System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving input text; means for analyzing a context of the received text; means for generating an optimal translation based on the analyzed context; means for generating a reply based on the generated translation; means for displaying the generated reply; and means for storing past interactions.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] Conventional text translation systems could only translate based on a default context, and were unable to provide translations appropriate for different contexts, such as business, casual, or formal. They also lacked the ability to properly manage past chat history and maintain consistent communication with customers. This made it difficult for many people and companies to communicate appropriately in different situations and contexts. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including a means for receiving input text, a means for analyzing the context of the received text, a means for generating an optimal translation based on the analyzed context, a means for generating a response based on the generated translation, a means for displaying the generated response, and a means for saving past exchanges. This system is characterized in that the context analysis means has an algorithm for classifying text as business, casual, or formal, and the generated translation is text optimized for the context using a natural language processing algorithm. This enables appropriate translation and consistent communication in various contexts.

[0006] "Means for receiving input text" refers to an interface and functionality for receiving text data sent by a user.

[0007] "Means for analyzing the context of received text" refers to algorithms and functions that determine the context of input text data and classify it into specific uses such as business, casual, formal, etc.

[0008] "Means for generating optimal translations based on analyzed context" refers to machine translation and natural language processing technologies for generating the most appropriate language expression based on the identified context.

[0009] The term "means for generating a response based on the generated translation" refers to an algorithm and function that generates a sentence to be displayed as a response to the user based on the text output by the translation engine.

[0010] "Means for displaying a generated response" refers to systems and devices for displaying a generated text response on a user interface.

[0011] "Means for storing past interactions" refers to systems and technologies that record chat history and past interactions in a database or storage and make them available for later reference. [Brief explanation of the drawings]

[0012] [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

[0013] 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.

[0014] First, the terms used in the following description will be explained.

[0015] 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).

[0016] 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.

[0017] 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.

[0018] 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.

[0019] 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."

[0020] [First embodiment]

[0021] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0022] 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.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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."

[0033] The natural language processing system of the present invention is composed of multiple modules, specifically including a means for receiving input text, a means for analyzing the context of the received text, a means for generating an optimal translation based on the analyzed context, a means for generating a response based on the generated translation, a means for displaying the generated response, and a means for saving past exchanges.

[0034] Receiving text input

[0035] A user inputs text through the terminal. For example, if the user inputs "Please check the schedule for business meetings," the terminal receives this text and prepares to send it to the server.

[0036] Contextual analysis

[0037] After the server receives the text sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to analyze the text and classify the context as business, casual, or formal. To do this, it performs keyword extraction and phrase analysis. For example, the text "Please confirm the business meeting schedule" is determined to be a business context.

[0038] Translation Generation

[0039] The server then invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. Using a generative AI model, a phrase optimized for the context is output. For example, "Please confirm the business meeting schedule" is translated to "Could you please confirm the meeting schedule?"

[0040] Response Generation

[0041] Based on the generated translation, the server uses the response generation module to create an appropriate response, which then suggests a specific action to the user. For example, after receiving the translation result, the response generated is "The meeting schedule is confirmed for tomorrow at 3 PM."

[0042] Viewing the response

[0043] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the translation and the response in real time.

[0044] Save chat history

[0045] All conversations are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations are searchable and can be used for future communication.

[0046] As a concrete example of implementation, if someone inputs "Please send the meeting materials" during the preparation stage of a business meeting, the system will perform appropriate context analysis and translation, and ultimately display a reply with an English translation of "Please send the meeting documents." This exchange is also saved on the server for future reference.

[0047] The processing flow will be explained below.

[0048] Step 1:

[0049] The user inputs text through the terminal. For example, the user inputs "Please check the schedule for business meetings."

[0050] Step 2:

[0051] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[0052] Step 3:

[0053] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[0054] Step 4:

[0055] The context analysis module installed on the server analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the schedule for the business meeting" is classified as a business context.

[0056] Step 5:

[0057] The server calls the translation engine module based on the results of the context analysis. The translation engine module generates a translation that is optimal for the business context. It uses a generative AI model to select the most appropriate expression. For example, "Please confirm the business meeting schedule." is output as "Could you please confirm the meeting schedule?"

[0058] Step 6:

[0059] The server receives the translated text generated by the translation engine module and passes it to the response generation module, which then creates a specific response for the user based on the translated text. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM."

[0060] Step 7:

[0061] The server sends the generated response to the terminal.

[0062] Step 8:

[0063] The terminal receives the response sent from the server and displays it to the user, allowing the user to check the response in real time.

[0064] Step 9:

[0065] The server records and stores all conversations using a chat history management module. The stored data includes details such as date and time, input text, analysis results, translation results, and final responses. Past conversations are made available for future communication.

[0066] Example 1

[0067] 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."

[0068] Conventional natural language processing systems have difficulty properly analyzing the context of text and generating optimal translations and responses. They also lack the ability to efficiently store and later search the history of interactions with users. As a result, there are problems with reduced accuracy and efficiency of communication.

[0069] 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.

[0070] In this invention, the server includes means for receiving text entered by a user through a terminal, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response on the terminal, and means for saving past exchanges. This makes it possible to appropriately analyze the context of the text and generate optimal translations and responses. Furthermore, the history of exchanges with the user can be efficiently saved and later searched for and used.

