Analyzes the received feedback and updates the generative AI engine's model, thereby improving the quality of future responses.

[0053] Specific examples

[0054] Scenario 1: Senior user learning how to use a smartphone camera

[0055] 1. User Input:

[0056] User: Uses the voice recognition feature on their smartphone to ask, "How do I use the camera?"

[0057] Terminal: Converts voice data into text and sends it to the server.

[0058] 2. Question analysis and answer generation:

[0059] Server: Analyzes the text "How do I use the camera?" and determines that it should explain how to use the camera app.

[0060] Server: Generates a detailed guide that explains how to start the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the camera app's interface.

[0061] 3. Providing a response:

[0062] Server: Sends the generated operation guide and multimedia materials to the terminal.

[0063] Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[0064] 4. Gathering Feedback:

[0065] Users: Rate their satisfaction with the guide and enter follow-up questions if needed.

[0066] Device: Sends user feedback to the server.

[0067] Server: Analyzes the feedback and updates the generative AI engine model.

[0068] In this way, the present invention provides a customer support system using generative AI, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality.

[0069] The processing flow will be explained below.

[0070] Step 1:

[0071] System startup and initialization

[0072] Terminal: Starts the system and loads libraries and required modules.

[0073] Terminal: Initialize the generation AI engine and perform the necessary settings.

[0074] Step 2:

[0075] Accepting user input

[0076] Terminal: displays the interface and allows the user to input by voice or text.

[0077] User: Asks a question or requests a command by voice or text.

[0078] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0079] Step 3:

[0080] Data reception and analysis

[0081] Server: Receives text data sent from the device.

[0082] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0083] Step 4:

[0084] Response Generation

[0085] Server: Generates appropriate responses and instructions based on the intent of the question. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app and take a photo.

[0086] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0087] Step 5:

[0088] Sending a Response

[0089] Server: Sends the generated response and multimedia material to the terminal.

[0090] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[0091] Step 6:

[0092] Collecting user feedback

[0093] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[0094] Terminal: Receives user feedback and sends it to the server.

[0095] Step 7:

[0096] Feedback analysis and model updating

[0097] Server: Analyzes the received feedback and updates the generative AI engine's model, making the next response more accurate and relevant.

[0098] This specific process flow allows the customer support system to efficiently provide instructions to customers, reduce the burden on store staff, and continuously improve the system based on feedback.

[0099] Example 1

[0100] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0101] Conventional customer support systems require a significant amount of human resources to provide operational instructions and respond to basic questions, making them particularly difficult to respond to during the initial learning phase when multiple questions frequently arise, or when product problems occur. Furthermore, conventional systems make it difficult to ensure the accuracy and consistency of responses, potentially leading to a decline in customer satisfaction. Furthermore, they are unable to effectively utilize feedback, which delays the continuous improvement of the system.

[0102] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0103] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for converting voice input from a user into text, means for analyzing the text and generating appropriate responses, means for providing the generated responses and multimedia materials to the user, and means for collecting feedback from users and updating the model of the generative AI engine. This makes it possible to quickly and accurately respond to a variety of questions from customers and continuously improve the system's response quality by utilizing the feedback.

[0104] "Generative AI" refers to a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[0105] "Generative AI engine" refers to the software components and hardware environment for operating and managing generative AI.

[0106] "Multimedia material" refers to data that contains multiple forms of digital content, such as images, video, and audio.

[0107] "Users" refers to people who use the system and provide instructions on how to use it or respond to questions.

[0108] "Feedback" refers to user responses and opinions and evaluations of system performance.

[0109] "Voice input" refers to the voice spoken by the user through a microphone.

[0110] "Text-to-text" refers to the process of converting voice input into written information.

[0111] "Analysis" refers to the process of structurally understanding and processing text data and feedback.

[0112] "Response" refers to information or instructions generated by a generative AI engine and provided to the user.

[0113] "Model updating" refers to the process of improving and adjusting the algorithms and parameters of a generative AI engine based on new data and feedback.

[0114] A specific embodiment of the present invention will be described below. A system for realizing the present invention comprises three elements: a server, a terminal, and a user.

[0115] The server runs a generative AI engine, analyzes user input data sent from the device, and generates a response. This generative AI engine can be based on commonly used artificial intelligence technologies (e.g., OpenAI's GPT-3). Furthermore, the server collects feedback from users and updates the generative AI engine's model to improve the quality of the response.

[0116] The terminal accepts user input data, communicates with the server, and ultimately presents a response to the user. Terminals are equipped with speech recognition and text input functions, and are devices such as smartphones and tablets. Google Cloud Speech-to-Text and other voice recognition APIs are used.

[0117] The user inputs questions or requests to the system through the terminal interface. The input method can be voice input or text input, depending on the user's preference. For example, the user may input "Please tell me how to use the camera" by voice.

[0118] The server converts the voice data received from the device into text using a speech recognition API, and then analyzes the text data using a generative AI engine. Generative AI such as GPT-3 generates an appropriate response based on the analysis results. Responses may include not only text information, but also multimedia materials such as images and videos to supplement the operating instructions. These multimedia materials are collected and managed on the server as needed.

[0119] The terminal receives the response and multimedia material sent from the server and displays them to the user. For example, it displays text on the smartphone screen and plays a video showing the operation procedure. This allows the user to intuitively and visually understand how to operate the device.

[0120] The user inputs feedback on the response and sends it from their device to the server. The server analyzes this feedback and updates the generative AI engine's model to improve the quality of the response from the next time onwards. In this way, the system has the ability to self-learn and continuously improve its performance.

[0121] Specific examples

[0122] Scenario 1: Senior user learning how to use a smartphone camera

[0123] 1. User input: The user speaks to their smartphone, asking, "How do I use the camera?"

[0124] 2. Server analysis: The voice data received from the device is converted into text, and the text "Please tell me how to use the camera" is analyzed using a generative AI engine (e.g., GPT-3).

[0125] 3. Response generation: Based on the analysis results, the server generates a text response that specifically explains how to use the camera app. For example, it generates a response such as "Open the camera app and press the center shutter button." It also includes an image showing the camera app's interface and a video explaining the operation procedure.

[0126] 4. Response provision: The response generated by the server and the multimedia material are sent to the terminal.

[0127] 5. User response confirmation: The device displays text, images, and videos to the user, and also plays audio instructions explaining how to operate the device.

[0128] This invention is implemented in the above-described manner, and a customer support system using generative AI can respond to user questions quickly and accurately, and utilize feedback to continuously improve the system's response quality.

[0129] Prompt Sentence Examples

[0130] "Please tell me how to use the camera. Specifically, please explain how to start the camera app, take a photo, and zoom in and out."

[0131] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0132] Step 1:

[0133] System startup and initialization

[0134] Device: When booting the system, the device first loads the necessary libraries and modules, specifically initializing the software components that run the speech recognition and generative AI engines (e.g., Python's tensorflow and numpy libraries).

[0135] Terminal: Next, the terminal establishes a network connection with the server and loads the configuration file (e.g., config.json), which completes the basic configuration of the system.

[0136] Input: The action the user takes to start the system.

[0137] Output: The required libraries and modules are loaded and a connection to the server is established.

[0138] Step 2:

[0139] Accepting user input

[0140] Terminal: A screen that allows voice recognition and text input is displayed through the user interface. The user inputs "How do I use the camera?" into the interface by voice or text.

[0141] Device: For voice input, use a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data.

[0142] Input: User voice or text input.

[0143] Output: User question data converted to text.

[0144] Step 3:

[0145] Question analysis and answer generation

[0146] Server: Receives text data received from the device and performs grammatical analysis of the text using a natural language processing library (e.g., spaCy). Then, a generative AI engine (e.g., OpenAI GPT-3) generates an appropriate response based on the text.

[0147] Server: In some cases, the generated response may include multimedia materials such as images or videos. For example, a screenshot of the camera app interface may be attached to the text response "Open the camera app and press the center shutter button."

[0148] Input: User question data converted to text.

[0149] Output: Appropriate response text and necessary multimedia materials.

[0150] Step 4:

[0151] Providing a response

[0152] Server: Sends the generated response and multimedia material to the terminal. The response is often packaged in JSON format or similar.

[0153] Terminal: The terminal displays the response received from the server to the user. Specifically, it displays text in the interface, plays images and videos, and outputs audio guidance, if available.

[0154] Input: The response data sent by the server.

[0155] Output: The response content (text, image, video) that is provided to the user.

[0156] Step 5:

[0157] Gathering feedback and learning

[0158] Users: Enter feedback on the response, adding comments such as "This response was helpful" or "Please explain in more detail."

[0159] Terminal: Sends the feedback data entered by the user to the server.

[0160] Server: Analyzes the received feedback and updates the generative AI engine model. Specifically, the feedback data is stored in a database and the model is trained based on that data to improve the quality of future responses.

[0161] Input: Feedback data from users.

[0162] Output: Feedback data parsed and stored, and an updated generative AI model.

[0163] (Application example 1)

[0164] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0165] Efficiently obtaining product information and operation instructions in physical stores often places a heavy burden on store staff. Providing fast and accurate support is difficult, especially during busy times. Furthermore, there is a lack of efficient customer support systems that utilize voice recognition and generative AI, raising concerns about a decline in customer satisfaction.

[0166] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0167] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from a user, means for displaying or playing responses to the user, means for converting voice input from the user into text data using a speech recognition function, means for using a generative AI model to generate optimal responses to user questions, and means for providing responses to the user in real time using a smartphone. This allows customers to quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[0168] "Generative AI" is an artificial intelligence technology that uses generative models to automatically generate content such as natural language text and images.

[0169] The "generative AI engine" is the core part of the system that runs the generative AI and generates responses to requests from users.

[0170] "Multimedia material" is digital content that includes multiple media formats such as images, video, and audio.

[0171] "Speech recognition" is a technology that analyzes voice input and converts it into text data.

[0172] A "smartphone" is a mobile device that has advanced computing power and Internet connection capabilities in addition to telephone functions.

[0173] "Self-learning" is a feature that allows a system to improve its own performance based on past data and feedback.

[0174] The "chat support function" is a function that responds to user questions in real time in a text-based chat format.

[0175] "Product placement information" is information relating to the location and placement of products within a store.

[0176] A specific embodiment of a system for realizing this application example will be described below.

[0177] Overall system configuration

[0178] This system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and processes requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[0179] Hardware and software used

[0180] Hardware: built-in microphone on smartphone, smartphone itself

[0181] software:

[0182] Speech recognition: speech_recognition

[0183] Generative AI model: HuggingFace Transformers library, specifically the bert-large-uncased-whole-word-masking-finetuned-squad model

[0184] Data processing and calculation

[0185] 1. Accepting voice input

[0186] The device uses a speech recognition module to collect voice input from the user and convert it into text data, allowing the user to voice questions or requests.

[0187] 2. Speech-to-text

[0188] The converted text data is sent to a generative AI model for analysis and response generation.

[0189] 3. Question Analysis and Response Generation Using Generative AI

[0190] The server analyzes the intent of the user's question based on the received text data and generates the optimal response. The generative AI model is responsible for this analysis and response generation.

[0191] 4. Providing a Response

[0192] The generated response is sent from the server to the terminal and displayed or played aloud to the user by the terminal, thereby providing the user with an immediate answer.

[0193] Examples of concrete examples and prompts

[0194] As a concrete example, consider the case where a user asks "Where is this item?" in a physical store. This question is handled as follows:

[0195] Examples:

[0196] The user uses their smartphone to voice-input "Where is this product?" The speech is converted into text, which is analyzed by the server's generative AI engine, which generates the optimal response based on product location information within the store. Finally, this response is presented to the user.

[0197] Example prompt sentence:

[0198] "Where is this item?"

[0199] "Food is on the first floor, and electrical appliances are on the second floor. Specifically, the TV is on the left at the back of the second floor."

[0200] In this way, customers can quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[0201] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0202] Step 1:

[0203] The device initializes the voice recognition module and waits for voice input. When the user speaks a question, the device collects the voice data through the microphone. The input is the voice data, and the output is the analysis of the voice data by the voice recognition module.

[0204] Step 2:

[0205] The collected voice data is converted into text data through the device's voice recognition module. In this conversion process, the voice recognition algorithm analyzes each phoneme from the user's voice and generates the corresponding text. The input is voice data, and the output is text data generated based on that voice.

[0206] Step 3:

[0207] The device sends the converted text data to the server, which then passes the received text data to the generative AI model to analyze the question and generate a response. The input is the text data sent from the device, and the output is the analysis result and the generated response.

[0208] Step 4:

[0209] The generative AI model analyzes the intent of the user's question based on the input text data and generates the optimal response. In this process, the generative AI model utilizes knowledge it has previously learned to analyze the text data. The input is text data, and the output is the generated response text.

[0210] Step 5:

[0211] The server sends the generated response text to the terminal, which receives the response text and prepares it to present to the user. The input is the response text sent from the server, and the output is data prepared by the terminal for display or audio playback.

[0212] Step 6:

[0213] The terminal displays the received response text to the user. Specifically, it displays the response as text on the smartphone screen or plays the response as audio using the audio playback function. The input is the response text, and the output is the display or audio data presented to the user.

[0214] Step 7:

[0215] The user can input feedback for the presented response through the terminal. The terminal sends this feedback as text data to the server. The input is the feedback (text data) from the user, and the output is the feedback data sent to the server.

[0216] Step 8:

[0217] The server analyzes the received feedback data and trains the generative AI model to self-train. In this process, the model's parameters are adjusted based on the feedback data to improve the accuracy of future responses. The input is the feedback data, and the output is an updated generative AI model.

[0218] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0219] This invention provides a customer support system that combines generative AI and an emotion engine. The system consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and emotion engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions and requests into the system through the terminal interface.

[0220] Basic operation of the system

[0221] 1. System startup and initialization

[0222] - Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine and emotion engine, and performs the necessary configuration.

[0223] 2. Accepting user input

[0224] - Terminal: displays the interface and allows the user to input by voice or text.

[0225] - User: Ask a question or request an action by voice or text.

[0226] - Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0227] 3. Data Reception and Analysis

[0228] - Server: Receives text data sent from the device.

[0229] - Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0230] - Server: Analyzes user emotions using an emotion engine and adjusts the tone and content of responses.

[0231] 4. Response Generation

[0232] - Server: Generates appropriate responses and instructions based on the intent of the question and the results of emotion analysis. For example, if the question is "Please tell me how to use the camera," the server generates instructions such as how to launch the camera app and how to take a photo, along with a tone that reflects the user's emotions.

[0233] - Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0234] 5. Providing a Response

[0235] - Server: Sends the generated response and multimedia material to the terminal.

[0236] - Terminal: Displays the received response to the user, and plays the audio prompt if available.

[0237] 6. Collecting User Feedback

[0238] - User: Enter feedback on the provided response (e.g. satisfaction, additional questions, etc.).

[0239] - Terminal: Receives user feedback and sends it to the server.

[0240] 7. Feedback analysis and model updating

[0241] - Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[0242] Specific examples

[0243] Scenario 1: Senior user learning how to use a smartphone camera

[0244] 1. User Input:

[0245] - User: Use your smartphone's voice recognition function to ask, "How do I use the camera?"

[0246] - Terminal: Converts voice data into text and sends it to the server.

[0247] 2. Data reception and analysis:

[0248] - Server: Analyzes the text "Please tell me how to use the camera" and determines that it should explain how to use the camera app.

[0249] - Server: Uses an emotion engine to recognize the user's current emotional state (e.g., anxiety, confusion, anticipation).

[0250] 3. Response generation:

[0251] - Server: Generates detailed instructions on how to launch the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the interface of the camera app.

[0252] - Server: Based on the analysis results of the emotion engine, generate a response in a tone that corresponds to the user's emotion (e.g., a calm tone, an encouraging tone).

[0253] 4. Providing a response:

[0254] - Server: Sends the generated operation guide and multimedia materials to the terminal.

[0255] - Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[0256] 5. Collecting User Feedback:

[0257] - User: Rate your satisfaction with the guide and enter follow-up questions if necessary.

[0258] - Device: Sends user feedback to the server.

[0259] - Server: Analyzes the feedback and updates the models of the generative AI engine and emotion engine.

[0260] In this way, the present invention provides a customer support system using generative AI and an emotion engine, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality. The emotion engine enables responses based on the user's emotions, further increasing customer satisfaction.

[0261] The processing flow will be explained below.

[0262] Step 1:

[0263] System startup and initialization

[0264] Terminal: Starts the system and loads libraries and required modules.

[0265] Terminal: Initialize the generative AI engine and emotion engine and perform the necessary settings.

[0266] Step 2:

[0267] Accepting user input

[0268] Terminal: Displays an interface that allows voice recognition and text input.

[0269] User: Asks a question or requests a command by voice or text.

[0270] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0271] Step 3:

[0272] Data reception and analysis

[0273] Server: Receives text data sent from the device.

[0274] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0275] Server: Analyzes the user's emotions using an emotion engine and identifies the user's emotional state (anxiety, confusion, expectation, etc.).

[0276] Step 4:

[0277] Response Generation

[0278] Server: Generates appropriate responses and instructions based on the intent of the question and the results of sentiment analysis. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app, how to take a photo, etc.

[0279] Server: Adjust the tone of your response depending on the user's emotional state. For example, if the user is feeling anxious, explain things in a calmer tone.