[0071] "User" refers to an individual or organization that uses the system by inputting text through a terminal.

[0072] "Terminal" refers to a computer device or personal digital assistant that a user uses to enter and send text.

[0073] "Text" refers to a character string or sentence that a user inputs through a terminal.

[0074] "Server" refers to a computer system that processes text received from a terminal and performs analysis, translation, and response generation.

[0075] The "context analysis means" refers to a mechanism that has an algorithm that analyzes received text and classifies the context of the text into business, casual, formal, etc.

[0076] "Translation generation means" refers to a mechanism that has an algorithm that generates an optimal translation based on analyzed context information.

[0077] "Generative AI model" refers to an algorithmic model that uses artificial intelligence techniques to generate translations and responses that are optimized for contextual information.

[0078] A "prompt" refers to a command or question input to a generative AI model.

[0079] The "response generation means" refers to a mechanism having an algorithm that automatically creates a response to the user based on the generated translation.

[0080] The "chat history management means" refers to a database system that stores the history of interactions with users and manages it so that it can be searched and referenced later.

[0081] The natural language processing system of the present invention is composed of multiple modules. In this system, the user, terminal, and server each play their respective roles and operate smoothly to analyze input text, translate, generate responses, and manage history.

[0082] First, a user inputs text through a device such as a PC or smartphone. For example, if the user inputs "Please confirm the business meeting schedule," the device packages the input text in JSON format and sends it to the server as an HTTP POST request. The device used can be a general computer or a mobile information terminal, and the software can be React or Vue.js.

[0083] Next, the server receives this HTTP request sent from the device. The server passes the received text data to a context analysis module. This module uses a natural language processing library (e.g., spaCy or NLTK) to tokenize the text and extract keywords, then classifies the context as business, casual, or formal. For example, "Please confirm the schedule for the business meeting" is considered a business context.

[0084] The server then invokes a translation engine module based on the context analysis. This module generates the optimal translation based on the analyzed context information. Specifically, a generative AI model (e.g., OpenAI GPT-3) is used to output a phrase optimized for the context. For example, "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?" An example of this prompt is "Translate this business-related text to English: Please confirm the business meeting schedule."

[0085] The server then uses a response generation module to generate an appropriate response based on the translated text. The response generation module generates a response that indicates a specific action based on the translated text. For example, in response to the translated text "Could you please confirm the meeting schedule?", the server generates the response "The meeting schedule is confirmed for tomorrow at 3 PM."

[0086] The generated response is sent from the server to the device, which then displays it to the user using a front-end framework such as React or Vue.js, providing the response to the user in real time.

[0087] Finally, the server stores all conversations through a chat history management module, which records details of each chat, such as the date and time, content, translation results, and responses, in a database (e.g., MySQL or PostgreSQL), efficiently storing past conversations so that future communications can be easily accessed and utilized.

[0088] As a concrete example of implementation, when preparing for a business meeting, a user may input "Please send the meeting materials." The system will perform appropriate context analysis and translation, and ultimately display a response to the user with an English translation: "Please send the meeting documents." All of this communication is also stored on the server for future reference.

[0089] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0090] Step 1:

[0091] The user inputs text through the terminal. Specifically, the user enters "Please confirm the business meeting schedule" into the text input field on the terminal and clicks the send button. The terminal packages this input text in JSON format and sends it to the server as an HTTP POST request. The input is the text from the user, and the output is an HTTP request to the server.

[0092] Step 2:

[0093] The server receives an HTTP request sent from a terminal. The server parses the received JSON formatted text data and passes it to the context analysis module. The input is the content of the HTTP request, and the output is the input data for the context analysis module.

[0094] Step 3:

[0095] The context analysis module analyzes the received text data using a natural language processing library (e.g., spaCy or NLTK). Specifically, it tokenizes the text, extracts keywords, and classifies the context as business, casual, or formal. The input is the text to be analyzed, and the output is context information. For example, "Please confirm the schedule for the business meeting" is classified as a business context.

[0096] Step 4:

[0097] The server calls the translation engine module based on the context information obtained from the context analysis module. The translation engine module uses a generative AI model (e.g., OpenAI GPT-3) to generate a translation optimized for the context information. The input is the context information and the original text, and the output is the translation. "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?"

[0098] Step 5:

[0099] The server uses a response generation module to create an appropriate response based on the generated translation. The response generation module generates a response indicating a specific action based on the translated text. The input is the translation and the output is the response. For example, in response to the question, "Could you please confirm the meeting schedule?", the response generated is, "The meeting schedule is confirmed for tomorrow at 3 PM."

[0100] Step 6:

[0101] The server sends the generated response to the terminal. The terminal receives it and displays it on the user interface. The input is the response from the server, and the output is the response displayed on the user interface. A front-end framework such as React or Vue.js is used for display.

[0102] Step 7:

[0103] All conversations are saved in the chat history management module on the server. The chat history management module records detailed information such as the date and time of each chat, content, translation results, and responses in a database (e.g., MySQL or PostgreSQL). The input is chat details, and the output is historical data saved in the database. This allows for later search and reference.

[0104] (Application example 1)

[0105] 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."

[0106] The purpose of this invention is to provide a customer support system that can quickly and accurately automatically generate translations and responses to multilingual inquiries from users. In particular, it solves the problem of efficiently responding to inquiries from a variety of users by performing context analysis, generating appropriate translations, and then generating responses based on those translations.

[0107] 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.

[0108] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past exchanges, means for receiving, analyzing, and translating user input in multiple languages ​​to generate a response, and means for displaying the generated translation and response in a chat format and maintaining a history, thereby enabling a prompt and accurate response to user inquiries.

[0109] "Entered text" means textual information entered by a user using a device.

[0110] "Context" is background information that helps understand the meaning and intent of the input text.

[0111] A "translation" is a text that is converted from an input text into a different language.

[0112] A "response" is a response provided by the system to a user's inquiry or request.

[0113] "Display means" means a device or software that displays the generated translation or response in a user-friendly format.

[0114] "Storage means" is a function for recording past interactions and data so that they can be retrieved later.

[0115] "Contextual analysis means" is an analytical function that understands the background and meaning of input text and performs appropriate classification and interpretation.

[0116] "Multilingual input" refers to a user entering text in multiple different languages.

[0117] A "natural language processing algorithm" is a set of procedures and calculations that allow a computer to understand and process human language.

[0118] A "generative AI model" is a trained model that uses artificial intelligence to analyze text and generate appropriate translations and responses.

[0119] "Chat format" is a format that displays alternating dialogue between the user and the system.

[0120] "History retention" is a function that records past inquiries and responses so that they can be accessed later.

[0121] The present invention has as its main object to provide a multilingual customer support system. Detailed embodiments of the invention will be described below.

[0122] This system implements a series of processes that receive input from users, analyze the context, generate appropriate translations, and return responses. The system consists of a server and a user terminal.

[0123] Program Overview

[0124] First, the user inputs text using the terminal. For example, if the user inputs "Please check the delivery status of the product," the terminal prepares to send this text to the server.

[0125] When the server receives the text from the device, it passes it to a context analysis module. This module uses natural language processing algorithms to analyze the context of the text and generate a translation appropriate for that context. For example, the text "Please check the delivery status of your item" is determined to be in a business context.

[0126] The server then invokes a translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used to output a translation optimized for the context.

[0127] Based on the generated translation, the server uses a response generation module to create an appropriate response, which then suggests a specific action to the user. For example, in response to the query "Please check the delivery status of your product," the server generates the response "The delivery status of your product has been confirmed and it is on its way."

[0128] The generated response is sent from the server to the terminal, which displays it to the user, who can check the translation and the response in real time.

[0129] Furthermore, all conversations are saved by a chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[0130] Hardware and Software Configuration

[0131] Hardware:

[0132] User device (smartphone or PC)

[0133] Server (Cloud server or on-premise server)

[0134] software:

[0135] Natural Language Processing Algorithms

[0136] Google Translation API (translation engine)

[0137] Python 3.x

[0138] The Requests library (for sending HTTP requests)

[0139] The server analyzes and translates the input text to quickly and accurately respond to user inquiries, generating and displaying a response, enabling users to receive smooth support even for inquiries in multiple languages.

[0140] Examples and prompts

[0141] For example, if a user types "Please check the delivery status of your product," the system analyzes the context, determines it is a business context, and calls the translation engine to translate it into English. The final response displayed to the user is "The delivery status of your product has been confirmed and it is on its way."

[0142] An example of an input prompt for the generative AI model is as follows:

[0143] text

[0144] If a user types "Please check the delivery status of my item," analyze the context, determine that it is a business context, call the translation engine to translate it into English, and generate an appropriate response to display to the user.

[0145] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0146] Step 1:

[0147] A user inputs text using a terminal. For example, the user inputs "Please check the delivery status of the product." This input text is received by the terminal as an instruction from the user. The received text is prepared to be sent to the server as is.

[0148] Step 2:

[0149] The terminal sends the input text to the server. The text sent from the terminal is received by the server. The received text proceeds to the next processing stage.

[0150] Step 3:

[0151] The server passes the received text to the context analysis module, which uses natural language processing algorithms to analyze the context of the text. For example, the text "Please check the delivery status of your item" is determined to be a business context. The input is text, and the output is context information for the text.

[0152] Step 4:

[0153] The server invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used here to output a context-optimized translation (e.g., "Could you please confirm the delivery status of the product?"). The input is context information and text, and the output is the translated text.

[0154] Step 5:

[0155] The server uses a response generation module to create an appropriate response based on the generated translation. This module uses an AI-based response algorithm to generate a response that suggests a specific action for the user. For example: "The delivery status of your product has been confirmed and it is on its way." The input is the translation, and the output is a response to the user.

[0156] Step 6:

[0157] The generated response is sent from the server to the device. The device receives the response and displays it to the user. The user can see the translated response in real time. The input is the response from the server, and the output is the text displayed to the user.

[0158] Step 7:

[0159] All exchanges are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. This allows past exchanges to be searched and used for future communication. The input is the exchange data, and the output is the saved chat history.

[0160] 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.

[0161] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, enabling more effective and natural communication. Specifically, it is composed of the following multiple modules:

[0162] A means of receiving input text

[0163] A means of analyzing the context of received text

[0164] Emotion analysis means to recognize user emotions

[0165] A means of generating optimal translations based on analyzed context

[0166] A means of tailoring responses based on perceived emotions

[0167] A means of generating a response based on the generated translation

[0168] A means of displaying the generated response

[0169] A way to preserve past interactions

[0170] Receiving text input

[0171] A user inputs text through the terminal. For example, if the user inputs "Please check the meeting schedule," the terminal receives this text and prepares to send it to the server.