[0280] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0281] Step 5:

[0282] Sending a Response

[0283] Server: Sends the generated response and multimedia material to the terminal.

[0284] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[0285] Step 6:

[0286] Collecting user feedback

[0287] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[0288] Terminal: Receives user feedback and sends it to the server.

[0289] Step 7:

[0290] Feedback analysis and model updating

[0291] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[0292] Specific examples

[0293] Scenario 1: Senior user learning how to use a smartphone camera

[0294] Step 1:

[0295] System startup and initialization

[0296] Device: Start the smartphone system and initialize the generative AI engine and emotion engine.

[0297] Step 2:

[0298] Accepting user input

[0299] Terminal: Displays a voice recognition interface and an interface for the user to request instructions.

[0300] User: Ask aloud, "How do I use the camera?"

[0301] Terminal: Converts voice data into text and sends it to the server.

[0302] Step 3:

[0303] Data reception and analysis

[0304] Server: Receives text data saying "Please tell me how to use the camera."

[0305] Server: A generative AI engine analyzes the text data and identifies that the user is asking about how to use the camera app.

[0306] Server: The emotion engine identifies emotions such as anxiety and confusion from the user's voice.

[0307] Step 4:

[0308] Response Generation

[0309] Server: Generates a guide including how to launch the camera app, take a photo, and zoom in and out. For example, generate instructions such as "Open the camera app and press the center shutter button."

[0310] Server: Use a calming tone to explain things to the user to ease their concerns.

[0311] Server: Add images and videos showing the camera app interface.

[0312] Step 5:

[0313] Sending a Response

[0314] Server: Sends the generated explanation and multimedia materials to the terminal.

[0315] Device: Displays images and videos along with text instructions, and also provides audio instructions.

[0316] Step 6:

[0317] Collecting user feedback

[0318] Users: Leave feedback about how satisfied you are with the guide and any follow-up questions you may have.

[0319] Device: Sends user feedback to the server.

[0320] Step 7:

[0321] Feedback analysis and model updating

[0322] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, adjusting the next response to be more accurate and appropriate.

[0323] With this specific processing flow, a customer support system that utilizes generative AI and an emotion engine can efficiently provide customers with operating instructions, reduce the burden on store staff, and improve customer satisfaction by responding according to the user's emotions.

[0324] Example 2

[0325] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0326] Conventional customer support systems rely on human intervention to respond to customer questions and provide operational instructions, which tends to result in long response times and reduced customer satisfaction. Furthermore, the response content is fixed, making it difficult to respond appropriately to individual customers' emotions and situations. Furthermore, the system lacks a self-learning function and cannot utilize past feedback, making it difficult to improve system performance.

[0327] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0328] In this invention, the server includes a means for automating customer operation instructions and responses to basic questions using a generative AI, a means for initializing and managing the generative AI engine and the emotion analysis engine, a means for converting voice data to text and receiving and analyzing input data, a means for analyzing the user's emotional state and adjusting the response content, a means for generating a response based on the intent of the question and the results of the emotion analysis, a means for collecting and displaying multimedia materials, a means for displaying or playing the response to the user, and a means for receiving and analyzing user feedback and updating the generative AI engine model. This enables the server to provide prompt and appropriate responses to customer questions and respond according to the emotions and circumstances of each individual customer. Furthermore, the system's performance can be continuously improved based on past feedback.

[0329] "Generative AI" is a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[0330] A "generative AI engine" is software or a platform for implementing generative AI functions, specifically text generation and image generation using large-scale neural network models.

[0331] An "emotion analysis engine" is software or a system that analyzes user input data and identifies the user's emotional state, allowing the tone and content of responses to be tailored to the user's emotions.

[0332] "Multimedia materials" is a general term for digital content that includes various media formats such as images, videos, and audio. By using these, information provided to users can be enriched visually and aurally.

[0333] "Self-learning" is the ability of a system to improve its performance based on past data and feedback, allowing it to provide increasingly accurate and appropriate responses over time.

[0334] "Feedback" refers to input data such as user ratings of responses and follow-up questions, which are used to improve the system.

[0335] "Chat Support" refers to the overall interface and system for interacting with customers in real time, allowing for immediate answers to customer questions.

[0336] This invention relates to a customer support system that combines generative AI and a sentiment analysis engine. The system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and a sentiment analysis engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions or requests into the system through the terminal interface.

[0337] System configuration

[0338] 1. Hardware and Software Configuration

[0339] Server: The server is a powerful computer running the Linux operating system. A programming language such as Python is used to run the generative AI engine and sentiment analysis engine. Deep learning frameworks such as TensorFlow and PyTorch are used.

[0340] Device: A device is a device such as a smartphone, tablet, or PC. A web interface using HTML / CSS / JavaScript or an Android / iOS app runs on the device.

[0341] User: The user operates a terminal to input questions into the system and receive responses.

[0342] System Operation

[0343] 2. Data processing and calculation

[0344] Accepting user input:

[0345] Terminal: The terminal displays an HTML interface and allows users to enter questions by voice or text, converting the voice data to text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[0346] Data reception and analysis:

[0347] Server: Receives text data sent from the device, analyzes the text data using a generative AI engine (e.g., GPT-3), and identifies the intent of the question or request. Analyzes the user's emotions using a sentiment analysis engine and adjusts the response accordingly.

[0348] Response generation:

[0349] Server: Generates appropriate responses based on the intent of the question and the results of sentiment analysis. Selects and adds multimedia materials such as images and videos as needed.

[0350] Providing a response:

[0351] Server: Sends the generated response and multimedia material to the terminal, which displays the received response to the user and plays audio guidance, if any.

[0352] Collecting and analyzing user feedback:

[0353] Terminal: Receives user feedback and sends it to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine.

[0354] Specific examples

[0355] Prompt Sentence Examples

[0356] Prompt 1:

[0357] "How do I use the camera?" the user asks. This is the first time they've used a smartphone camera, and they seem nervous. Use a calming tone to provide step-by-step instructions on how to launch the camera app and take a photo.

[0358] The above is an embodiment of the present invention. This system enables quick and appropriate responses to customer questions, responding to individual customer emotions and situations. It also allows the system's performance to be continuously improved based on past feedback.

[0359] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0360] Step 1: System startup and initialization

[0361] Terminal: Starts the system and loads the necessary libraries and modules, such as Python, HTML, JavaScript, etc. During this process, the generative AI engine and sentiment analysis engine are also initialized, and API keys and configuration files are loaded.

[0362] Input: System startup command.

[0363] Output: Initialized generative AI engine and sentiment analysis engine.

[0364] Step 2: Accepting User Input

[0365] Terminal: Displays the user interface and allows the user to enter questions by voice or text, converting the voice data into text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[0366] User: Using the device interface, ask a question by voice or text, such as "How do I use the camera?"

[0367] Input: User's voice or text question.

[0368] Output: User question data converted to text.

[0369] Step 3: Data reception and analysis

[0370] Server: Receives text data sent from the device. Based on this data, it uses a generative AI engine (e.g., GPT-3) to analyze the intent of the question. It also uses an emotion analysis engine to analyze the user's emotions and adjust the response accordingly.

[0371] Input: Textualized user question data.

[0372] Output: Question intent and sentiment analysis results.

[0373] Step 4: Response Generation

[0374] Server: Generates an appropriate response based on the intent of the question and the results of sentiment analysis. For example, it generates specific instructions such as "Open the camera app and press the center shutter button." It also selects multimedia materials such as images and videos as needed.

[0375] Input: Question intent and sentiment analysis results.

[0376] Output: Generated response text and multimedia materials.

[0377] Step 5: Providing a response

[0378] Server: Sends the generated response and multimedia material to the terminal.

[0379] Terminal: Uses a speech synthesis engine to display the received response to the user and play audio prompts, if any.

[0380] Input: Generated response text and multimedia material.

[0381] Output: The response displayed to the user and the audio prompt played.

[0382] Step 6: Gather user feedback

[0383] User: Enter feedback on the response provided, for example, a satisfaction rating or a follow-up question.

[0384] Terminal: Receives user feedback and sends it to the server.

[0385] Input: User feedback or follow-up questions.

[0386] Output: Feedback data.

[0387] Step 7: Feedback analysis and model updating

[0388] Server: Analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine based on past questions and feedback.

[0389] Input: Feedback data.

[0390] Output: Updated generative AI engine and sentiment analysis engine.

[0391] The above is the specific processing flow of the system program and the specific operations performed at each step.

[0392] (Application example 2)

[0393] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0394] Conventional customer support systems have the problem that they respond mechanically to user questions and requests and are unable to respond in a way that reflects the user's emotions. Another problem is that a lack of visual guidance makes it difficult to provide efficient support when customers navigate in a store or search for products. This invention aims to provide a customer support system that can analyze a user's emotions and provide responses in real time with appropriate tone and content, as well as to improve the efficiency of in-store support by using visual guidance with smart glasses.

[0395] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automating operation instructions for customers and responses to basic questions using a generation AI, means for initializing and managing the generation AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from the user, means for displaying or playing responses to the user, means for analyzing the user's emotions and adjusting the tone and content of the responses, and means for providing visual guidance via the smart glasses. This allows the user to receive an appropriate response in a tone that corresponds to their emotions, while also allowing them to receive efficient support in the store through visual guidance.

[0396] "Generative AI" is an artificial intelligence technology that analyzes text data and generates appropriate answers and content.

[0397] A "customer" is a user of the system who inputs questions and requests.

[0398] "Instructions" refers to providing detailed instructions or guidance on the functionality and use of a system or device.

[0399] A "generative AI engine" is a software component that actually operates generative AI technology and performs analysis and response generation.

[0400] An "emotion analysis engine" is a software component that analyzes emotions from user input data and identifies their state.

[0401] "Tone" refers to the tone or expression of a response that corresponds to the user's emotion.

[0402] "Multimedia material" is data that includes information in a variety of formats, such as text, images, audio, and video.

[0403] "Input data" refers to the content of questions or requests entered by the user via voice or text.

[0404] "Smart glasses" are portable display devices capable of displaying visual information.

[0405] "Visual guide" refers to guide information that visually indicates product locations and directions within a store.

[0406] The "server" is a computer that runs the generative AI engine, emotion analysis engine, and related modules, and manages and operates the entire system.

[0407] "Real time" refers to immediate response or processing in response to user input.

[0408] A "store" refers to a physical location where users (customers) can actually visit to shop or obtain information.

[0409] "Response" means an answer or instruction provided by the system in response to a user's question or request.

[0410] "Self-learning" refers to the system's ability to automatically improve its performance based on past questions and feedback.

[0411] This invention provides a customer support system that combines generative AI and an emotion analysis engine. The system allows users to wear smart glasses and provides instant, appropriate responses and visual guidance when they ask questions or receive directions in a physical store.

[0412] System Configuration

[0413] The system consists of a server, a terminal, and a user. The server runs a generative AI engine and an emotion analysis engine, and the terminal (smart glasses) receives input from the user. The user inputs questions and requests into the system via the smart glasses.

[0414] Initialization and Management

[0415] The server initializes the generative AI engine and sentiment analysis engine and configures them appropriately. The server also updates the generative AI engine's model based on user feedback and past questions, enabling it to self-learn.

[0416] Voice Recognition

[0417] User input in the form of voice is collected as voice data by the device (smart glasses) and sent to the server, which then converts the voice data into text data using a speech recognition library. The speech_recognition library is used for this purpose.

[0418] Intention and emotion analysis

[0419] The server analyzes the received text data through a generative AI model (OpenAI's text-davinci-003) to identify the intent of the question, and uses a sentiment analysis engine to analyze the user's emotional state, which determines the tone of the response.

[0420] Response Generation

[0421] The generative AI engine generates appropriate responses based on the identified intent and the results of sentiment analysis. For example, in response to the question, "Where is the red sweater?", the generative AI engine generates a specific response such as, "The red sweater is in the women's clothing department on the second floor," and adjusts the tone accordingly.

[0422] Providing a visual guide

[0423] In addition to the generated textual response, the device provides visual guidance information, such as a store map and location guide. Visual guidance to assist with store navigation is provided using the MapDisplay module.

[0424] Specific examples

[0425] For example, a user puts on smart glasses and asks, "Where is the red sweater?" The voice data is converted to text through a speech recognition library and sent to a server. The server uses a generative AI model to analyze the intent and determine that directions to the red sweater should be provided. At the same time, it uses an emotion analysis engine to recognize the user's current emotional state and generate a response in a calm tone.

[0426] An example prompt for this would be:

[0427] Analyze the user's question to identify intent and sentiment and generate a response: Question: Where is the red sweater?

[0428] The generated response is accompanied by visual information, such as "The red sweater is in the women's clothing section on the second floor." As a visual guide, a route from the current location to the second floor is displayed on a map of the store. This series of processes allows the user to receive an appropriate response based on their emotions while also providing visual support for in-store navigation, resulting in comfortable and efficient customer support.

[0429] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0430] Step 1:

[0431] The terminal (smart glasses) collects questions entered by the user through voice, and this voice data becomes the input.

[0432] Step 2:

[0433] The device uses a speech recognition library (speech_recognition) to convert voice data into text data. During this conversion process, the device analyzes the voice waveform data and outputs the corresponding text.

[0434] Step 3:

[0435] The terminal sends the converted text data to the server, which receives the data and proceeds to the next analysis step.

[0436] Step 4:

[0437] The server uses a generative AI model (OpenAI's text-davinci-003) to analyze the text data and identify the intent of the user's question. It uses a prompt sentence for the analysis and outputs a model for generating an appropriate response from the user's question. An example of a prompt sentence in this case is, "Analyze the user's question to identify the intent and sentiment, and generate a response. Question: Where is the red sweater?"

[0438] Step 5:

[0439] The server uses an emotion analysis engine to analyze emotions from the user's text data, and the emotion analysis engine analyzes emotional expressions contained in the input text and outputs the user's emotional state.

[0440] Step 6:

[0441] The server generates a response in an appropriate tone based on the results of the generative AI model's analysis and the results of emotion analysis. During this process, it selects an appropriate tone (e.g., calm tone, encouraging tone) according to the intention and emotion, and outputs the response text data.

[0442] Step 7:

[0443] The server generates visual guide information based on the response text. Using the MapDisplay module, the server creates guide information for visually displaying the route from the user's current location to the destination, and sends the data to the terminal.

[0444] Step 8:

[0445] The device provides a response to the user based on the response text and visual guide information received from the server, specifically by playing back a voice response and displaying visual guide information on the smart glasses display.

[0446] Step 9:

[0447] The user inputs feedback on the provided response into the terminal, and this feedback data becomes new input.

[0448] Step 10:

[0449] The device sends user feedback data to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine, thereby continuously improving the accuracy and appropriateness of the system's responses.

[0450] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0451] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0452] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0453] [Second embodiment]

[0454] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0455] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0456] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0457] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0458] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0459] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0460] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0461] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0462] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0463] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0464] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0465] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0466] Below, we will describe a specific embodiment of a customer support system that utilizes generative AI.

[0467] First, the overall system configuration consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and is responsible for processing requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[0468] Basic operation of the system

[0469] 1. System startup and initialization

[0470] Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine, and performs the necessary settings and configuration.

[0471] 2. Accepting user input

[0472] Terminal: Displays an interface that allows speech recognition and text input. Users can enter operation instructions and questions by voice or text.

[0473] Device: If the input is voice, it is converted into text and the input data is sent to the server.

[0474] 3. Question Analysis and Answer Generation

[0475] Server: Analyzes the text data received from the device. The generative AI engine identifies the intent of the question and generates appropriate responses and operating instructions.

[0476] Server: Optionally, add multimedia material such as images or videos to the response.

[0477] 4. Providing a Response

[0478] Server: Sends the generated response and multimedia material to the terminal.

[0479] Terminal: Displays the received response to the user and plays the audio prompt, if available.

[0480] 5. Gather feedback and learn

[0481] User: Can provide feedback on the response.

[0482] Device: Sends user feedback to the server.

[0483] Server: Analyzes the received feedback and updates the generative AI engine's model, thereby improving the quality of future responses.

[0484] Specific examples

[0485] Scenario 1: Senior user learning how to use a smartphone camera

[0486] 1. User Input:

[0487] User: Uses the voice recognition feature on their smartphone to ask, "How do I use the camera?"

[0488] Terminal: Converts voice data into text and sends it to the server.

[0489] 2. Question analysis and answer generation:

[0490] Server: Analyzes the text "How do I use the camera?" and determines that it should explain how to use the camera app.

[0491] Server: Generates a detailed guide that explains how to start the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the camera app's interface.

[0492] 3. Providing a response:

[0493] Server: Sends the generated operation guide and multimedia materials to the terminal.

[0494] Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[0495] 4. Gathering Feedback:

[0496] Users: Rate their satisfaction with the guide and enter follow-up questions if needed.

[0497] Device: Sends user feedback to the server.

[0498] Server: Analyzes the feedback and updates the generative AI engine model.

[0499] In this way, the present invention provides a customer support system using generative AI, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality.

[0500] The processing flow will be explained below.

[0501] Step 1:

[0502] System startup and initialization

[0503] Terminal: Starts the system and loads libraries and required modules.

[0504] Terminal: Initialize the generation AI engine and perform the necessary settings.

[0505] Step 2:

[0506] Accepting user input

[0507] Terminal: displays the interface and allows the user to input by voice or text.