[0172] Context and Sentiment Analysis

[0173] After the server receives the text data sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. At the same time, the sentiment analysis module identifies the user's sentiment (positive, negative, neutral) from the text. For example, the text "Please confirm the meeting schedule" is in a business context and the sentiment is determined to be slightly negative.

[0174] Translation Generation

[0175] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for the business context while also taking negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0176] Response Generation and Adjustment

[0177] Based on the translation generated by the translation engine module, the server uses the response generation module to generate an appropriate response, taking into account the results of sentiment analysis and adjusting it based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0178] Viewing the response

[0179] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the tailored response in real time.

[0180] Save chat history

[0181] All interactions are recorded and saved by the server's chat history management module. This module records detailed information about each chat, such as the date and time, content, context analysis results, sentiment analysis results, translation results, and final responses, in a database. The saved data can be used for future communications.

[0182] As a concrete example of its implementation, when a user inputs something like "Please prepare the materials" during the preparation stage of a business meeting, the system will analyze the user's emotions and provide an appropriately tailored translation and response, resulting in more useful communication for the user. This exchange is also saved on the server for future reference.

[0183] The processing flow will be explained below.

[0184] Step 1:

[0185] A user inputs text through a terminal. For example, the user inputs "Please check the meeting schedule."

[0186] Step 2:

[0187] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[0188] Step 3:

[0189] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[0190] Step 4:

[0191] The server's context analysis module analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the meeting schedule" would be classified as a business context.

[0192] Step 5:

[0193] Based on the results of the context analysis, the server calls the sentiment analysis module. This module determines the user's sentiment from the text. It uses a natural language processing algorithm to classify it as positive, negative, or neutral. For example, the phrase "please" (please) detects an overall negative sentiment.

[0194] Step 6:

[0195] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for business contexts and takes negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0196] Step 7:

[0197] The server receives the translated text generated by the translation engine module and passes it to the response generation module. The response generation module creates a specific response for the user based on the translated text. The response generation module also takes into account the results of sentiment analysis and adjusts based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0198] Step 8:

[0199] The server sends the generated response to the terminal.

[0200] Step 9:

[0201] The terminal receives the response sent from the server and displays it to the user, allowing the user to see the adjusted response in real time.

[0202] Step 10:

[0203] The server records and stores all interactions using a chat history management module. The stored data includes details such as date and time, input text, context analysis results, sentiment analysis results, translation results, and final responses. Past interactions are made available for future communication.

[0204] Example 2

[0205] 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."

[0206] While conventional natural language processing systems can analyze the context of input text and generate translations, it has been difficult to realize communication that takes into account the user's emotions. Therefore, there is a need for a system that can generate natural and effective responses based on the user's emotions.

[0207] 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.

[0208] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the user's emotions, means for generating an optimal translation based on the analyzed context and emotions, means for generating a response based on the generated translation, means for displaying the generated response, and means for saving past exchanges, thereby enabling natural communication that takes into account the user's emotions.

[0209] The "means for receiving input text" is a means for receiving text data input by a user through a terminal and transmitting it for subsequent processing.

[0210] The "means for analyzing the context of received text" refers to a means for extracting keywords and phrases from the received text data and determining the context and category (business, casual, formal, etc.).

[0211] The "means for analyzing user emotions" refers to a means for identifying user emotions (positive, negative, neutral) from received text data using a natural language processing algorithm.

[0212] "Means for generating optimal translations based on analyzed context and sentiment" refers to means for generating appropriate translations based on the results of contextual and sentiment analysis. This is done using a generative AI model.

[0213] The "means for generating a response based on the generated translation" refers to a means for creating an appropriate response using the generated translation, taking into account the results of sentiment analysis.

[0214] The "means for displaying the generated response" is a means for transmitting the generated response text to the terminal and displaying it to the user.

[0215] "Means for storing past interactions" refers to a means for recording and storing details of all interactions (date and time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database.

[0216] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, thereby realizing more effective and natural communication. This system is composed of the following multiple modules.

[0217] System Components

[0218] 1. A means of receiving input text

[0219] When a user inputs text through the terminal, for example, "Please check the meeting schedule," the terminal receives this text and transmits it to the server through the network.

[0220] 2. Means of analyzing the context of received text

[0221] The server receives the text data sent from the device and passes it to a context analysis module, which uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine its context.

[0222] 3. Means of analyzing user emotions

[0223] The server uses a sentiment analysis module to identify the user's sentiment (positive, negative, neutral) from the text. Sentiment analysis algorithms (e.g., VADER and TextBlob) are used to precisely analyze the sentiment of the input text.

[0224] 4. A means to generate optimal translations based on analyzed context and sentiment

[0225] The server then calls the translation engine module based on the analyzed context and sentiment information. Using a generative AI model (e.g., GPT-3), it generates a translation that best fits the context and sentiment. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0226] 5. A means of generating a response based on the generated translation

[0227] The server uses the response generation module to generate an appropriate response based on the translation. Based on the results of sentiment analysis, the response is fine-tuned based on the sentiment. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0228] 6. A way to display the generated response

[0229] The generated response is sent from the server to the terminal, which displays the received response on its screen so that the user can check it in real time.