[0508] User: Asks a question or requests a command by voice or text.

[0509] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0510] Step 3:

[0511] Data reception and analysis

[0512] Server: Receives text data sent from the device.

[0513] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0514] Step 4:

[0515] Response Generation

[0516] Server: Generates appropriate responses and instructions based on the intent of the question. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app and take a photo.

[0517] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0518] Step 5:

[0519] Sending a Response

[0520] Server: Sends the generated response and multimedia material to the terminal.

[0521] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[0522] Step 6:

[0523] Collecting user feedback

[0524] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[0525] Terminal: Receives user feedback and sends it to the server.

[0526] Step 7:

[0527] Feedback analysis and model updating

[0528] Server: Analyzes the received feedback and updates the generative AI engine's model, making the next response more accurate and relevant.

[0529] This specific process flow allows the customer support system to efficiently provide instructions to customers, reduce the burden on store staff, and continuously improve the system based on feedback.

[0530] Example 1

[0531] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0532] Conventional customer support systems require a significant amount of human resources to provide operational instructions and respond to basic questions, making them particularly difficult to respond to during the initial learning phase when multiple questions frequently arise, or when product problems occur. Furthermore, conventional systems make it difficult to ensure the accuracy and consistency of responses, potentially leading to a decline in customer satisfaction. Furthermore, they are unable to effectively utilize feedback, which delays the continuous improvement of the system.

[0533] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0534] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for converting voice input from a user into text, means for analyzing the text and generating appropriate responses, means for providing the generated responses and multimedia materials to the user, and means for collecting feedback from users and updating the model of the generative AI engine. This makes it possible to quickly and accurately respond to a variety of questions from customers and continuously improve the system's response quality by utilizing the feedback.

[0535] "Generative AI" refers to a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[0536] "Generative AI engine" refers to the software components and hardware environment for operating and managing generative AI.

[0537] "Multimedia material" refers to data that contains multiple forms of digital content, such as images, video, and audio.

[0538] "Users" refers to people who use the system and provide instructions on how to use it or respond to questions.

[0539] "Feedback" refers to user responses and opinions and evaluations of system performance.

[0540] "Voice input" refers to the voice spoken by the user through a microphone.

[0541] "Text-to-text" refers to the process of converting voice input into written information.

[0542] "Analysis" refers to the process of structurally understanding and processing text data and feedback.

[0543] "Response" refers to information or instructions generated by a generative AI engine and provided to the user.

[0544] "Model updating" refers to the process of improving and adjusting the algorithms and parameters of a generative AI engine based on new data and feedback.

[0545] A specific embodiment of the present invention will be described below. A system for realizing the present invention comprises three elements: a server, a terminal, and a user.

[0546] The server runs a generative AI engine, analyzes user input data sent from the device, and generates a response. This generative AI engine can be based on commonly used artificial intelligence technologies (e.g., OpenAI's GPT-3). Furthermore, the server collects feedback from users and updates the generative AI engine's model to improve the quality of the response.

[0547] The terminal accepts user input data, communicates with the server, and ultimately presents a response to the user. Terminals are equipped with speech recognition and text input functions, and are devices such as smartphones and tablets. Google Cloud Speech-to-Text and other voice recognition APIs are used.

[0548] The user inputs questions or requests to the system through the terminal interface. The input method can be voice input or text input, depending on the user's preference. For example, the user may input "Please tell me how to use the camera" by voice.

[0549] The server converts the voice data received from the device into text using a speech recognition API, and then analyzes the text data using a generative AI engine. Generative AI such as GPT-3 generates an appropriate response based on the analysis results. Responses may include not only text information, but also multimedia materials such as images and videos to supplement the operating instructions. These multimedia materials are collected and managed on the server as needed.

[0550] The terminal receives the response and multimedia material sent from the server and displays them to the user. For example, it displays text on the smartphone screen and plays a video showing the operation procedure. This allows the user to intuitively and visually understand how to operate the device.

[0551] The user inputs feedback on the response and sends it from their device to the server. The server analyzes this feedback and updates the generative AI engine's model to improve the quality of the response from the next time onwards. In this way, the system has the ability to self-learn and continuously improve its performance.

[0552] Specific examples

[0553] Scenario 1: Senior user learning how to use a smartphone camera

[0554] 1. User input: The user speaks to their smartphone, asking, "How do I use the camera?"

[0555] 2. Server analysis: The voice data received from the device is converted into text, and the text "Please tell me how to use the camera" is analyzed using a generative AI engine (e.g., GPT-3).

[0556] 3. Response generation: Based on the analysis results, the server generates a text response that specifically explains how to use the camera app. For example, it generates a response such as "Open the camera app and press the center shutter button." It also includes an image showing the camera app's interface and a video explaining the operation procedure.

[0557] 4. Response provision: The response generated by the server and the multimedia material are sent to the terminal.

[0558] 5. User response confirmation: The device displays text, images, and videos to the user, and also plays audio instructions explaining how to operate the device.

[0559] This invention is implemented in the above-described manner, and a customer support system using generative AI can respond to user questions quickly and accurately, and utilize feedback to continuously improve the system's response quality.

[0560] Prompt Sentence Examples

[0561] "Please tell me how to use the camera. Specifically, please explain how to start the camera app, take a photo, and zoom in and out."

[0562] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0563] Step 1:

[0564] System startup and initialization

[0565] Device: When booting the system, the device first loads the necessary libraries and modules, specifically initializing the software components that run the speech recognition and generative AI engines (e.g., Python's tensorflow and numpy libraries).

[0566] Terminal: Next, the terminal establishes a network connection with the server and loads the configuration file (e.g., config.json), which completes the basic configuration of the system.

[0567] Input: The action the user takes to start the system.

[0568] Output: The required libraries and modules are loaded and a connection to the server is established.

[0569] Step 2:

[0570] Accepting user input

[0571] Terminal: A screen that allows voice recognition and text input is displayed through the user interface. The user inputs "How do I use the camera?" into the interface by voice or text.

[0572] Device: For voice input, use a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data.

[0573] Input: User voice or text input.

[0574] Output: User question data converted to text.

[0575] Step 3:

[0576] Question analysis and answer generation

[0577] Server: Receives text data received from the device and performs grammatical analysis of the text using a natural language processing library (e.g., spaCy). Then, a generative AI engine (e.g., OpenAI GPT-3) generates an appropriate response based on the text.

[0578] Server: In some cases, the generated response may include multimedia materials such as images or videos. For example, a screenshot of the camera app interface may be attached to the text response "Open the camera app and press the center shutter button."

[0579] Input: User question data converted to text.

[0580] Output: Appropriate response text and necessary multimedia materials.

[0581] Step 4:

[0582] Providing a response

[0583] Server: Sends the generated response and multimedia material to the terminal. The response is often packaged in JSON format or similar.

[0584] Terminal: The terminal displays the response received from the server to the user. Specifically, it displays text in the interface, plays images and videos, and outputs audio guidance, if available.

[0585] Input: The response data sent by the server.

[0586] Output: The response content (text, image, video) that is provided to the user.

[0587] Step 5:

[0588] Gathering feedback and learning

[0589] Users: Enter feedback on the response, adding comments such as "This response was helpful" or "Please explain in more detail."

[0590] Terminal: Sends the feedback data entered by the user to the server.

[0591] Server: Analyzes the received feedback and updates the generative AI engine model. Specifically, the feedback data is stored in a database and the model is trained based on that data to improve the quality of future responses.

[0592] Input: Feedback data from users.

[0593] Output: Feedback data parsed and stored, and an updated generative AI model.

[0594] (Application example 1)

[0595] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0596] Efficiently obtaining product information and operation instructions in physical stores often places a heavy burden on store staff. Providing fast and accurate support is difficult, especially during busy times. Furthermore, there is a lack of efficient customer support systems that utilize voice recognition and generative AI, raising concerns about a decline in customer satisfaction.

[0597] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0598] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from a user, means for displaying or playing responses to the user, means for converting voice input from the user into text data using a speech recognition function, means for using a generative AI model to generate optimal responses to user questions, and means for providing responses to the user in real time using a smartphone. This allows customers to quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[0599] "Generative AI" is an artificial intelligence technology that uses generative models to automatically generate content such as natural language text and images.

[0600] The "generative AI engine" is the core part of the system that runs the generative AI and generates responses to requests from users.

[0601] "Multimedia material" is digital content that includes multiple media formats such as images, video, and audio.

[0602] "Speech recognition" is a technology that analyzes voice input and converts it into text data.

[0603] A "smartphone" is a mobile device that has advanced computing power and Internet connection capabilities in addition to telephone functions.

[0604] "Self-learning" is a feature that allows a system to improve its own performance based on past data and feedback.

[0605] The "chat support function" is a function that responds to user questions in real time in a text-based chat format.

[0606] "Product placement information" is information relating to the location and placement of products within a store.

[0607] A specific embodiment of a system for realizing this application example will be described below.

[0608] Overall system configuration

[0609] This system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and processes requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[0610] Hardware and software used

[0611] Hardware: built-in microphone on smartphone, smartphone itself

[0612] software:

[0613] Speech recognition: speech_recognition

[0614] Generative AI model: HuggingFace Transformers library, specifically the bert-large-uncased-whole-word-masking-finetuned-squad model

[0615] Data processing and calculation

[0616] 1. Accepting voice input

[0617] The device uses a speech recognition module to collect voice input from the user and convert it into text data, allowing the user to voice questions or requests.

[0618] 2. Speech-to-text

[0619] The converted text data is sent to a generative AI model for analysis and response generation.

[0620] 3. Question Analysis and Response Generation Using Generative AI

[0621] The server analyzes the intent of the user's question based on the received text data and generates the optimal response. The generative AI model is responsible for this analysis and response generation.

[0622] 4. Providing a Response

[0623] The generated response is sent from the server to the terminal and displayed or played aloud to the user by the terminal, thereby providing the user with an immediate answer.

[0624] Examples of concrete examples and prompts

[0625] As a concrete example, consider the case where a user asks "Where is this item?" in a physical store. This question is handled as follows:

[0626] Examples:

[0627] The user uses their smartphone to voice-input "Where is this product?" The speech is converted into text, which is analyzed by the server's generative AI engine, which generates the optimal response based on product location information within the store. Finally, this response is presented to the user.

[0628] Example prompt sentence:

[0629] "Where is this item?"

[0630] "Food is on the first floor, and electrical appliances are on the second floor. Specifically, the TV is on the left at the back of the second floor."

[0631] In this way, customers can quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[0632] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0633] Step 1:

[0634] The device initializes the voice recognition module and waits for voice input. When the user speaks a question, the device collects the voice data through the microphone. The input is the voice data, and the output is the analysis of the voice data by the voice recognition module.

[0635] Step 2:

[0636] The collected voice data is converted into text data through the device's voice recognition module. In this conversion process, the voice recognition algorithm analyzes each phoneme from the user's voice and generates the corresponding text. The input is voice data, and the output is text data generated based on that voice.

[0637] Step 3:

[0638] The device sends the converted text data to the server, which then passes the received text data to the generative AI model to analyze the question and generate a response. The input is the text data sent from the device, and the output is the analysis result and the generated response.

[0639] Step 4:

[0640] The generative AI model analyzes the intent of the user's question based on the input text data and generates the optimal response. In this process, the generative AI model utilizes knowledge it has previously learned to analyze the text data. The input is text data, and the output is the generated response text.

[0641] Step 5:

[0642] The server sends the generated response text to the terminal, which receives the response text and prepares it to present to the user. The input is the response text sent from the server, and the output is data prepared by the terminal for display or audio playback.

[0643] Step 6:

[0644] The terminal displays the received response text to the user. Specifically, it displays the response as text on the smartphone screen or plays the response as audio using the audio playback function. The input is the response text, and the output is the display or audio data presented to the user.

[0645] Step 7:

[0646] The user can input feedback for the presented response through the terminal. The terminal sends this feedback as text data to the server. The input is the feedback (text data) from the user, and the output is the feedback data sent to the server.

[0647] Step 8:

[0648] The server analyzes the received feedback data and trains the generative AI model to self-train. In this process, the model's parameters are adjusted based on the feedback data to improve the accuracy of future responses. The input is the feedback data, and the output is an updated generative AI model.

[0649] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0650] This invention provides a customer support system that combines generative AI and an emotion engine. The system consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and emotion engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions and requests into the system through the terminal interface.

[0651] Basic operation of the system

[0652] 1. System startup and initialization

[0653] - Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine and emotion engine, and performs the necessary configuration.

[0654] 2. Accepting user input

[0655] - Terminal: displays the interface and allows the user to input by voice or text.

[0656] - User: Ask a question or request an action by voice or text.

[0657] - Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0658] 3. Data Reception and Analysis

[0659] - Server: Receives text data sent from the device.

[0660] - Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0661] - Server: Analyzes user emotions using an emotion engine and adjusts the tone and content of responses.

[0662] 4. Response Generation

[0663] - Server: Generates appropriate responses and instructions based on the intent of the question and the results of emotion analysis. For example, if the question is "Please tell me how to use the camera," the server generates instructions such as how to launch the camera app and how to take a photo, along with a tone that reflects the user's emotions.

[0664] - Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0665] 5. Providing a Response

[0666] - Server: Sends the generated response and multimedia material to the terminal.

[0667] - Terminal: Displays the received response to the user, and plays the audio prompt if available.

[0668] 6. Collecting User Feedback

[0669] - User: Enter feedback on the provided response (e.g. satisfaction, additional questions, etc.).

[0670] - Terminal: Receives user feedback and sends it to the server.

[0671] 7. Feedback analysis and model updating

[0672] - Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[0673] Specific examples

[0674] Scenario 1: Senior user learning how to use a smartphone camera

[0675] 1. User Input:

[0676] - User: Use your smartphone's voice recognition function to ask, "How do I use the camera?"

[0677] - Terminal: Converts voice data into text and sends it to the server.

[0678] 2. Data reception and analysis:

[0679] - Server: Analyzes the text "Please tell me how to use the camera" and determines that it should explain how to use the camera app.

[0680] - Server: Uses an emotion engine to recognize the user's current emotional state (e.g., anxiety, confusion, anticipation).

[0681] 3. Response generation:

[0682] - Server: Generates detailed instructions on how to launch the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the interface of the camera app.

[0683] - Server: Based on the analysis results of the emotion engine, generate a response in a tone that corresponds to the user's emotion (e.g., a calm tone, an encouraging tone).

[0684] 4. Providing a response:

[0685] - Server: Sends the generated operation guide and multimedia materials to the terminal.

[0686] - Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[0687] 5. Collecting User Feedback:

[0688] - User: Rate your satisfaction with the guide and enter follow-up questions if necessary.

[0689] - Device: Sends user feedback to the server.

[0690] - Server: Analyzes the feedback and updates the models of the generative AI engine and emotion engine.

[0691] In this way, the present invention provides a customer support system using generative AI and an emotion engine, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality. The emotion engine enables responses based on the user's emotions, further increasing customer satisfaction.

[0692] The processing flow will be explained below.

[0693] Step 1:

[0694] System startup and initialization

[0695] Terminal: Starts the system and loads libraries and required modules.

[0696] Terminal: Initialize the generative AI engine and emotion engine and perform the necessary settings.

[0697] Step 2:

[0698] Accepting user input

[0699] Terminal: Displays an interface that allows voice recognition and text input.

[0700] User: Asks a question or requests a command by voice or text.

[0701] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0702] Step 3:

[0703] Data reception and analysis

[0704] Server: Receives text data sent from the device.

[0705] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0706] Server: Analyzes the user's emotions using an emotion engine and identifies the user's emotional state (anxiety, confusion, expectation, etc.).

[0707] Step 4:

[0708] Response Generation

[0709] Server: Generates appropriate responses and instructions based on the intent of the question and the results of sentiment analysis. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app, how to take a photo, etc.

[0710] Server: Adjust the tone of your response depending on the user's emotional state. For example, if the user is feeling anxious, explain things in a calmer tone.

[0711] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0712] Step 5:

[0713] Sending a Response

[0714] Server: Sends the generated response and multimedia material to the terminal.

[0715] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[0716] Step 6:

[0717] Collecting user feedback

[0718] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[0719] Terminal: Receives user feedback and sends it to the server.

[0720] Step 7:

[0721] Feedback analysis and model updating

[0722] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[0723] Specific examples

[0724] Scenario 1: Senior user learning how to use a smartphone camera

[0725] Step 1:

[0726] System startup and initialization

[0727] Device: Start the smartphone system and initialize the generative AI engine and emotion engine.

[0728] Step 2:

[0729] Accepting user input

[0730] Terminal: Displays a voice recognition interface and an interface for the user to request instructions.

[0731] User: Ask aloud, "How do I use the camera?"

[0732] Terminal: Converts voice data into text and sends it to the server.

[0733] Step 3:

[0734] Data reception and analysis

[0735] Server: Receives text data saying "Please tell me how to use the camera."

[0736] Server: A generative AI engine analyzes the text data and identifies that the user is asking about how to use the camera app.

[0737] Server: The emotion engine identifies emotions such as anxiety and confusion from the user's voice.

[0738] Step 4:

[0739] Response Generation

[0740] Server: Generates a guide including how to launch the camera app, take a photo, and zoom in and out. For example, generate instructions such as "Open the camera app and press the center shutter button."

[0741] Server: Use a calming tone to explain things to the user to ease their concerns.

[0742] Server: Add images and videos showing the camera app interface.

[0743] Step 5:

[0744] Sending a Response

[0745] Server: Sends the generated explanation and multimedia materials to the terminal.

[0746] Device: Displays images and videos along with text instructions, and also provides audio instructions.

[0747] Step 6:

[0748] Collecting user feedback

[0749] Users: Leave feedback about how satisfied you are with the guide and any follow-up questions you may have.

[0750] Device: Sends user feedback to the server.

[0751] Step 7:

[0752] Feedback analysis and model updating

[0753] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, adjusting the next response to be more accurate and appropriate.

[0754] With this specific processing flow, a customer support system that utilizes generative AI and an emotion engine can efficiently provide customers with operating instructions, reduce the burden on store staff, and improve customer satisfaction by responding according to the user's emotions.

[0755] Example 2

[0756] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0757] Conventional customer support systems rely on human intervention to respond to customer questions and provide operational instructions, which tends to result in long response times and reduced customer satisfaction. Furthermore, the response content is fixed, making it difficult to respond appropriately to individual customers' emotions and situations. Furthermore, the system lacks a self-learning function and cannot utilize past feedback, making it difficult to improve system performance.

[0758] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0759] In this invention, the server includes a means for automating customer operation instructions and responses to basic questions using a generative AI, a means for initializing and managing the generative AI engine and the emotion analysis engine, a means for converting voice data to text and receiving and analyzing input data, a means for analyzing the user's emotional state and adjusting the response content, a means for generating a response based on the intent of the question and the results of the emotion analysis, a means for collecting and displaying multimedia materials, a means for displaying or playing the response to the user, and a means for receiving and analyzing user feedback and updating the generative AI engine model. This enables the server to provide prompt and appropriate responses to customer questions and respond according to the emotions and circumstances of each individual customer. Furthermore, the system's performance can be continuously improved based on past feedback.

[0760] "Generative AI" is a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[0761] A "generative AI engine" is software or a platform for implementing generative AI functions, specifically text generation and image generation using large-scale neural network models.

[0762] An "emotion analysis engine" is software or a system that analyzes user input data and identifies the user's emotional state, allowing the tone and content of responses to be tailored to the user's emotions.

[0763] "Multimedia materials" is a general term for digital content that includes various media formats such as images, videos, and audio. By using these, information provided to users can be enriched visually and aurally.

[0764] "Self-learning" is the ability of a system to improve its performance based on past data and feedback, allowing it to provide increasingly accurate and appropriate responses over time.

[0765] "Feedback" refers to input data such as user ratings of responses and follow-up questions, which are used to improve the system.

[0766] "Chat Support" refers to the overall interface and system for interacting with customers in real time, allowing for immediate answers to customer questions.

[0767] This invention relates to a customer support system that combines generative AI and a sentiment analysis engine. The system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and a sentiment analysis engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions or requests into the system through the terminal interface.

[0768] System configuration

[0769] 1. Hardware and Software Configuration

[0770] Server: The server is a powerful computer running the Linux operating system. A programming language such as Python is used to run the generative AI engine and sentiment analysis engine. Deep learning frameworks such as TensorFlow and PyTorch are used.

[0771] Device: A device is a device such as a smartphone, tablet, or PC. A web interface using HTML / CSS / JavaScript or an Android / iOS app runs on the device.

[0772] User: The user operates a terminal to input questions into the system and receive responses.

[0773] System Operation

[0774] 2. Data processing and calculation

[0775] Accepting user input:

[0776] Terminal: The terminal displays an HTML interface and allows users to enter questions by voice or text, converting the voice data to text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[0777] Data reception and analysis:

[0778] Server: Receives text data sent from the device, analyzes the text data using a generative AI engine (e.g., GPT-3), and identifies the intent of the question or request. Analyzes the user's emotions using a sentiment analysis engine and adjusts the response accordingly.

[0779] Response generation:

[0780] Server: Generates appropriate responses based on the intent of the question and the results of sentiment analysis. Selects and adds multimedia materials such as images and videos as needed.

[0781] Providing a response:

[0782] Server: Sends the generated response and multimedia material to the terminal, which displays the received response to the user and plays audio guidance, if any.

[0783] Collecting and analyzing user feedback:

[0784] Terminal: Receives user feedback and sends it to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine.

[0785] Specific examples

[0786] Prompt Sentence Examples

[0787] Prompt 1:

[0788] "How do I use the camera?" the user asks. This is the first time they've used a smartphone camera, and they seem nervous. Use a calming tone to provide step-by-step instructions on how to launch the camera app and take a photo.

[0789] The above is an embodiment of the present invention. This system enables quick and appropriate responses to customer questions, responding to individual customer emotions and situations. It also allows the system's performance to be continuously improved based on past feedback.

[0790] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0791] Step 1: System startup and initialization

[0792] Terminal: Starts the system and loads the necessary libraries and modules, such as Python, HTML, JavaScript, etc. During this process, the generative AI engine and sentiment analysis engine are also initialized, and API keys and configuration files are loaded.

[0793] Input: System startup command.

[0794] Output: Initialized generative AI engine and sentiment analysis engine.

[0795] Step 2: Accepting User Input

[0796] Terminal: Displays the user interface and allows the user to enter questions by voice or text, converting the voice data into text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[0797] User: Using the device interface, ask a question by voice or text, such as "How do I use the camera?"

[0798] Input: User's voice or text question.

[0799] Output: User question data converted to text.

[0800] Step 3: Data reception and analysis

[0801] Server: Receives text data sent from the device. Based on this data, it uses a generative AI engine (e.g., GPT-3) to analyze the intent of the question. It also uses an emotion analysis engine to analyze the user's emotions and adjust the response accordingly.

[0802] Input: Textualized user question data.

[0803] Output: Question intent and sentiment analysis results.

[0804] Step 4: Response Generation

[0805] Server: Generates an appropriate response based on the intent of the question and the results of sentiment analysis. For example, it generates specific instructions such as "Open the camera app and press the center shutter button." It also selects multimedia materials such as images and videos as needed.

[0806] Input: Question intent and sentiment analysis results.

[0807] Output: Generated response text and multimedia materials.

[0808] Step 5: Providing a response

[0809] Server: Sends the generated response and multimedia material to the terminal.

[0810] Terminal: Uses a speech synthesis engine to display the received response to the user and play audio prompts, if any.

[0811] Input: Generated response text and multimedia material.

[0812] Output: The response displayed to the user and the audio prompt played.

[0813] Step 6: Gather user feedback

[0814] User: Enter feedback on the response provided, for example, a satisfaction rating or a follow-up question.

[0815] Terminal: Receives user feedback and sends it to the server.

[0816] Input: User feedback or follow-up questions.

[0817] Output: Feedback data.

[0818] Step 7: Feedback analysis and model updating

[0819] Server: Analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine based on past questions and feedback.

[0820] Input: Feedback data.

[0821] Output: Updated generative AI engine and sentiment analysis engine.

[0822] The above is the specific processing flow of the system program and the specific operations performed at each step.

[0823] (Application example 2)

[0824] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0825] Conventional customer support systems have the problem that they respond mechanically to user questions and requests and are unable to respond in a way that reflects the user's emotions. Another problem is that a lack of visual guidance makes it difficult to provide efficient support when customers navigate in a store or search for products. This invention aims to provide a customer support system that can analyze a user's emotions and provide responses in real time with appropriate tone and content, as well as to improve the efficiency of in-store support by using visual guidance with smart glasses.

[0826] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automating operation instructions for customers and responses to basic questions using a generation AI, means for initializing and managing the generation AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from the user, means for displaying or playing responses to the user, means for analyzing the user's emotions and adjusting the tone and content of the responses, and means for providing visual guidance via the smart glasses. This allows the user to receive an appropriate response in a tone that corresponds to their emotions, while also allowing them to receive efficient support in the store through visual guidance.

[0827] "Generative AI" is an artificial intelligence technology that analyzes text data and generates appropriate answers and content.

[0828] A "customer" is a user of the system who inputs questions and requests.

[0829] "Instructions" refers to providing detailed instructions or guidance on the functionality and use of a system or device.

[0830] A "generative AI engine" is a software component that actually operates generative AI technology and performs analysis and response generation.

[0831] An "emotion analysis engine" is a software component that analyzes emotions from user input data and identifies their state.

[0832] "Tone" refers to the tone or expression of a response that corresponds to the user's emotion.

[0833] "Multimedia material" is data that includes information in a variety of formats, such as text, images, audio, and video.

[0834] "Input data" refers to the content of questions or requests entered by the user via voice or text.

[0835] "Smart glasses" are portable display devices capable of displaying visual information.

[0836] "Visual guide" refers to guide information that visually indicates product locations and directions within a store.

[0837] The "server" is a computer that runs the generative AI engine, emotion analysis engine, and related modules, and manages and operates the entire system.

[0838] "Real time" refers to immediate response or processing in response to user input.

[0839] A "store" refers to a physical location where users (customers) can actually visit to shop or obtain information.

[0840] "Response" means an answer or instruction provided by the system in response to a user's question or request.

[0841] "Self-learning" refers to the system's ability to automatically improve its performance based on past questions and feedback.

[0842] This invention provides a customer support system that combines generative AI and an emotion analysis engine. The system allows users to wear smart glasses and provides instant, appropriate responses and visual guidance when they ask questions or receive directions in a physical store.

[0843] System Configuration

[0844] The system consists of a server, a terminal, and a user. The server runs a generative AI engine and an emotion analysis engine, and the terminal (smart glasses) receives input from the user. The user inputs questions and requests into the system via the smart glasses.

[0845] Initialization and Management

[0846] The server initializes the generative AI engine and sentiment analysis engine and configures them appropriately. The server also updates the generative AI engine's model based on user feedback and past questions, enabling it to self-learn.

[0847] Voice Recognition

[0848] User input in the form of voice is collected as voice data by the device (smart glasses) and sent to the server, which then converts the voice data into text data using a speech recognition library. The speech_recognition library is used for this purpose.

[0849] Intention and emotion analysis

[0850] The server analyzes the received text data through a generative AI model (OpenAI's text-davinci-003) to identify the intent of the question, and uses a sentiment analysis engine to analyze the user's emotional state, which determines the tone of the response.

[0851] Response Generation

[0852] The generative AI engine generates appropriate responses based on the identified intent and the results of sentiment analysis. For example, in response to the question, "Where is the red sweater?", the generative AI engine generates a specific response such as, "The red sweater is in the women's clothing department on the second floor," and adjusts the tone accordingly.

[0853] Providing a visual guide

[0854] In addition to the generated textual response, the device provides visual guidance information, such as a store map and location guide. Visual guidance to assist with store navigation is provided using the MapDisplay module.

[0855] Specific examples

[0856] For example, a user puts on smart glasses and asks, "Where is the red sweater?" The voice data is converted to text through a speech recognition library and sent to a server. The server uses a generative AI model to analyze the intent and determine that directions to the red sweater should be provided. At the same time, it uses an emotion analysis engine to recognize the user's current emotional state and generate a response in a calm tone.

[0857] An example prompt for this would be:

[0858] Analyze the user's question to identify intent and sentiment and generate a response: Question: Where is the red sweater?

[0859] The generated response is accompanied by visual information, such as "The red sweater is in the women's clothing section on the second floor." As a visual guide, a route from the current location to the second floor is displayed on a map of the store. This series of processes allows the user to receive an appropriate response based on their emotions while also providing visual support for in-store navigation, resulting in comfortable and efficient customer support.

[0860] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0861] Step 1:

[0862] The terminal (smart glasses) collects questions entered by the user through voice, and this voice data becomes the input.

[0863] Step 2:

[0864] The device uses a speech recognition library (speech_recognition) to convert voice data into text data. During this conversion process, the device analyzes the voice waveform data and outputs the corresponding text.

[0865] Step 3:

[0866] The terminal sends the converted text data to the server, which receives the data and proceeds to the next analysis step.

[0867] Step 4:

[0868] The server uses a generative AI model (OpenAI's text-davinci-003) to analyze the text data and identify the intent of the user's question. It uses a prompt sentence for the analysis and outputs a model for generating an appropriate response from the user's question. An example of a prompt sentence in this case is, "Analyze the user's question to identify the intent and sentiment, and generate a response. Question: Where is the red sweater?"

[0869] Step 5:

[0870] The server uses an emotion analysis engine to analyze emotions from the user's text data, and the emotion analysis engine analyzes emotional expressions contained in the input text and outputs the user's emotional state.

[0871] Step 6:

[0872] The server generates a response in an appropriate tone based on the results of the generative AI model's analysis and the results of emotion analysis. During this process, it selects an appropriate tone (e.g., calm tone, encouraging tone) according to the intention and emotion, and outputs the response text data.

[0873] Step 7:

[0874] The server generates visual guide information based on the response text. Using the MapDisplay module, the server creates guide information for visually displaying the route from the user's current location to the destination, and sends the data to the terminal.

[0875] Step 8:

[0876] The device provides a response to the user based on the response text and visual guide information received from the server, specifically by playing back a voice response and displaying visual guide information on the smart glasses display.

[0877] Step 9:

[0878] The user inputs feedback on the provided response into the terminal, and this feedback data becomes new input.

[0879] Step 10:

[0880] The device sends user feedback data to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine, thereby continuously improving the accuracy and appropriateness of the system's responses.

[0881] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0882] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0883] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0884] [Third embodiment]

[0885] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0886] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0887] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0888] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0889] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0890] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0891] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0892] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0893] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0894] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0895] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0896] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0897] Below, we will describe a specific embodiment of a customer support system that utilizes generative AI.

[0898] First, the overall system configuration consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and is responsible for processing requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[0899] Basic operation of the system

[0900] 1. System startup and initialization

[0901] Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine, and performs the necessary settings and configuration.

[0902] 2. Accepting user input

[0903] Terminal: Displays an interface that allows speech recognition and text input. Users can enter operation instructions and questions by voice or text.

[0904] Device: If the input is voice, it is converted into text and the input data is sent to the server.

[0905] 3. Question Analysis and Answer Generation

[0906] Server: Analyzes the text data received from the device. The generative AI engine identifies the intent of the question and generates appropriate responses and operating instructions.

[0907] Server: Optionally, add multimedia material such as images or videos to the response.

[0908] 4. Providing a Response

[0909] Server: Sends the generated response and multimedia material to the terminal.

[0910] Terminal: Displays the received response to the user and plays the audio prompt, if available.

[0911] 5. Gather feedback and learn

[0912] User: Can provide feedback on the response.

[0913] Device: Sends user feedback to the server.

[0914] Server: Analyzes the received feedback and updates the generative AI engine's model, thereby improving the quality of future responses.

[0915] Specific examples

[0916] Scenario 1: Senior user learning how to use a smartphone camera

[0917] 1. User Input:

[0918] User: Uses the voice recognition feature on their smartphone to ask, "How do I use the camera?"

[0919] Terminal: Converts voice data into text and sends it to the server.

[0920] 2. Question analysis and answer generation:

[0921] Server: Analyzes the text "How do I use the camera?" and determines that it should explain how to use the camera app.

[0922] Server: Generates a detailed guide that explains how to start the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the camera app's interface.

[0923] 3. Providing a response:

[0924] Server: Sends the generated operation guide and multimedia materials to the terminal.

[0925] Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[0926] 4. Gathering Feedback:

[0927] Users: Rate their satisfaction with the guide and enter follow-up questions if needed.

[0928] Device: Sends user feedback to the server.

[0929] Server: Analyzes the feedback and updates the generative AI engine model.

[0930] In this way, the present invention provides a customer support system using generative AI, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality.

[0931] The processing flow will be explained below.

[0932] Step 1:

[0933] System startup and initialization

[0934] Terminal: Starts the system and loads libraries and required modules.

[0935] Terminal: Initialize the generation AI engine and perform the necessary settings.

[0936] Step 2:

[0937] Accepting user input

[0938] Terminal: displays the interface and allows the user to input by voice or text.

[0939] User: Asks a question or requests a command by voice or text.

[0940] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[0941] Step 3:

[0942] Data reception and analysis

[0943] Server: Receives text data sent from the device.

[0944] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[0945] Step 4:

[0946] Response Generation

[0947] Server: Generates appropriate responses and instructions based on the intent of the question. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app and take a photo.

[0948] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[0949] Step 5:

[0950] Sending a Response

[0951] Server: Sends the generated response and multimedia material to the terminal.

[0952] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[0953] Step 6:

[0954] Collecting user feedback

[0955] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[0956] Terminal: Receives user feedback and sends it to the server.

[0957] Step 7:

[0958] Feedback analysis and model updating

[0959] Server: Analyzes the received feedback and updates the generative AI engine's model, making the next response more accurate and relevant.

[0960] This specific process flow allows the customer support system to efficiently provide instructions to customers, reduce the burden on store staff, and continuously improve the system based on feedback.

[0961] Example 1

[0962] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0963] Conventional customer support systems require a significant amount of human resources to provide operational instructions and respond to basic questions, making them particularly difficult to respond to during the initial learning phase when multiple questions frequently arise, or when product problems occur. Furthermore, conventional systems make it difficult to ensure the accuracy and consistency of responses, potentially leading to a decline in customer satisfaction. Furthermore, they are unable to effectively utilize feedback, which delays the continuous improvement of the system.