[0230] 7. A way to preserve past interactions

[0231] All interactions are recorded and stored by the chat history management module on the server. Details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) are stored in a database (e.g., MySQL or MongoDB) for future reference and analysis.

[0232] Specific examples

[0233] If a user types "Please prepare the materials" while preparing for a business meeting, the system will generate an appropriate response through the following steps. First, the device sends the input text to the server, which performs contextual and sentiment analysis. Contextual analysis determines the context as "business," and sentiment analysis identifies it as "somewhat urgent." The translation engine translates it as "Please prepare the materials urgently," and the response generation module generates the reply, "The materials will be prepared and sent to you shortly. We appreciate your prompt request." The device displays this response to the user, and the details of the exchange are stored on the server.

[0234] Prompt Sentence Examples

[0235] "Please check the meeting schedule, and I'd also like to discuss a new project."

[0236] "Please prepare the materials."

[0237] In this way, the present invention is a system that realizes natural and effective communication for users.

[0238] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0239] The flow of this system's program processing

[0240] Step 1:

[0241] A user inputs text through a terminal. The input text (e.g., "Please check the meeting schedule.") is received by the terminal and temporarily stored in memory. The terminal then converts the received text into network packets and prepares them to be sent to the server. The input is text data, and the output is network packets for transmission.

[0242] Step 2:

[0243] The server receives text data sent from the device via the network. The server converts the received data into an appropriate format for analysis and passes it to the context analysis and sentiment analysis modules. The input is network packets, and the output is a parseable data structure (e.g., JSON format).

[0244] Step 3:

[0245] The server's context analysis module uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine their context. The server then passes the extracted keywords, phrases, and context categories to the next processing step. The input is the text data, and the output is the context analysis results.

[0246] Step 4:

[0247] The server's sentiment analysis module uses a natural language processing algorithm (e.g., VADER or TextBlob) to identify the user's sentiment from the text. The server passes the result of the sentiment analysis (positive, negative, or neutral) to the next processing step. The input is the text data, and the output is the sentiment analysis result.

[0248] Step 5:

[0249] Based on the analyzed context and sentiment information, the server invokes the translation engine module and generates the optimal translation using a generative AI model (e.g., GPT-3). The generated translation (e.g., "Could you please confirm the meeting schedule? Thank you for your understanding.") is passed to the next step. The input is the results of context analysis and sentiment analysis, and the output is the optimized translation.

[0250] Step 6:

[0251] The server uses a response generation module to generate an appropriate response based on the generated translation. Based on the results of sentiment analysis, the response is fine-tuned based on sentiment (e.g., "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."). The input is the translation, and the output is the generated response.

[0252] Step 7:

[0253] The generated response is sent from the server to the terminal. The server converts the response into a network packet and sends it to the terminal. The input is the generated response, and the output is the network packet for transmission.

[0254] Step 8:

[0255] The terminal receives the response sent from the server and displays it on the screen. The input is the network packet of the response, and the output is the displayed text that the user can see. This allows the user to see the adjusted response in real time.

[0256] Step 9:

[0257] All interactions are recorded and stored by the server's chat history management module. The server stores details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database (e.g., MySQL or MongoDB) for future reference and analysis. The input is chat details data, and the output is stored database entries.

[0258] Through the above process, this system realizes natural communication that takes into account the user's emotions and context.

[0259] (Application example 2)

[0260] 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."

[0261] To ensure smooth communication with customers in physical stores, it is necessary to respond quickly and appropriately to their questions and requests. However, it is not easy to accurately understand the customer's emotions and context and respond based on them. In particular, it is difficult to respond appropriately according to the customer's emotional state, which can lead to a decrease in customer satisfaction and an unsatisfactory experience.

[0262] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0263] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the emotion of the user, means for generating an optimal translation based on the analyzed context, means for adjusting a response based on the recognized emotion, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past interactions, and means for supporting customer service in a physical store. This allows for the rapid generation of appropriate responses that take into account the emotion and context of the customer, thereby improving the efficiency of customer service in a physical store and customer satisfaction.

[0264] The "means for receiving input text" is a function for receiving text input from a user and sending the data to a server for analysis.

[0265] The "means for analyzing the context of received text" is a function for analyzing received text data and understanding the context according to the meaning and use of the text.

[0266] "Emotion analysis means for recognizing user emotions" is a function for analyzing and identifying user emotions (positive, negative, neutral, etc.) based on keywords and phrases in the text.

[0267] The "means for generating optimal translations based on analyzed context" is a function for generating appropriate and natural translations according to the context and sentiment.

[0268] The "means for adjusting a response based on a recognized emotion" is a function for taking into account the emotional state of the user and adjusting the response so that an optimized response is generated.

[0269] The "means for generating a response based on the generated translation" is a function for generating an appropriate response based on the translated text.

[0270] The "means for displaying the generated response" is a function for displaying the generated response in a form that is easy for the user to see.

[0271] "Means for storing past interactions" is a function that records all interactions as data and stores them for future reference.

[0272] "Means to support customer service in physical stores" refers to support functions that facilitate smooth communication with customers in physical stores.