[0964] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0965] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for converting voice input from a user into text, means for analyzing the text and generating appropriate responses, means for providing the generated responses and multimedia materials to the user, and means for collecting feedback from users and updating the model of the generative AI engine. This makes it possible to quickly and accurately respond to a variety of questions from customers and continuously improve the system's response quality by utilizing the feedback.

[0966] "Generative AI" refers to a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[0967] "Generative AI engine" refers to the software components and hardware environment for operating and managing generative AI.

[0968] "Multimedia material" refers to data that contains multiple forms of digital content, such as images, video, and audio.

[0969] "Users" refers to people who use the system and provide instructions on how to use it or respond to questions.

[0970] "Feedback" refers to user responses and opinions and evaluations of system performance.

[0971] "Voice input" refers to the voice spoken by the user through a microphone.

[0972] "Text-to-text" refers to the process of converting voice input into written information.

[0973] "Analysis" refers to the process of structurally understanding and processing text data and feedback.

[0974] "Response" refers to information or instructions generated by a generative AI engine and provided to the user.

[0975] "Model updating" refers to the process of improving and adjusting the algorithms and parameters of a generative AI engine based on new data and feedback.

[0976] A specific embodiment of the present invention will be described below. A system for realizing the present invention comprises three elements: a server, a terminal, and a user.

[0977] The server runs a generative AI engine, analyzes user input data sent from the device, and generates a response. This generative AI engine can be based on commonly used artificial intelligence technologies (e.g., OpenAI's GPT-3). Furthermore, the server collects feedback from users and updates the generative AI engine's model to improve the quality of the response.

[0978] The terminal accepts user input data, communicates with the server, and ultimately presents a response to the user. Terminals are equipped with speech recognition and text input functions, and are devices such as smartphones and tablets. Google Cloud Speech-to-Text and other voice recognition APIs are used.

[0979] The user inputs questions or requests to the system through the terminal interface. The input method can be voice input or text input, depending on the user's preference. For example, the user may input "Please tell me how to use the camera" by voice.

[0980] The server converts the voice data received from the device into text using a speech recognition API, and then analyzes the text data using a generative AI engine. Generative AI such as GPT-3 generates an appropriate response based on the analysis results. Responses may include not only text information, but also multimedia materials such as images and videos to supplement the operating instructions. These multimedia materials are collected and managed on the server as needed.

[0981] The terminal receives the response and multimedia material sent from the server and displays them to the user. For example, it displays text on the smartphone screen and plays a video showing the operation procedure. This allows the user to intuitively and visually understand how to operate the device.

[0982] The user inputs feedback on the response and sends it from their device to the server. The server analyzes this feedback and updates the generative AI engine's model to improve the quality of the response from the next time onwards. In this way, the system has the ability to self-learn and continuously improve its performance.

[0983] Specific examples

[0984] Scenario 1: Senior user learning how to use a smartphone camera

[0985] 1. User input: The user speaks to their smartphone, asking, "How do I use the camera?"

[0986] 2. Server analysis: The voice data received from the device is converted into text, and the text "Please tell me how to use the camera" is analyzed using a generative AI engine (e.g., GPT-3).

[0987] 3. Response generation: Based on the analysis results, the server generates a text response that specifically explains how to use the camera app. For example, it generates a response such as "Open the camera app and press the center shutter button." It also includes an image showing the camera app's interface and a video explaining the operation procedure.

[0988] 4. Response provision: The response generated by the server and the multimedia material are sent to the terminal.

[0989] 5. User response confirmation: The device displays text, images, and videos to the user, and also plays audio instructions explaining how to operate the device.

[0990] This invention is implemented in the above-described manner, and a customer support system using generative AI can respond to user questions quickly and accurately, and utilize feedback to continuously improve the system's response quality.

[0991] Prompt Sentence Examples

[0992] "Please tell me how to use the camera. Specifically, please explain how to start the camera app, take a photo, and zoom in and out."

[0993] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0994] Step 1:

[0995] System startup and initialization

[0996] Device: When booting the system, the device first loads the necessary libraries and modules, specifically initializing the software components that run the speech recognition and generative AI engines (e.g., Python's tensorflow and numpy libraries).

[0997] Terminal: Next, the terminal establishes a network connection with the server and loads the configuration file (e.g., config.json), which completes the basic configuration of the system.

[0998] Input: The action the user takes to start the system.

[0999] Output: The required libraries and modules are loaded and a connection to the server is established.

[1000] Step 2:

[1001] Accepting user input

[1002] Terminal: A screen that allows voice recognition and text input is displayed through the user interface. The user inputs "How do I use the camera?" into the interface by voice or text.

[1003] Device: For voice input, use a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data.

[1004] Input: User voice or text input.

[1005] Output: User question data converted to text.

[1006] Step 3:

[1007] Question analysis and answer generation

[1008] Server: Receives text data received from the device and performs grammatical analysis of the text using a natural language processing library (e.g., spaCy). Then, a generative AI engine (e.g., OpenAI GPT-3) generates an appropriate response based on the text.

[1009] Server: In some cases, the generated response may include multimedia materials such as images or videos. For example, a screenshot of the camera app interface may be attached to the text response "Open the camera app and press the center shutter button."

[1010] Input: User question data converted to text.

[1011] Output: Appropriate response text and necessary multimedia materials.

[1012] Step 4:

[1013] Providing a response

[1014] Server: Sends the generated response and multimedia material to the terminal. The response is often packaged in JSON format or similar.

[1015] Terminal: The terminal displays the response received from the server to the user. Specifically, it displays text in the interface, plays images and videos, and outputs audio guidance, if available.

[1016] Input: The response data sent by the server.

[1017] Output: The response content (text, image, video) that is provided to the user.

[1018] Step 5:

[1019] Gathering feedback and learning

[1020] Users: Enter feedback on the response, adding comments such as "This response was helpful" or "Please explain in more detail."

[1021] Terminal: Sends the feedback data entered by the user to the server.

[1022] Server: Analyzes the received feedback and updates the generative AI engine model. Specifically, the feedback data is stored in a database and the model is trained based on that data to improve the quality of future responses.

[1023] Input: Feedback data from users.

[1024] Output: Feedback data parsed and stored, and an updated generative AI model.

[1025] (Application example 1)

[1026] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1027] Efficiently obtaining product information and operation instructions in physical stores often places a heavy burden on store staff. Providing fast and accurate support is difficult, especially during busy times. Furthermore, there is a lack of efficient customer support systems that utilize voice recognition and generative AI, raising concerns about a decline in customer satisfaction.

[1028] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1029] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from a user, means for displaying or playing responses to the user, means for converting voice input from the user into text data using a speech recognition function, means for using a generative AI model to generate optimal responses to user questions, and means for providing responses to the user in real time using a smartphone. This allows customers to quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[1030] "Generative AI" is an artificial intelligence technology that uses generative models to automatically generate content such as natural language text and images.

[1031] The "generative AI engine" is the core part of the system that runs the generative AI and generates responses to requests from users.

[1032] "Multimedia material" is digital content that includes multiple media formats such as images, video, and audio.

[1033] "Speech recognition" is a technology that analyzes voice input and converts it into text data.

[1034] A "smartphone" is a mobile device that has advanced computing power and Internet connection capabilities in addition to telephone functions.

[1035] "Self-learning" is a feature that allows a system to improve its own performance based on past data and feedback.

[1036] The "chat support function" is a function that responds to user questions in real time in a text-based chat format.

[1037] "Product placement information" is information relating to the location and placement of products within a store.

[1038] A specific embodiment of a system for realizing this application example will be described below.

[1039] Overall system configuration

[1040] This system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and processes requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[1041] Hardware and software used

[1042] Hardware: built-in microphone on smartphone, smartphone itself

[1043] software:

[1044] Speech recognition: speech_recognition

[1045] Generative AI model: HuggingFace Transformers library, specifically the bert-large-uncased-whole-word-masking-finetuned-squad model

[1046] Data processing and calculation

[1047] 1. Accepting voice input

[1048] The device uses a speech recognition module to collect voice input from the user and convert it into text data, allowing the user to voice questions or requests.

[1049] 2. Speech-to-text

[1050] The converted text data is sent to a generative AI model for analysis and response generation.

[1051] 3. Question Analysis and Response Generation Using Generative AI

[1052] The server analyzes the intent of the user's question based on the received text data and generates the optimal response. The generative AI model is responsible for this analysis and response generation.

[1053] 4. Providing a Response

[1054] The generated response is sent from the server to the terminal and displayed or played aloud to the user by the terminal, thereby providing the user with an immediate answer.

[1055] Examples of concrete examples and prompts

[1056] As a concrete example, consider the case where a user asks "Where is this item?" in a physical store. This question is handled as follows:

[1057] Examples:

[1058] The user uses their smartphone to voice-input "Where is this product?" The speech is converted into text, which is analyzed by the server's generative AI engine, which generates the optimal response based on product location information within the store. Finally, this response is presented to the user.

[1059] Example prompt sentence:

[1060] "Where is this item?"

[1061] "Food is on the first floor, and electrical appliances are on the second floor. Specifically, the TV is on the left at the back of the second floor."

[1062] In this way, customers can quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[1063] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1064] Step 1:

[1065] The device initializes the voice recognition module and waits for voice input. When the user speaks a question, the device collects the voice data through the microphone. The input is the voice data, and the output is the analysis of the voice data by the voice recognition module.

[1066] Step 2:

[1067] The collected voice data is converted into text data through the device's voice recognition module. In this conversion process, the voice recognition algorithm analyzes each phoneme from the user's voice and generates the corresponding text. The input is voice data, and the output is text data generated based on that voice.

[1068] Step 3:

[1069] The device sends the converted text data to the server, which then passes the received text data to the generative AI model to analyze the question and generate a response. The input is the text data sent from the device, and the output is the analysis result and the generated response.

[1070] Step 4:

[1071] The generative AI model analyzes the intent of the user's question based on the input text data and generates the optimal response. In this process, the generative AI model utilizes knowledge it has previously learned to analyze the text data. The input is text data, and the output is the generated response text.

[1072] Step 5:

[1073] The server sends the generated response text to the terminal, which receives the response text and prepares it to present to the user. The input is the response text sent from the server, and the output is data prepared by the terminal for display or audio playback.

[1074] Step 6:

[1075] The terminal displays the received response text to the user. Specifically, it displays the response as text on the smartphone screen or plays the response as audio using the audio playback function. The input is the response text, and the output is the display or audio data presented to the user.

[1076] Step 7:

[1077] The user can input feedback for the presented response through the terminal. The terminal sends this feedback as text data to the server. The input is the feedback (text data) from the user, and the output is the feedback data sent to the server.

[1078] Step 8:

[1079] The server analyzes the received feedback data and trains the generative AI model to self-train. In this process, the model's parameters are adjusted based on the feedback data to improve the accuracy of future responses. The input is the feedback data, and the output is an updated generative AI model.

[1080] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1081] This invention provides a customer support system that combines generative AI and an emotion engine. The system consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and emotion engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions and requests into the system through the terminal interface.

[1082] Basic operation of the system

[1083] 1. System startup and initialization

[1084] - Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine and emotion engine, and performs the necessary configuration.

[1085] 2. Accepting user input

[1086] - Terminal: displays the interface and allows the user to input by voice or text.

[1087] - User: Ask a question or request an action by voice or text.

[1088] - Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[1089] 3. Data Reception and Analysis

[1090] - Server: Receives text data sent from the device.

[1091] - Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[1092] - Server: Analyzes user emotions using an emotion engine and adjusts the tone and content of responses.

[1093] 4. Response Generation

[1094] - Server: Generates appropriate responses and instructions based on the intent of the question and the results of emotion analysis. For example, if the question is "Please tell me how to use the camera," the server generates instructions such as how to launch the camera app and how to take a photo, along with a tone that reflects the user's emotions.

[1095] - Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[1096] 5. Providing a Response

[1097] - Server: Sends the generated response and multimedia material to the terminal.

[1098] - Terminal: Displays the received response to the user, and plays the audio prompt if available.

[1099] 6. Collecting User Feedback

[1100] - User: Enter feedback on the provided response (e.g. satisfaction, additional questions, etc.).

[1101] - Terminal: Receives user feedback and sends it to the server.

[1102] 7. Feedback analysis and model updating

[1103] - Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[1104] Specific examples

[1105] Scenario 1: Senior user learning how to use a smartphone camera

[1106] 1. User Input:

[1107] - User: Use your smartphone's voice recognition function to ask, "How do I use the camera?"

[1108] - Terminal: Converts voice data into text and sends it to the server.

[1109] 2. Data reception and analysis:

[1110] - Server: Analyzes the text "Please tell me how to use the camera" and determines that it should explain how to use the camera app.

[1111] - Server: Uses an emotion engine to recognize the user's current emotional state (e.g., anxiety, confusion, anticipation).

[1112] 3. Response generation:

[1113] - Server: Generates detailed instructions on how to launch the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the interface of the camera app.

[1114] - Server: Based on the analysis results of the emotion engine, generate a response in a tone that corresponds to the user's emotion (e.g., a calm tone, an encouraging tone).

[1115] 4. Providing a response:

[1116] - Server: Sends the generated operation guide and multimedia materials to the terminal.

[1117] - Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[1118] 5. Collecting User Feedback:

[1119] - User: Rate your satisfaction with the guide and enter follow-up questions if necessary.

[1120] - Device: Sends user feedback to the server.

[1121] - Server: Analyzes the feedback and updates the models of the generative AI engine and emotion engine.

[1122] In this way, the present invention provides a customer support system using generative AI and an emotion engine, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality. The emotion engine enables responses based on the user's emotions, further increasing customer satisfaction.

[1123] The processing flow will be explained below.

[1124] Step 1:

[1125] System startup and initialization

[1126] Terminal: Starts the system and loads libraries and required modules.

[1127] Terminal: Initialize the generative AI engine and emotion engine and perform the necessary settings.

[1128] Step 2:

[1129] Accepting user input

[1130] Terminal: Displays an interface that allows voice recognition and text input.

[1131] User: Asks a question or requests a command by voice or text.

[1132] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[1133] Step 3:

[1134] Data reception and analysis

[1135] Server: Receives text data sent from the device.

[1136] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[1137] Server: Analyzes the user's emotions using an emotion engine and identifies the user's emotional state (anxiety, confusion, expectation, etc.).

[1138] Step 4:

[1139] Response Generation

[1140] Server: Generates appropriate responses and instructions based on the intent of the question and the results of sentiment analysis. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app, how to take a photo, etc.

[1141] Server: Adjust the tone of your response depending on the user's emotional state. For example, if the user is feeling anxious, explain things in a calmer tone.

[1142] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[1143] Step 5:

[1144] Sending a Response

[1145] Server: Sends the generated response and multimedia material to the terminal.

[1146] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[1147] Step 6:

[1148] Collecting user feedback

[1149] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[1150] Terminal: Receives user feedback and sends it to the server.

[1151] Step 7:

[1152] Feedback analysis and model updating

[1153] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[1154] Specific examples

[1155] Scenario 1: Senior user learning how to use a smartphone camera

[1156] Step 1:

[1157] System startup and initialization

[1158] Device: Start the smartphone system and initialize the generative AI engine and emotion engine.

[1159] Step 2:

[1160] Accepting user input

[1161] Terminal: Displays a voice recognition interface and an interface for the user to request instructions.

[1162] User: Ask aloud, "How do I use the camera?"

[1163] Terminal: Converts voice data into text and sends it to the server.

[1164] Step 3:

[1165] Data reception and analysis

[1166] Server: Receives text data saying "Please tell me how to use the camera."

[1167] Server: A generative AI engine analyzes the text data and identifies that the user is asking about how to use the camera app.

[1168] Server: The emotion engine identifies emotions such as anxiety and confusion from the user's voice.

[1169] Step 4:

[1170] Response Generation

[1171] Server: Generates a guide including how to launch the camera app, take a photo, and zoom in and out. For example, generate instructions such as "Open the camera app and press the center shutter button."

[1172] Server: Use a calming tone to explain things to the user to ease their concerns.

[1173] Server: Add images and videos showing the camera app interface.

[1174] Step 5:

[1175] Sending a Response

[1176] Server: Sends the generated explanation and multimedia materials to the terminal.

[1177] Device: Displays images and videos along with text instructions, and also provides audio instructions.

[1178] Step 6:

[1179] Collecting user feedback

[1180] Users: Leave feedback about how satisfied you are with the guide and any follow-up questions you may have.

[1181] Device: Sends user feedback to the server.

[1182] Step 7:

[1183] Feedback analysis and model updating

[1184] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, adjusting the next response to be more accurate and appropriate.

[1185] With this specific processing flow, a customer support system that utilizes generative AI and an emotion engine can efficiently provide customers with operating instructions, reduce the burden on store staff, and improve customer satisfaction by responding according to the user's emotions.

[1186] Example 2

[1187] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1188] Conventional customer support systems rely on human intervention to respond to customer questions and provide operational instructions, which tends to result in long response times and reduced customer satisfaction. Furthermore, the response content is fixed, making it difficult to respond appropriately to individual customers' emotions and situations. Furthermore, the system lacks a self-learning function and cannot utilize past feedback, making it difficult to improve system performance.

[1189] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1190] In this invention, the server includes a means for automating customer operation instructions and responses to basic questions using a generative AI, a means for initializing and managing the generative AI engine and the emotion analysis engine, a means for converting voice data to text and receiving and analyzing input data, a means for analyzing the user's emotional state and adjusting the response content, a means for generating a response based on the intent of the question and the results of the emotion analysis, a means for collecting and displaying multimedia materials, a means for displaying or playing the response to the user, and a means for receiving and analyzing user feedback and updating the generative AI engine model. This enables the server to provide prompt and appropriate responses to customer questions and respond according to the emotions and circumstances of each individual customer. Furthermore, the system's performance can be continuously improved based on past feedback.

[1191] "Generative AI" is a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[1192] A "generative AI engine" is software or a platform for implementing generative AI functions, specifically text generation and image generation using large-scale neural network models.

[1193] An "emotion analysis engine" is software or a system that analyzes user input data and identifies the user's emotional state, allowing the tone and content of responses to be tailored to the user's emotions.

[1194] "Multimedia materials" is a general term for digital content that includes various media formats such as images, videos, and audio. By using these, information provided to users can be enriched visually and aurally.

[1195] "Self-learning" is the ability of a system to improve its performance based on past data and feedback, allowing it to provide increasingly accurate and appropriate responses over time.

[1196] "Feedback" refers to input data such as user ratings of responses and follow-up questions, which are used to improve the system.

[1197] "Chat Support" refers to the overall interface and system for interacting with customers in real time, allowing for immediate answers to customer questions.

[1198] This invention relates to a customer support system that combines generative AI and a sentiment analysis engine. The system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and a sentiment analysis engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions or requests into the system through the terminal interface.

[1199] System configuration

[1200] 1. Hardware and Software Configuration

[1201] Server: The server is a powerful computer running the Linux operating system. A programming language such as Python is used to run the generative AI engine and sentiment analysis engine. Deep learning frameworks such as TensorFlow and PyTorch are used.

[1202] Device: A device is a device such as a smartphone, tablet, or PC. A web interface using HTML / CSS / JavaScript or an Android / iOS app runs on the device.

[1203] User: The user operates a terminal to input questions into the system and receive responses.

[1204] System Operation

[1205] 2. Data processing and calculation

[1206] Accepting user input:

[1207] Terminal: The terminal displays an HTML interface and allows users to enter questions by voice or text, converting the voice data to text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[1208] Data reception and analysis:

[1209] Server: Receives text data sent from the device, analyzes the text data using a generative AI engine (e.g., GPT-3), and identifies the intent of the question or request. Analyzes the user's emotions using a sentiment analysis engine and adjusts the response accordingly.

[1210] Response generation:

[1211] Server: Generates appropriate responses based on the intent of the question and the results of sentiment analysis. Selects and adds multimedia materials such as images and videos as needed.

[1212] Providing a response:

[1213] Server: Sends the generated response and multimedia material to the terminal, which displays the received response to the user and plays audio guidance, if any.

[1214] Collecting and analyzing user feedback:

[1215] Terminal: Receives user feedback and sends it to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine.

[1216] Specific examples

[1217] Prompt Sentence Examples

[1218] Prompt 1:

[1219] "How do I use the camera?" the user asks. This is the first time they've used a smartphone camera, and they seem nervous. Use a calming tone to provide step-by-step instructions on how to launch the camera app and take a photo.

[1220] The above is an embodiment of the present invention. This system enables quick and appropriate responses to customer questions, responding to individual customer emotions and situations. It also allows the system's performance to be continuously improved based on past feedback.

[1221] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1222] Step 1: System startup and initialization

[1223] Terminal: Starts the system and loads the necessary libraries and modules, such as Python, HTML, JavaScript, etc. During this process, the generative AI engine and sentiment analysis engine are also initialized, and API keys and configuration files are loaded.

[1224] Input: System startup command.

[1225] Output: Initialized generative AI engine and sentiment analysis engine.

[1226] Step 2: Accepting User Input

[1227] Terminal: Displays the user interface and allows the user to enter questions by voice or text, converting the voice data into text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[1228] User: Using the device interface, ask a question by voice or text, such as "How do I use the camera?"

[1229] Input: User's voice or text question.

[1230] Output: User question data converted to text.

[1231] Step 3: Data reception and analysis

[1232] Server: Receives text data sent from the device. Based on this data, it uses a generative AI engine (e.g., GPT-3) to analyze the intent of the question. It also uses an emotion analysis engine to analyze the user's emotions and adjust the response accordingly.

[1233] Input: Textualized user question data.

[1234] Output: Question intent and sentiment analysis results.

[1235] Step 4: Response Generation

[1236] Server: Generates an appropriate response based on the intent of the question and the results of sentiment analysis. For example, it generates specific instructions such as "Open the camera app and press the center shutter button." It also selects multimedia materials such as images and videos as needed.

[1237] Input: Question intent and sentiment analysis results.

[1238] Output: Generated response text and multimedia materials.

[1239] Step 5: Providing a response

[1240] Server: Sends the generated response and multimedia material to the terminal.

[1241] Terminal: Uses a speech synthesis engine to display the received response to the user and play audio prompts, if any.

[1242] Input: Generated response text and multimedia material.

[1243] Output: The response displayed to the user and the audio prompt played.

[1244] Step 6: Gather user feedback

[1245] User: Enter feedback on the response provided, for example, a satisfaction rating or a follow-up question.

[1246] Terminal: Receives user feedback and sends it to the server.

[1247] Input: User feedback or follow-up questions.

[1248] Output: Feedback data.

[1249] Step 7: Feedback analysis and model updating

[1250] Server: Analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine based on past questions and feedback.

[1251] Input: Feedback data.

[1252] Output: Updated generative AI engine and sentiment analysis engine.

[1253] The above is the specific processing flow of the system program and the specific operations performed at each step.

[1254] (Application example 2)

[1255] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1256] Conventional customer support systems have the problem that they respond mechanically to user questions and requests and are unable to respond in a way that reflects the user's emotions. Another problem is that a lack of visual guidance makes it difficult to provide efficient support when customers navigate in a store or search for products. This invention aims to provide a customer support system that can analyze a user's emotions and provide responses in real time with appropriate tone and content, as well as to improve the efficiency of in-store support by using visual guidance with smart glasses.

[1257] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automating operation instructions for customers and responses to basic questions using a generation AI, means for initializing and managing the generation AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from the user, means for displaying or playing responses to the user, means for analyzing the user's emotions and adjusting the tone and content of the responses, and means for providing visual guidance via the smart glasses. This allows the user to receive an appropriate response in a tone that corresponds to their emotions, while also allowing them to receive efficient support in the store through visual guidance.

[1258] "Generative AI" is an artificial intelligence technology that analyzes text data and generates appropriate answers and content.

[1259] A "customer" is a user of the system who inputs questions and requests.

[1260] "Instructions" refers to providing detailed instructions or guidance on the functionality and use of a system or device.

[1261] A "generative AI engine" is a software component that actually operates generative AI technology and performs analysis and response generation.

[1262] An "emotion analysis engine" is a software component that analyzes emotions from user input data and identifies their state.

[1263] "Tone" refers to the tone or expression of a response that corresponds to the user's emotion.

[1264] "Multimedia material" is data that includes information in a variety of formats, such as text, images, audio, and video.

[1265] "Input data" refers to the content of questions or requests entered by the user via voice or text.

[1266] "Smart glasses" are portable display devices capable of displaying visual information.

[1267] "Visual guide" refers to guide information that visually indicates product locations and directions within a store.

[1268] The "server" is a computer that runs the generative AI engine, emotion analysis engine, and related modules, and manages and operates the entire system.

[1269] "Real time" refers to immediate response or processing in response to user input.

[1270] A "store" refers to a physical location where users (customers) can actually visit to shop or obtain information.

[1271] "Response" means an answer or instruction provided by the system in response to a user's question or request.

[1272] "Self-learning" refers to the system's ability to automatically improve its performance based on past questions and feedback.

[1273] This invention provides a customer support system that combines generative AI and an emotion analysis engine. The system allows users to wear smart glasses and provides instant, appropriate responses and visual guidance when they ask questions or receive directions in a physical store.

[1274] System Configuration

[1275] The system consists of a server, a terminal, and a user. The server runs a generative AI engine and an emotion analysis engine, and the terminal (smart glasses) receives input from the user. The user inputs questions and requests into the system via the smart glasses.

[1276] Initialization and Management

[1277] The server initializes the generative AI engine and sentiment analysis engine and configures them appropriately. The server also updates the generative AI engine's model based on user feedback and past questions, enabling it to self-learn.

[1278] Voice Recognition

[1279] User input in the form of voice is collected as voice data by the device (smart glasses) and sent to the server, which then converts the voice data into text data using a speech recognition library. The speech_recognition library is used for this purpose.

[1280] Intention and emotion analysis

[1281] The server analyzes the received text data through a generative AI model (OpenAI's text-davinci-003) to identify the intent of the question, and uses a sentiment analysis engine to analyze the user's emotional state, which determines the tone of the response.

[1282] Response Generation

[1283] The generative AI engine generates appropriate responses based on the identified intent and the results of sentiment analysis. For example, in response to the question, "Where is the red sweater?", the generative AI engine generates a specific response such as, "The red sweater is in the women's clothing department on the second floor," and adjusts the tone accordingly.

[1284] Providing a visual guide

[1285] In addition to the generated textual response, the device provides visual guidance information, such as a store map and location guide. Visual guidance to assist with store navigation is provided using the MapDisplay module.

[1286] Specific examples

[1287] For example, a user puts on smart glasses and asks, "Where is the red sweater?" The voice data is converted to text through a speech recognition library and sent to a server. The server uses a generative AI model to analyze the intent and determine that directions to the red sweater should be provided. At the same time, it uses an emotion analysis engine to recognize the user's current emotional state and generate a response in a calm tone.

[1288] An example prompt for this would be:

[1289] Analyze the user's question to identify intent and sentiment and generate a response: Question: Where is the red sweater?

[1290] The generated response is accompanied by visual information, such as "The red sweater is in the women's clothing section on the second floor." As a visual guide, a route from the current location to the second floor is displayed on a map of the store. This series of processes allows the user to receive an appropriate response based on their emotions while also providing visual support for in-store navigation, resulting in comfortable and efficient customer support.

[1291] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1292] Step 1:

[1293] The terminal (smart glasses) collects questions entered by the user through voice, and this voice data becomes the input.

[1294] Step 2:

[1295] The device uses a speech recognition library (speech_recognition) to convert voice data into text data. During this conversion process, the device analyzes the voice waveform data and outputs the corresponding text.

[1296] Step 3:

[1297] The terminal sends the converted text data to the server, which receives the data and proceeds to the next analysis step.

[1298] Step 4:

[1299] The server uses a generative AI model (OpenAI's text-davinci-003) to analyze the text data and identify the intent of the user's question. It uses a prompt sentence for the analysis and outputs a model for generating an appropriate response from the user's question. An example of a prompt sentence in this case is, "Analyze the user's question to identify the intent and sentiment, and generate a response. Question: Where is the red sweater?"

[1300] Step 5:

[1301] The server uses an emotion analysis engine to analyze emotions from the user's text data, and the emotion analysis engine analyzes emotional expressions contained in the input text and outputs the user's emotional state.

[1302] Step 6:

[1303] The server generates a response in an appropriate tone based on the results of the generative AI model's analysis and the results of emotion analysis. During this process, it selects an appropriate tone (e.g., calm tone, encouraging tone) according to the intention and emotion, and outputs the response text data.

[1304] Step 7:

[1305] The server generates visual guide information based on the response text. Using the MapDisplay module, the server creates guide information for visually displaying the route from the user's current location to the destination, and sends the data to the terminal.

[1306] Step 8:

[1307] The device provides a response to the user based on the response text and visual guide information received from the server, specifically by playing back a voice response and displaying visual guide information on the smart glasses display.

[1308] Step 9:

[1309] The user inputs feedback on the provided response into the terminal, and this feedback data becomes new input.

[1310] Step 10:

[1311] The device sends user feedback data to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine, thereby continuously improving the accuracy and appropriateness of the system's responses.

[1312] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1313] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1314] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1315] [Fourth embodiment]

[1316] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1317] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1318] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1319] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1320] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1321] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1322] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1323] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1324] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1325] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1326] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1327] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1328] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1329] Below, we will describe a specific embodiment of a customer support system that utilizes generative AI.

[1330] First, the overall system configuration consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and is responsible for processing requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[1331] Basic operation of the system

[1332] 1. System startup and initialization

[1333] Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine, and performs the necessary settings and configuration.

[1334] 2. Accepting user input

[1335] Terminal: Displays an interface that allows speech recognition and text input. Users can enter operation instructions and questions by voice or text.

[1336] Device: If the input is voice, it is converted into text and the input data is sent to the server.

[1337] 3. Question Analysis and Answer Generation

[1338] Server: Analyzes the text data received from the device. The generative AI engine identifies the intent of the question and generates appropriate responses and operating instructions.

[1339] Server: Optionally, add multimedia material such as images or videos to the response.

[1340] 4. Providing a Response

[1341] Server: Sends the generated response and multimedia material to the terminal.

[1342] Terminal: Displays the received response to the user and plays the audio prompt, if available.

[1343] 5. Gather feedback and learn

[1344] User: Can provide feedback on the response.

[1345] Device: Sends user feedback to the server.

[1346] Server: Analyzes the received feedback and updates the generative AI engine's model, thereby improving the quality of future responses.

[1347] Specific examples

[1348] Scenario 1: Senior user learning how to use a smartphone camera

[1349] 1. User Input:

[1350] User: Uses the voice recognition feature on their smartphone to ask, "How do I use the camera?"

[1351] Terminal: Converts voice data into text and sends it to the server.

[1352] 2. Question analysis and answer generation:

[1353] Server: Analyzes the text "How do I use the camera?" and determines that it should explain how to use the camera app.

[1354] Server: Generates a detailed guide that explains how to start the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the camera app's interface.

[1355] 3. Providing a response:

[1356] Server: Sends the generated operation guide and multimedia materials to the terminal.

[1357] Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[1358] 4. Gathering Feedback:

[1359] Users: Rate their satisfaction with the guide and enter follow-up questions if needed.

[1360] Device: Sends user feedback to the server.

[1361] Server: Analyzes the feedback and updates the generative AI engine model.

[1362] In this way, the present invention provides a customer support system using generative AI, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality.

[1363] The processing flow will be explained below.

[1364] Step 1:

[1365] System startup and initialization

[1366] Terminal: Starts the system and loads libraries and required modules.

[1367] Terminal: Initialize the generation AI engine and perform the necessary settings.

[1368] Step 2:

[1369] Accepting user input

[1370] Terminal: displays the interface and allows the user to input by voice or text.

[1371] User: Asks a question or requests a command by voice or text.

[1372] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[1373] Step 3:

[1374] Data reception and analysis

[1375] Server: Receives text data sent from the device.

[1376] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[1377] Step 4:

[1378] Response Generation

[1379] Server: Generates appropriate responses and instructions based on the intent of the question. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app and take a photo.

[1380] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[1381] Step 5:

[1382] Sending a Response

[1383] Server: Sends the generated response and multimedia material to the terminal.

[1384] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[1385] Step 6:

[1386] Collecting user feedback

[1387] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[1388] Terminal: Receives user feedback and sends it to the server.

[1389] Step 7:

[1390] Feedback analysis and model updating

[1391] Server: Analyzes the received feedback and updates the generative AI engine's model, making the next response more accurate and relevant.

[1392] This specific process flow allows the customer support system to efficiently provide instructions to customers, reduce the burden on store staff, and continuously improve the system based on feedback.

[1393] Example 1

[1394] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1395] Conventional customer support systems require a significant amount of human resources to provide operational instructions and respond to basic questions, making them particularly difficult to respond to during the initial learning phase when multiple questions frequently arise, or when product problems occur. Furthermore, conventional systems make it difficult to ensure the accuracy and consistency of responses, potentially leading to a decline in customer satisfaction. Furthermore, they are unable to effectively utilize feedback, which delays the continuous improvement of the system.

[1396] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1397] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for converting voice input from a user into text, means for analyzing the text and generating appropriate responses, means for providing the generated responses and multimedia materials to the user, and means for collecting feedback from users and updating the model of the generative AI engine. This makes it possible to quickly and accurately respond to a variety of questions from customers and continuously improve the system's response quality by utilizing the feedback.

[1398] "Generative AI" refers to a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[1399] "Generative AI engine" refers to the software components and hardware environment for operating and managing generative AI.

[1400] "Multimedia material" refers to data that contains multiple forms of digital content, such as images, video, and audio.

[1401] "Users" refers to people who use the system and provide instructions on how to use it or respond to questions.

[1402] "Feedback" refers to user responses and opinions and evaluations of system performance.

[1403] "Voice input" refers to the voice spoken by the user through a microphone.

[1404] "Text-to-text" refers to the process of converting voice input into written information.

[1405] "Analysis" refers to the process of structurally understanding and processing text data and feedback.

[1406] "Response" refers to information or instructions generated by a generative AI engine and provided to the user.

[1407] "Model updating" refers to the process of improving and adjusting the algorithms and parameters of a generative AI engine based on new data and feedback.

[1408] A specific embodiment of the present invention will be described below. A system for realizing the present invention comprises three elements: a server, a terminal, and a user.