[0273] This invention is a system for facilitating customer service in brick-and-mortar stores. The system receives user input text, analyzes context and sentiment to generate an optimal translation, and provides a series of functions to display a response based on that. Specific embodiments are described below.

[0274] System configuration

[0275] The system mainly consists of the following elements:

[0276] A means of receiving input text

[0277] A means of analyzing the context of received text

[0278] Emotion analysis means to recognize user emotions

[0279] A means of generating optimal translations based on analyzed context

[0280] A means of tailoring responses based on perceived emotions

[0281] A means of generating a response based on the generated translation

[0282] A means of displaying the generated response

[0283] A way to preserve past interactions

[0284] A means to support customer service in physical stores

[0285] System operation details

[0286] Receiving text input

[0287] A user inputs text using a terminal (e.g., a smartphone) in a physical store. For example, if the user inputs "Please check product inventory," the terminal receives this text and prepares it for transmission to the server.

[0288] Context and Sentiment Analysis

[0289] When the server receives text from a device, it first passes it to a context analysis module, which uses natural language processing algorithms (for example, Python's Hugging Face library) to extract keywords and phrases from the text and analyze its context. At the same time, a sentiment analysis module identifies the user's sentiment from the text.

[0290] For example, the text "Please check product availability" is a request to check product availability, and the sentiment is judged to be slightly negative.

[0291] Translation generation and adjustment

[0292] Based on the analyzed context and sentiment information, the server calls the translation engine module, which generates a translation that best suits the context and sentiment, such as "Could you please confirm the stock availability? Thank you."

[0293] The system then tailors the response based on the recognized sentiment. For example, for text with a negative sentiment like "Please check the product availability," it generates a sentiment-sensitive response. In this case, it generates a response like "The stock availability has been checked. Thank you for your patience."

[0294] View and save responses

[0295] The generated response is sent from the server to the terminal, which then displays it to the user, allowing the user to see the response tailored to the situation in real time.

[0296] All interactions are recorded and saved by the server's chat history management module, which records detailed information such as the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses in a database for future reference and analysis.

[0297] Examples of concrete examples and prompts

[0298] For example, consider the following exchange:

[0299] User Input:

[0300] "Please check product availability."

[0301] System response:

[0302] "The stock availability has been checked. Thank you for your patience."

[0303] Example prompt sentence:

[0304] "A user types, 'Check product availability.' Use contextual analysis and sentiment identification to generate an appropriate response and translation."

[0305] Contextual analysis results: Check product availability

[0306] Emotion identification result: Negative

[0307] Translation to generate: The stock availability has been checked.

[0308] Generate response: We have confirmed the product is in stock. Thank you.

[0309] This system will enable the rapid generation of appropriate responses that take into account customer emotions and context, improving customer service efficiency and customer satisfaction in physical stores.

[0310] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0311] Step 1:

[0312] The user inputs text. The user uses a device such as a smartphone in a physical store to input a question or request about a product. An example of input text is "Please check product availability." The device receives this text.

[0313] Input: User text input

[0314] Output: Received text data

[0315] Step 2:

[0316] The terminal sends the received text to the server. At this time, the text data is safely delivered to the server via the network. The server receives this text data.

[0317] Input: Text data sent from the terminal

[0318] Output: Text data received by the server

[0319] Step 3:

[0320] The server passes the received text data to a context analysis module, which uses a natural language processing algorithm (e.g., Python's Hugging Face library) to extract keywords and phrases from the text. For example, it extracts "product inventory" and determines that the context is a business context.

[0321] Input: Text data received by the server

[0322] Output: Context information (e.g. product availability, business context)

[0323] Step 4:

[0324] The server simultaneously passes the text to a sentiment analysis module, which identifies the user's sentiment from the text. For example, a request like "Please check product availability" is identified as a slightly negative sentiment. This analysis is performed using a generative AI model.

[0325] Input: Text data received by the server

[0326] Output: Emotional information (e.g., negative)

[0327] Step 5:

[0328] The server passes the results of the contextual and sentiment analysis to the translation engine module, which generates the optimal translation based on the analysis results. For example, the generated text might be "Can you please check the stock availability? Thank you." A generative AI model is also used at this stage.

[0329] Input: Contextual and emotional information

[0330] Output:Translated text

[0331] Step 6:

[0332] The translated text is passed to a response generation module, which adjusts the response based on the sentiment analysis results. For example, if the sentiment is negative, a more considerate response such as "The stock availability has been checked. Thank you for your patience" will be generated.

[0333] Input: Translation text and sentiment information

[0334] Output: Adjusted response text

[0335] Step 7:

[0336] The server sends the generated response text to the terminal, which displays it to the user, who can see the appropriately tailored response in real time.

[0337] Input: Reply text sent from the server

[0338] Output: The response text displayed to the user

[0339] Step 8:

[0340] All interactions are recorded and saved by the chat history management module on the server, including the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses. The saved data is available for future reference and analysis.

[0341] Input: Various generated data (date, time, content, analysis results, final response, etc.)

[0342] Output: Saved chat history

[0343] 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.

[0344] 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.

[0345] 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.

[0346] [Second embodiment]

[0347] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0348] 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.

[0349] 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).

[0350] 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.

[0351] 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.

[0352] 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).

[0353] 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.

[0354] 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.