[1409] The server runs a generative AI engine, analyzes user input data sent from the device, and generates a response. This generative AI engine can be based on commonly used artificial intelligence technologies (e.g., OpenAI's GPT-3). Furthermore, the server collects feedback from users and updates the generative AI engine's model to improve the quality of the response.

[1410] The terminal accepts user input data, communicates with the server, and ultimately presents a response to the user. Terminals are equipped with speech recognition and text input functions, and are devices such as smartphones and tablets. Google Cloud Speech-to-Text and other voice recognition APIs are used.

[1411] The user inputs questions or requests to the system through the terminal interface. The input method can be voice input or text input, depending on the user's preference. For example, the user may input "Please tell me how to use the camera" by voice.

[1412] The server converts the voice data received from the device into text using a speech recognition API, and then analyzes the text data using a generative AI engine. Generative AI such as GPT-3 generates an appropriate response based on the analysis results. Responses may include not only text information, but also multimedia materials such as images and videos to supplement the operating instructions. These multimedia materials are collected and managed on the server as needed.

[1413] The terminal receives the response and multimedia material sent from the server and displays them to the user. For example, it displays text on the smartphone screen and plays a video showing the operation procedure. This allows the user to intuitively and visually understand how to operate the device.

[1414] The user inputs feedback on the response and sends it from their device to the server. The server analyzes this feedback and updates the generative AI engine's model to improve the quality of the response from the next time onwards. In this way, the system has the ability to self-learn and continuously improve its performance.

[1415] Specific examples

[1416] Scenario 1: Senior user learning how to use a smartphone camera

[1417] 1. User input: The user speaks to their smartphone, asking, "How do I use the camera?"

[1418] 2. Server analysis: The voice data received from the device is converted into text, and the text "Please tell me how to use the camera" is analyzed using a generative AI engine (e.g., GPT-3).

[1419] 3. Response generation: Based on the analysis results, the server generates a text response that specifically explains how to use the camera app. For example, it generates a response such as "Open the camera app and press the center shutter button." It also includes an image showing the camera app's interface and a video explaining the operation procedure.

[1420] 4. Response provision: The response generated by the server and the multimedia material are sent to the terminal.

[1421] 5. User response confirmation: The device displays text, images, and videos to the user, and also plays audio instructions explaining how to operate the device.

[1422] This invention is implemented in the above-described manner, and a customer support system using generative AI can respond to user questions quickly and accurately, and utilize feedback to continuously improve the system's response quality.

[1423] Prompt Sentence Examples

[1424] "Please tell me how to use the camera. Specifically, please explain how to start the camera app, take a photo, and zoom in and out."

[1425] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1426] Step 1:

[1427] System startup and initialization

[1428] Device: When booting the system, the device first loads the necessary libraries and modules, specifically initializing the software components that run the speech recognition and generative AI engines (e.g., Python's tensorflow and numpy libraries).

[1429] Terminal: Next, the terminal establishes a network connection with the server and loads the configuration file (e.g., config.json), which completes the basic configuration of the system.

[1430] Input: The action the user takes to start the system.

[1431] Output: The required libraries and modules are loaded and a connection to the server is established.

[1432] Step 2:

[1433] Accepting user input

[1434] Terminal: A screen that allows voice recognition and text input is displayed through the user interface. The user inputs "How do I use the camera?" into the interface by voice or text.

[1435] Device: For voice input, use a speech recognition API (e.g., Google Cloud Speech-to-Text) to convert the voice data into text data.

[1436] Input: User voice or text input.

[1437] Output: User question data converted to text.

[1438] Step 3:

[1439] Question analysis and answer generation

[1440] Server: Receives text data received from the device and performs grammatical analysis of the text using a natural language processing library (e.g., spaCy). Then, a generative AI engine (e.g., OpenAI GPT-3) generates an appropriate response based on the text.

[1441] Server: In some cases, the generated response may include multimedia materials such as images or videos. For example, a screenshot of the camera app interface may be attached to the text response "Open the camera app and press the center shutter button."

[1442] Input: User question data converted to text.

[1443] Output: Appropriate response text and necessary multimedia materials.

[1444] Step 4:

[1445] Providing a response

[1446] Server: Sends the generated response and multimedia material to the terminal. The response is often packaged in JSON format or similar.

[1447] Terminal: The terminal displays the response received from the server to the user. Specifically, it displays text in the interface, plays images and videos, and outputs audio guidance, if available.

[1448] Input: The response data sent by the server.

[1449] Output: The response content (text, image, video) that is provided to the user.

[1450] Step 5:

[1451] Gathering feedback and learning

[1452] Users: Enter feedback on the response, adding comments such as "This response was helpful" or "Please explain in more detail."

[1453] Terminal: Sends the feedback data entered by the user to the server.

[1454] Server: Analyzes the received feedback and updates the generative AI engine model. Specifically, the feedback data is stored in a database and the model is trained based on that data to improve the quality of future responses.

[1455] Input: Feedback data from users.

[1456] Output: Feedback data parsed and stored, and an updated generative AI model.

[1457] (Application example 1)

[1458] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1459] Efficiently obtaining product information and operation instructions in physical stores often places a heavy burden on store staff. Providing fast and accurate support is difficult, especially during busy times. Furthermore, there is a lack of efficient customer support systems that utilize voice recognition and generative AI, raising concerns about a decline in customer satisfaction.

[1460] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1461] In this invention, the server includes means for automating operation instructions and responses to basic questions for customers using a generative AI, means for initializing and managing the generative AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from a user, means for displaying or playing responses to the user, means for converting voice input from the user into text data using a speech recognition function, means for using a generative AI model to generate optimal responses to user questions, and means for providing responses to the user in real time using a smartphone. This allows customers to quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[1462] "Generative AI" is an artificial intelligence technology that uses generative models to automatically generate content such as natural language text and images.

[1463] The "generative AI engine" is the core part of the system that runs the generative AI and generates responses to requests from users.

[1464] "Multimedia material" is digital content that includes multiple media formats such as images, video, and audio.

[1465] "Speech recognition" is a technology that analyzes voice input and converts it into text data.

[1466] A "smartphone" is a mobile device that has advanced computing power and Internet connection capabilities in addition to telephone functions.

[1467] "Self-learning" is a feature that allows a system to improve its own performance based on past data and feedback.

[1468] The "chat support function" is a function that responds to user questions in real time in a text-based chat format.

[1469] "Product placement information" is information relating to the location and placement of products within a store.

[1470] A specific embodiment of a system for realizing this application example will be described below.

[1471] Overall system configuration

[1472] This system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and processes requests from customers. The terminal accepts input from users, communicates with the server, and ultimately presents responses to the users. Users input questions and requests to the system through the terminal interface.

[1473] Hardware and software used

[1474] Hardware: built-in microphone on smartphone, smartphone itself

[1475] software:

[1476] Speech recognition: speech_recognition

[1477] Generative AI model: HuggingFace Transformers library, specifically the bert-large-uncased-whole-word-masking-finetuned-squad model

[1478] Data processing and calculation

[1479] 1. Accepting voice input

[1480] The device uses a speech recognition module to collect voice input from the user and convert it into text data, allowing the user to voice questions or requests.

[1481] 2. Speech-to-text

[1482] The converted text data is sent to a generative AI model for analysis and response generation.

[1483] 3. Question Analysis and Response Generation Using Generative AI

[1484] The server analyzes the intent of the user's question based on the received text data and generates the optimal response. The generative AI model is responsible for this analysis and response generation.

[1485] 4. Providing a Response

[1486] The generated response is sent from the server to the terminal and displayed or played aloud to the user by the terminal, thereby providing the user with an immediate answer.

[1487] Examples of concrete examples and prompts

[1488] As a concrete example, consider the case where a user asks "Where is this item?" in a physical store. This question is handled as follows:

[1489] Examples:

[1490] The user uses their smartphone to voice-input "Where is this product?" The speech is converted into text, which is analyzed by the server's generative AI engine, which generates the optimal response based on product location information within the store. Finally, this response is presented to the user.

[1491] Example prompt sentence:

[1492] "Where is this item?"

[1493] "Food is on the first floor, and electrical appliances are on the second floor. Specifically, the TV is on the left at the back of the second floor."

[1494] In this way, customers can quickly and accurately obtain the information they need in the store, reducing the burden on store staff and improving customer satisfaction.

[1495] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1496] Step 1:

[1497] The device initializes the voice recognition module and waits for voice input. When the user speaks a question, the device collects the voice data through the microphone. The input is the voice data, and the output is the analysis of the voice data by the voice recognition module.

[1498] Step 2:

[1499] The collected voice data is converted into text data through the device's voice recognition module. In this conversion process, the voice recognition algorithm analyzes each phoneme from the user's voice and generates the corresponding text. The input is voice data, and the output is text data generated based on that voice.

[1500] Step 3:

[1501] The device sends the converted text data to the server, which then passes the received text data to the generative AI model to analyze the question and generate a response. The input is the text data sent from the device, and the output is the analysis result and the generated response.

[1502] Step 4:

[1503] The generative AI model analyzes the intent of the user's question based on the input text data and generates the optimal response. In this process, the generative AI model utilizes knowledge it has previously learned to analyze the text data. The input is text data, and the output is the generated response text.

[1504] Step 5:

[1505] The server sends the generated response text to the terminal, which receives the response text and prepares it to present to the user. The input is the response text sent from the server, and the output is data prepared by the terminal for display or audio playback.

[1506] Step 6:

[1507] The terminal displays the received response text to the user. Specifically, it displays the response as text on the smartphone screen or plays the response as audio using the audio playback function. The input is the response text, and the output is the display or audio data presented to the user.

[1508] Step 7:

[1509] The user can input feedback for the presented response through the terminal. The terminal sends this feedback as text data to the server. The input is the feedback (text data) from the user, and the output is the feedback data sent to the server.

[1510] Step 8:

[1511] The server analyzes the received feedback data and trains the generative AI model to self-train. In this process, the model's parameters are adjusted based on the feedback data to improve the accuracy of future responses. The input is the feedback data, and the output is an updated generative AI model.

[1512] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1513] This invention provides a customer support system that combines generative AI and an emotion engine. The system consists of three components: a server, a terminal, and a user. The server runs the generative AI engine and emotion engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions and requests into the system through the terminal interface.

[1514] Basic operation of the system

[1515] 1. System startup and initialization

[1516] - Terminal: Starts the system, loads the necessary libraries and modules, initializes the generative AI engine and emotion engine, and performs the necessary configuration.

[1517] 2. Accepting user input

[1518] - Terminal: displays the interface and allows the user to input by voice or text.

[1519] - User: Ask a question or request an action by voice or text.

[1520] - Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[1521] 3. Data Reception and Analysis

[1522] - Server: Receives text data sent from the device.

[1523] - Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[1524] - Server: Analyzes user emotions using an emotion engine and adjusts the tone and content of responses.

[1525] 4. Response Generation

[1526] - Server: Generates appropriate responses and instructions based on the intent of the question and the results of emotion analysis. For example, if the question is "Please tell me how to use the camera," the server generates instructions such as how to launch the camera app and how to take a photo, along with a tone that reflects the user's emotions.

[1527] - Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[1528] 5. Providing a Response

[1529] - Server: Sends the generated response and multimedia material to the terminal.

[1530] - Terminal: Displays the received response to the user, and plays the audio prompt if available.

[1531] 6. Collecting User Feedback

[1532] - User: Enter feedback on the provided response (e.g. satisfaction, additional questions, etc.).

[1533] - Terminal: Receives user feedback and sends it to the server.

[1534] 7. Feedback analysis and model updating

[1535] - Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[1536] Specific examples

[1537] Scenario 1: Senior user learning how to use a smartphone camera

[1538] 1. User Input:

[1539] - User: Use your smartphone's voice recognition function to ask, "How do I use the camera?"

[1540] - Terminal: Converts voice data into text and sends it to the server.

[1541] 2. Data reception and analysis:

[1542] - Server: Analyzes the text "Please tell me how to use the camera" and determines that it should explain how to use the camera app.

[1543] - Server: Uses an emotion engine to recognize the user's current emotional state (e.g., anxiety, confusion, anticipation).

[1544] 3. Response generation:

[1545] - Server: Generates detailed instructions on how to launch the camera, take a photo, zoom in and out, etc. Specifically, it generates text such as "Open the camera app and press the center shutter button." It also adds images and videos showing the interface of the camera app.

[1546] - Server: Based on the analysis results of the emotion engine, generate a response in a tone that corresponds to the user's emotion (e.g., a calm tone, an encouraging tone).

[1547] 4. Providing a response:

[1548] - Server: Sends the generated operation guide and multimedia materials to the terminal.

[1549] - Device: In addition to text instructions, images and videos are displayed and audio instructions are provided.

[1550] 5. Collecting User Feedback:

[1551] - User: Rate your satisfaction with the guide and enter follow-up questions if necessary.

[1552] - Device: Sends user feedback to the server.

[1553] - Server: Analyzes the feedback and updates the models of the generative AI engine and emotion engine.

[1554] In this way, the present invention provides a customer support system using generative AI and an emotion engine, which efficiently provides operational instructions to customers and reduces the burden on store staff. The system also has a self-learning function, allowing it to continuously improve its response quality. The emotion engine enables responses based on the user's emotions, further increasing customer satisfaction.

[1555] The processing flow will be explained below.

[1556] Step 1:

[1557] System startup and initialization

[1558] Terminal: Starts the system and loads libraries and required modules.

[1559] Terminal: Initialize the generative AI engine and emotion engine and perform the necessary settings.

[1560] Step 2:

[1561] Accepting user input

[1562] Terminal: Displays an interface that allows voice recognition and text input.

[1563] User: Asks a question or requests a command by voice or text.

[1564] Terminal: In the case of voice input, the voice data is converted into text and the input data is sent to the server.

[1565] Step 3:

[1566] Data reception and analysis

[1567] Server: Receives text data sent from the device.

[1568] Server: Analyzes text data using a generative AI engine to identify the intent of questions and requests.

[1569] Server: Analyzes the user's emotions using an emotion engine and identifies the user's emotional state (anxiety, confusion, expectation, etc.).

[1570] Step 4:

[1571] Response Generation

[1572] Server: Generates appropriate responses and instructions based on the intent of the question and the results of sentiment analysis. For example, if the question is "Please tell me how to use the camera," it generates instructions on how to launch the camera app, how to take a photo, etc.

[1573] Server: Adjust the tone of your response depending on the user's emotional state. For example, if the user is feeling anxious, explain things in a calmer tone.

[1574] Server: Selects and adds multimedia materials such as images and videos to the response, if necessary.

[1575] Step 5:

[1576] Sending a Response

[1577] Server: Sends the generated response and multimedia material to the terminal.

[1578] Terminal: Displays the received response to the user, and plays the audio prompt if one is available.

[1579] Step 6:

[1580] Collecting user feedback

[1581] User: Enter feedback on the response provided (e.g. satisfaction, additional questions, etc.).

[1582] Terminal: Receives user feedback and sends it to the server.

[1583] Step 7:

[1584] Feedback analysis and model updating

[1585] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, making the next response more accurate and relevant.

[1586] Specific examples

[1587] Scenario 1: Senior user learning how to use a smartphone camera

[1588] Step 1:

[1589] System startup and initialization

[1590] Device: Start the smartphone system and initialize the generative AI engine and emotion engine.

[1591] Step 2:

[1592] Accepting user input

[1593] Terminal: Displays a voice recognition interface and an interface for the user to request instructions.

[1594] User: Ask aloud, "How do I use the camera?"

[1595] Terminal: Converts voice data into text and sends it to the server.

[1596] Step 3:

[1597] Data reception and analysis

[1598] Server: Receives text data saying "Please tell me how to use the camera."

[1599] Server: A generative AI engine analyzes the text data and identifies that the user is asking about how to use the camera app.

[1600] Server: The emotion engine identifies emotions such as anxiety and confusion from the user's voice.

[1601] Step 4:

[1602] Response Generation

[1603] Server: Generates a guide including how to launch the camera app, take a photo, and zoom in and out. For example, generate instructions such as "Open the camera app and press the center shutter button."

[1604] Server: Use a calming tone to explain things to the user to ease their concerns.

[1605] Server: Add images and videos showing the camera app interface.

[1606] Step 5:

[1607] Sending a Response

[1608] Server: Sends the generated explanation and multimedia materials to the terminal.

[1609] Device: Displays images and videos along with text instructions, and also provides audio instructions.

[1610] Step 6:

[1611] Collecting user feedback

[1612] Users: Leave feedback about how satisfied you are with the guide and any follow-up questions you may have.

[1613] Device: Sends user feedback to the server.

[1614] Step 7:

[1615] Feedback analysis and model updating

[1616] Server: Analyzes the received feedback and updates the models of the generative AI engine and emotion engine, adjusting the next response to be more accurate and appropriate.

[1617] With this specific processing flow, a customer support system that utilizes generative AI and an emotion engine can efficiently provide customers with operating instructions, reduce the burden on store staff, and improve customer satisfaction by responding according to the user's emotions.

[1618] Example 2

[1619] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1620] Conventional customer support systems rely on human intervention to respond to customer questions and provide operational instructions, which tends to result in long response times and reduced customer satisfaction. Furthermore, the response content is fixed, making it difficult to respond appropriately to individual customers' emotions and situations. Furthermore, the system lacks a self-learning function and cannot utilize past feedback, making it difficult to improve system performance.