[0355] 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.

[0356] 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.

[0357] 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.

[0358] 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."

[0359] The natural language processing system of the present invention is composed of multiple modules, specifically including a means for receiving input text, a means for analyzing the context of the received text, a means for generating an optimal translation based on the analyzed context, a means for generating a response based on the generated translation, a means for displaying the generated response, and a means for saving past exchanges.

[0360] Receiving text input

[0361] A user inputs text through the terminal. For example, if the user inputs "Please check the schedule for business meetings," the terminal receives this text and prepares to send it to the server.

[0362] Contextual analysis

[0363] After the server receives the text sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to analyze the text and classify the context as business, casual, or formal. To do this, it performs keyword extraction and phrase analysis. For example, the text "Please confirm the business meeting schedule" is determined to be a business context.

[0364] Translation Generation

[0365] The server then invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. Using a generative AI model, a phrase optimized for the context is output. For example, "Please confirm the business meeting schedule" is translated to "Could you please confirm the meeting schedule?"

[0366] Response Generation

[0367] Based on the generated translation, the server uses the response generation module to create an appropriate response, which then suggests a specific action to the user. For example, after receiving the translation result, the response generated is "The meeting schedule is confirmed for tomorrow at 3 PM."

[0368] Viewing the response

[0369] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the translation and the response in real time.

[0370] Save chat history

[0371] All conversations are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[0372] As a concrete example of implementation, if someone inputs "Please send the meeting materials" during the preparation stage of a business meeting, the system will perform appropriate context analysis and translation, and ultimately display a reply with an English translation of "Please send the meeting documents." This exchange is also saved on the server for future reference.

[0373] The processing flow will be explained below.

[0374] Step 1:

[0375] The user inputs text through the terminal. For example, the user inputs "Please check the schedule for business meetings."

[0376] Step 2:

[0377] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[0378] Step 3:

[0379] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[0380] Step 4:

[0381] The context analysis module installed on the server analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the schedule for the business meeting" is classified as a business context.

[0382] Step 5:

[0383] The server calls the translation engine module based on the results of the context analysis. The translation engine module generates a translation that is optimal for the business context. It uses a generative AI model to select the most appropriate expression. For example, "Please confirm the business meeting schedule." is output as "Could you please confirm the meeting schedule?"

[0384] Step 6:

[0385] The server receives the translated text generated by the translation engine module and passes it to the response generation module, which then creates a specific response for the user based on the translated text. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM."

[0386] Step 7:

[0387] The server sends the generated response to the terminal.

[0388] Step 8:

[0389] The terminal receives the response sent from the server and displays it to the user, allowing the user to check the response in real time.

[0390] Step 9:

[0391] The server records and stores all conversations using a chat history management module. The stored data includes details such as date and time, input text, analysis results, translation results, and final responses. Past conversations are made available for future communication.

[0392] Example 1

[0393] 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."

[0394] Conventional natural language processing systems have difficulty properly analyzing the context of text and generating optimal translations and responses. They also lack the ability to efficiently store and later search the history of interactions with users. As a result, there are problems with reduced accuracy and efficiency of communication.

[0395] 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.

[0396] In this invention, the server includes means for receiving text entered by a user through a terminal, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response on the terminal, and means for saving past exchanges. This makes it possible to appropriately analyze the context of the text and generate optimal translations and responses. Furthermore, the history of exchanges with the user can be efficiently saved and later searched for and used.

[0397] "User" refers to an individual or organization that uses the system by inputting text through a terminal.

[0398] "Terminal" refers to a computer device or personal digital assistant that a user uses to enter and send text.

[0399] "Text" refers to a character string or sentence that a user inputs through a terminal.

[0400] "Server" refers to a computer system that processes text received from a terminal and performs analysis, translation, and response generation.

[0401] The "context analysis means" refers to a mechanism that has an algorithm that analyzes received text and classifies the context of the text into business, casual, formal, etc.

[0402] "Translation generation means" refers to a mechanism that has an algorithm that generates an optimal translation based on analyzed context information.

[0403] "Generative AI model" refers to an algorithmic model that uses artificial intelligence techniques to generate translations and responses that are optimized for contextual information.

[0404] A "prompt" refers to a command or question input to a generative AI model.

[0405] The "response generation means" refers to a mechanism having an algorithm that automatically creates a response to the user based on the generated translation.

[0406] The "chat history management means" refers to a database system that stores the history of interactions with users and manages it so that it can be searched and referenced later.

[0407] The natural language processing system of the present invention is composed of multiple modules. In this system, the user, terminal, and server each play their respective roles and operate smoothly to analyze input text, translate, generate responses, and manage history.

[0408] First, a user inputs text through a device such as a PC or smartphone. For example, if the user inputs "Please confirm the business meeting schedule," the device packages the input text in JSON format and sends it to the server as an HTTP POST request. The device used can be a general computer or a mobile information terminal, and the software can be React or Vue.js.

[0409] Next, the server receives this HTTP request sent from the device. The server passes the received text data to a context analysis module. This module uses a natural language processing library (e.g., spaCy or NLTK) to tokenize the text and extract keywords, then classifies the context as business, casual, or formal. For example, "Please confirm the schedule for the business meeting" is considered a business context.