[1621] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1622] In this invention, the server includes a means for automating customer operation instructions and responses to basic questions using a generative AI, a means for initializing and managing the generative AI engine and the emotion analysis engine, a means for converting voice data to text and receiving and analyzing input data, a means for analyzing the user's emotional state and adjusting the response content, a means for generating a response based on the intent of the question and the results of the emotion analysis, a means for collecting and displaying multimedia materials, a means for displaying or playing the response to the user, and a means for receiving and analyzing user feedback and updating the generative AI engine model. This enables the server to provide prompt and appropriate responses to customer questions and respond according to the emotions and circumstances of each individual customer. Furthermore, the system's performance can be continuously improved based on past feedback.

[1623] "Generative AI" is a system that uses artificial intelligence technology to automatically generate content such as text, images, and audio.

[1624] A "generative AI engine" is software or a platform for implementing generative AI functions, specifically text generation and image generation using large-scale neural network models.

[1625] An "emotion analysis engine" is software or a system that analyzes user input data and identifies the user's emotional state, allowing the tone and content of responses to be tailored to the user's emotions.

[1626] "Multimedia materials" is a general term for digital content that includes various media formats such as images, videos, and audio. By using these, information provided to users can be enriched visually and aurally.

[1627] "Self-learning" is the ability of a system to improve its performance based on past data and feedback, allowing it to provide increasingly accurate and appropriate responses over time.

[1628] "Feedback" refers to input data such as user ratings of responses and follow-up questions, which are used to improve the system.

[1629] "Chat Support" refers to the overall interface and system for interacting with customers in real time, allowing for immediate answers to customer questions.

[1630] This invention relates to a customer support system that combines generative AI and a sentiment analysis engine. The system consists of three components: a server, a terminal, and a user. The server runs a generative AI engine and a sentiment analysis engine and is responsible for processing requests and feedback from customers. The terminal receives input from the user, communicates with the server, and finally presents a response to the user. The user inputs questions or requests into the system through the terminal interface.

[1631] System configuration

[1632] 1. Hardware and Software Configuration

[1633] Server: The server is a powerful computer running the Linux operating system. A programming language such as Python is used to run the generative AI engine and sentiment analysis engine. Deep learning frameworks such as TensorFlow and PyTorch are used.

[1634] Device: A device is a device such as a smartphone, tablet, or PC. A web interface using HTML / CSS / JavaScript or an Android / iOS app runs on the device.

[1635] User: The user operates a terminal to input questions into the system and receive responses.

[1636] System Operation

[1637] 2. Data processing and calculation

[1638] Accepting user input:

[1639] Terminal: The terminal displays an HTML interface and allows users to enter questions by voice or text, converting the voice data to text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[1640] Data reception and analysis:

[1641] Server: Receives text data sent from the device, analyzes the text data using a generative AI engine (e.g., GPT-3), and identifies the intent of the question or request. Analyzes the user's emotions using a sentiment analysis engine and adjusts the response accordingly.

[1642] Response generation:

[1643] Server: Generates appropriate responses based on the intent of the question and the results of sentiment analysis. Selects and adds multimedia materials such as images and videos as needed.

[1644] Providing a response:

[1645] Server: Sends the generated response and multimedia material to the terminal, which displays the received response to the user and plays audio guidance, if any.

[1646] Collecting and analyzing user feedback:

[1647] Terminal: Receives user feedback and sends it to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine.

[1648] Specific examples

[1649] Prompt Sentence Examples

[1650] Prompt 1:

[1651] "How do I use the camera?" the user asks. This is the first time they've used a smartphone camera, and they seem nervous. Use a calming tone to provide step-by-step instructions on how to launch the camera app and take a photo.

[1652] The above is an embodiment of the present invention. This system enables quick and appropriate responses to customer questions, responding to individual customer emotions and situations. It also allows the system's performance to be continuously improved based on past feedback.

[1653] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1654] Step 1: System startup and initialization

[1655] Terminal: Starts the system and loads the necessary libraries and modules, such as Python, HTML, JavaScript, etc. During this process, the generative AI engine and sentiment analysis engine are also initialized, and API keys and configuration files are loaded.

[1656] Input: System startup command.

[1657] Output: Initialized generative AI engine and sentiment analysis engine.

[1658] Step 2: Accepting User Input

[1659] Terminal: Displays the user interface and allows the user to enter questions by voice or text, converting the voice data into text using the Google Speech-to-Text API or Microsoft Azure Recognition Services.

[1660] User: Using the device interface, ask a question by voice or text, such as "How do I use the camera?"

[1661] Input: User's voice or text question.

[1662] Output: User question data converted to text.

[1663] Step 3: Data reception and analysis

[1664] Server: Receives text data sent from the device. Based on this data, it uses a generative AI engine (e.g., GPT-3) to analyze the intent of the question. It also uses an emotion analysis engine to analyze the user's emotions and adjust the response accordingly.

[1665] Input: Textualized user question data.

[1666] Output: Question intent and sentiment analysis results.

[1667] Step 4: Response Generation

[1668] Server: Generates an appropriate response based on the intent of the question and the results of sentiment analysis. For example, it generates specific instructions such as "Open the camera app and press the center shutter button." It also selects multimedia materials such as images and videos as needed.

[1669] Input: Question intent and sentiment analysis results.

[1670] Output: Generated response text and multimedia materials.

[1671] Step 5: Providing a response

[1672] Server: Sends the generated response and multimedia material to the terminal.

[1673] Terminal: Uses a speech synthesis engine to display the received response to the user and play audio prompts, if any.

[1674] Input: Generated response text and multimedia material.

[1675] Output: The response displayed to the user and the audio prompt played.

[1676] Step 6: Gather user feedback

[1677] User: Enter feedback on the response provided, for example, a satisfaction rating or a follow-up question.

[1678] Terminal: Receives user feedback and sends it to the server.

[1679] Input: User feedback or follow-up questions.

[1680] Output: Feedback data.

[1681] Step 7: Feedback analysis and model updating

[1682] Server: Analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine based on past questions and feedback.

[1683] Input: Feedback data.

[1684] Output: Updated generative AI engine and sentiment analysis engine.

[1685] The above is the specific processing flow of the system program and the specific operations performed at each step.

[1686] (Application example 2)

[1687] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1688] Conventional customer support systems have the problem that they respond mechanically to user questions and requests and are unable to respond in a way that reflects the user's emotions. Another problem is that a lack of visual guidance makes it difficult to provide efficient support when customers navigate in a store or search for products. This invention aims to provide a customer support system that can analyze a user's emotions and provide responses in real time with appropriate tone and content, as well as to improve the efficiency of in-store support by using visual guidance with smart glasses.

[1689] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for automating operation instructions for customers and responses to basic questions using a generation AI, means for initializing and managing the generation AI engine, means for collecting and displaying multimedia materials, means for receiving and analyzing input data from the user, means for displaying or playing responses to the user, means for analyzing the user's emotions and adjusting the tone and content of the responses, and means for providing visual guidance via the smart glasses. This allows the user to receive an appropriate response in a tone that corresponds to their emotions, while also allowing them to receive efficient support in the store through visual guidance.

[1690] "Generative AI" is an artificial intelligence technology that analyzes text data and generates appropriate answers and content.

[1691] A "customer" is a user of the system who inputs questions and requests.

[1692] "Instructions" refers to providing detailed instructions or guidance on the functionality and use of a system or device.

[1693] A "generative AI engine" is a software component that actually operates generative AI technology and performs analysis and response generation.

[1694] An "emotion analysis engine" is a software component that analyzes emotions from user input data and identifies their state.

[1695] "Tone" refers to the tone or expression of a response that corresponds to the user's emotion.

[1696] "Multimedia material" is data that includes information in a variety of formats, such as text, images, audio, and video.

[1697] "Input data" refers to the content of questions or requests entered by the user via voice or text.

[1698] "Smart glasses" are portable display devices capable of displaying visual information.

[1699] "Visual guide" refers to guide information that visually indicates product locations and directions within a store.

[1700] The "server" is a computer that runs the generative AI engine, emotion analysis engine, and related modules, and manages and operates the entire system.

[1701] "Real time" refers to immediate response or processing in response to user input.

[1702] A "store" refers to a physical location where users (customers) can actually visit to shop or obtain information.

[1703] "Response" means an answer or instruction provided by the system in response to a user's question or request.

[1704] "Self-learning" refers to the system's ability to automatically improve its performance based on past questions and feedback.

[1705] This invention provides a customer support system that combines generative AI and an emotion analysis engine. The system allows users to wear smart glasses and provides instant, appropriate responses and visual guidance when they ask questions or receive directions in a physical store.

[1706] System Configuration

[1707] The system consists of a server, a terminal, and a user. The server runs a generative AI engine and an emotion analysis engine, and the terminal (smart glasses) receives input from the user. The user inputs questions and requests into the system via the smart glasses.

[1708] Initialization and Management

[1709] The server initializes the generative AI engine and sentiment analysis engine and configures them appropriately. The server also updates the generative AI engine's model based on user feedback and past questions, enabling it to self-learn.

[1710] Voice Recognition

[1711] User input in the form of voice is collected as voice data by the device (smart glasses) and sent to the server, which then converts the voice data into text data using a speech recognition library. The speech_recognition library is used for this purpose.

[1712] Intention and emotion analysis

[1713] The server analyzes the received text data through a generative AI model (OpenAI's text-davinci-003) to identify the intent of the question, and uses a sentiment analysis engine to analyze the user's emotional state, which determines the tone of the response.

[1714] Response Generation

[1715] The generative AI engine generates appropriate responses based on the identified intent and the results of sentiment analysis. For example, in response to the question, "Where is the red sweater?", the generative AI engine generates a specific response such as, "The red sweater is in the women's clothing department on the second floor," and adjusts the tone accordingly.

[1716] Providing a visual guide

[1717] In addition to the generated textual response, the device provides visual guidance information, such as a store map and location guide. Visual guidance to assist with store navigation is provided using the MapDisplay module.

[1718] Specific examples

[1719] For example, a user puts on smart glasses and asks, "Where is the red sweater?" The voice data is converted to text through a speech recognition library and sent to a server. The server uses a generative AI model to analyze the intent and determine that directions to the red sweater should be provided. At the same time, it uses an emotion analysis engine to recognize the user's current emotional state and generate a response in a calm tone.

[1720] An example prompt for this would be:

[1721] Analyze the user's question to identify intent and sentiment and generate a response: Question: Where is the red sweater?

[1722] The generated response is accompanied by visual information, such as "The red sweater is in the women's clothing section on the second floor." As a visual guide, a route from the current location to the second floor is displayed on a map of the store. This series of processes allows the user to receive an appropriate response based on their emotions while also providing visual support for in-store navigation, resulting in comfortable and efficient customer support.

[1723] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1724] Step 1:

[1725] The terminal (smart glasses) collects questions entered by the user through voice, and this voice data becomes the input.

[1726] Step 2:

[1727] The device uses a speech recognition library (speech_recognition) to convert voice data into text data. During this conversion process, the device analyzes the voice waveform data and outputs the corresponding text.

[1728] Step 3:

[1729] The terminal sends the converted text data to the server, which receives the data and proceeds to the next analysis step.

[1730] Step 4:

[1731] The server uses a generative AI model (OpenAI's text-davinci-003) to analyze the text data and identify the intent of the user's question. It uses a prompt sentence for the analysis and outputs a model for generating an appropriate response from the user's question. An example of a prompt sentence in this case is, "Analyze the user's question to identify the intent and sentiment, and generate a response. Question: Where is the red sweater?"

[1732] Step 5:

[1733] The server uses an emotion analysis engine to analyze emotions from the user's text data, and the emotion analysis engine analyzes emotional expressions contained in the input text and outputs the user's emotional state.

[1734] Step 6:

[1735] The server generates a response in an appropriate tone based on the results of the generative AI model's analysis and the results of emotion analysis. During this process, it selects an appropriate tone (e.g., calm tone, encouraging tone) according to the intention and emotion, and outputs the response text data.

[1736] Step 7:

[1737] The server generates visual guide information based on the response text. Using the MapDisplay module, the server creates guide information for visually displaying the route from the user's current location to the destination, and sends the data to the terminal.

[1738] Step 8:

[1739] The device provides a response to the user based on the response text and visual guide information received from the server, specifically by playing back a voice response and displaying visual guide information on the smart glasses display.

[1740] Step 9:

[1741] The user inputs feedback on the provided response into the terminal, and this feedback data becomes new input.

[1742] Step 10:

[1743] The device sends user feedback data to the server, which analyzes the received feedback and updates the models of the generative AI engine and sentiment analysis engine, thereby continuously improving the accuracy and appropriateness of the system's responses.

[1744] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1745] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1746] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1747] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1748] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1749] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1750] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1751] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1752] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1753] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1754] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1755] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1756] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1757] 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.

[1758] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1759] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1760] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1761] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1762] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1763] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1764] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1765] The following is further disclosed regarding the above embodiment.

[1766] (Claim 1)

[1767] Using generative AI to automate customer operation instructions and responses to basic questions,

[1768] a means for initializing and managing the generative AI engine;

[1769] means for collecting and displaying multimedia material;

[1770] means for receiving and parsing input data from a user;

[1771] The system includes a means for displaying or playing the response to the user.

[1772] (Claim 2)

[1773] 10. The system of claim 1, further comprising means for self-learning and updating the model of the generative AI engine based on past questions and feedback.

[1774] (Claim 3)

[1775] 10. The system of claim 1, further comprising means for providing a chat support function for responding to customer questions in real time.

[1776] "Example 1"

[1777] (Claim 1)

[1778] Using generative AI to automate customer operation instructions and responses to basic questions,

[1779] a means for initializing and managing the generative AI engine;

[1780] means for collecting and displaying multimedia material;

[1781] means for converting voice input from a user into text;

[1782] a means for analyzing the text and generating an appropriate response;

[1783] means for providing the generated response and multimedia material to a user;

[1784] a means of collecting user feedback and updating the generative AI engine model; and

[1785] A system including:

[1786] (Claim 2)

[1787] 10. The system of claim 1, further comprising means for self-learning and updating the model of the generative AI engine based on past questions and feedback.

[1788] (Claim 3)

[1789] 10. The system of claim 1, further comprising means for providing a chat support function for responding to customer questions in real time.

[1790] "Application Example 1"

[1791] (Claim 1)

[1792] Using generative AI to automate customer operation instructions and responses to basic questions,

[1793] a means for initializing and managing the generative AI engine;

[1794] means for collecting and displaying multimedia material;

[1795] means for receiving and parsing input data from a user;

[1796] means for displaying or playing the response to the user;

[1797] means for converting a voice input from a user into text data using a voice recognition function;

[1798] a means for using a generative AI model to generate an optimal response to a user's question;

[1799] a means for providing a response to a user in real time using a smartphone;

[1800] A system including:

[1801] (Claim 2)

[1802] 10. The system of claim 1, further comprising means for self-learning and updating the model of the generative AI engine based on past questions and feedback.

[1803] (Claim 3)

[1804] The system according to claim 1, further comprising a means for providing a chat support function for responding to customer questions in real time and a guidance function based on product placement information within the store.

[1805] "Example 2: Combining Emotion Engines"

[1806] (Claim 1)

[1807] Using generative AI to automate customer operation instructions and responses to basic questions,

[1808] means for initializing and managing the generative AI engine and the sentiment analysis engine;

[1809] means for converting voice data to text and receiving and analyzing input data;

[1810] means for analyzing the emotional state of the user and adjusting the response content;

[1811] a means for generating a response based on the intent of the question and the results of sentiment analysis;

[1812] means for collecting and displaying multimedia material;

[1813] means for displaying or playing the response to the user;

[1814] A system including means for receiving and analyzing user feedback to update the model of the generative AI engine.

[1815] (Claim 2)

[1816] 10. The system of claim 1, further comprising means for self-learning and updating the models of the generative AI engine and sentiment analysis engine based on past questions and feedback.

[1817] (Claim 3)

[1818] 10. The system of claim 1, further comprising means for providing a chat support function for responding to customer questions in real time.

[1819] "Application example 2 when combining emotion engines"

[1820] (Claim 1)

[1821] Using generative AI to automate customer operation instructions and responses to basic questions,

[1822] a means for initializing and managing the generative AI engine;

[1823] means for collecting and displaying multimedia material;

[1824] means for receiving and parsing input data from a user;

[1825] means for displaying or playing the response to the user;

[1826] A means of analyzing user emotions and adjusting the tone and content of responses;

[1827] a means for providing visual guidance via the smart glasses;

[1828] A system including:

[1829] (Claim 2)

[1830] 10. The system of claim 1, further comprising means for self-learning and updating the model of the generative AI engine based on past questions and feedback.

[1831] (Claim 3)

[1832] 10. The system of claim 1, further comprising means for providing a chat support function for responding to customer questions in real time. [Explanation of symbols]

[1833] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. Using generative AI to automate customer operation instructions and responses to basic questions, a means for initializing and managing the generative AI engine; means for collecting and displaying multimedia material; means for receiving and parsing input data from a user; The system includes a means for displaying or playing the response to the user.

2. 10. The system of claim 1, further comprising means for self-learning and updating the model of the generative AI engine based on past questions and feedback.

3. The system of claim 1, further comprising means for providing a chat support function for responding to customer questions in real time.

Images