[0410] The server then invokes a translation engine module based on the context analysis. This module generates the optimal translation based on the analyzed context information. Specifically, a generative AI model (e.g., OpenAI GPT-3) is used to output a phrase optimized for the context. For example, "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?" An example of this prompt is "Translate this business-related text to English: Please confirm the business meeting schedule."

[0411] The server then uses a response generation module to generate an appropriate response based on the translated text. The response generation module generates a response that indicates a specific action based on the translated text. For example, in response to the translated text "Could you please confirm the meeting schedule?", the server generates the response "The meeting schedule is confirmed for tomorrow at 3 PM."

[0412] The generated response is sent from the server to the device, which then displays it to the user using a front-end framework such as React or Vue.js, providing the response to the user in real time.

[0413] Finally, the server stores all conversations through a chat history management module, which records details of each chat, such as the date and time, content, translation results, and responses, in a database (e.g., MySQL or PostgreSQL), efficiently storing past conversations so that future communications can be easily accessed and utilized.

[0414] As a concrete example of implementation, when preparing for a business meeting, a user may input "Please send the meeting materials." The system will perform appropriate context analysis and translation, and ultimately display a response to the user with an English translation: "Please send the meeting documents." All of this communication is also stored on the server for future reference.

[0415] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0416] Step 1:

[0417] The user inputs text through the terminal. Specifically, the user enters "Please confirm the business meeting schedule" into the text input field on the terminal and clicks the send button. The terminal packages this input text in JSON format and sends it to the server as an HTTP POST request. The input is the text from the user, and the output is an HTTP request to the server.

[0418] Step 2:

[0419] The server receives an HTTP request sent from a terminal. The server parses the received JSON formatted text data and passes it to the context analysis module. The input is the content of the HTTP request, and the output is the input data for the context analysis module.

[0420] Step 3:

[0421] The context analysis module analyzes the received text data using a natural language processing library (e.g., spaCy or NLTK). Specifically, it tokenizes the text, extracts keywords, and classifies the context as business, casual, or formal. The input is the text to be analyzed, and the output is context information. For example, "Please confirm the schedule for the business meeting" is classified as a business context.

[0422] Step 4:

[0423] The server calls the translation engine module based on the context information obtained from the context analysis module. The translation engine module uses a generative AI model (e.g., OpenAI GPT-3) to generate a translation optimized for the context information. The input is the context information and the original text, and the output is the translation. "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?"

[0424] Step 5:

[0425] The server uses a response generation module to create an appropriate response based on the generated translation. The response generation module generates a response indicating a specific action based on the translated text. The input is the translation and the output is the response. For example, in response to the question, "Could you please confirm the meeting schedule?", the response generated is, "The meeting schedule is confirmed for tomorrow at 3 PM."

[0426] Step 6:

[0427] The server sends the generated response to the terminal. The terminal receives it and displays it on the user interface. The input is the response from the server, and the output is the response displayed on the user interface. A front-end framework such as React or Vue.js is used for display.

[0428] Step 7:

[0429] All conversations are saved in the chat history management module on the server. The chat history management module records detailed information such as the date and time of each chat, content, translation results, and responses in a database (e.g., MySQL or PostgreSQL). The input is chat details, and the output is historical data saved in the database. This allows for later search and reference.

[0430] (Application example 1)

[0431] 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."

[0432] The purpose of this invention is to provide a customer support system that can quickly and accurately automatically generate translations and responses to multilingual inquiries from users. In particular, it solves the problem of efficiently responding to inquiries from a variety of users by performing context analysis, generating appropriate translations, and then generating responses based on those translations.

[0433] 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.

[0434] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past exchanges, means for receiving, analyzing, and translating user input in multiple languages ​​to generate a response, and means for displaying the generated translation and response in a chat format and maintaining a history, thereby enabling a prompt and accurate response to user inquiries.

[0435] "Entered text" means textual information entered by a user using a device.

[0436] "Context" is background information that helps understand the meaning and intent of the input text.

[0437] A "translation" is a text that is converted from an input text into a different language.

[0438] A "response" is a response provided by the system to a user's inquiry or request.

[0439] "Display means" means a device or software that displays the generated translation or response in a user-friendly format.

[0440] "Storage means" is a function for recording past interactions and data so that they can be retrieved later.

[0441] "Contextual analysis means" is an analytical function that understands the background and meaning of input text and performs appropriate classification and interpretation.

[0442] "Multilingual input" refers to a user entering text in multiple different languages.

[0443] A "natural language processing algorithm" is a set of procedures and calculations that allow a computer to understand and process human language.

[0444] A "generative AI model" is a trained model that uses artificial intelligence to analyze text and generate appropriate translations and responses.

[0445] "Chat format" is a format that displays alternating dialogue between the user and the system.

[0446] "History retention" is a function that records past inquiries and responses so that they can be accessed later.

[0447] The present invention has as its main object to provide a multilingual customer support system. Detailed embodiments of the invention will be described below.

[0448] This system implements a series of processes that receive input from users, analyze the context, generate appropriate translations, and return responses. The system consists of a server and a user terminal.

[0449] Program Overview

[0450] First, the user inputs text using the terminal. For example, if the user inputs "Please check the delivery status of the product," the terminal prepares to send this text to the server.

[0451] When the server receives the text from the device, it passes it to a context analysis module. This module uses natural language processing algorithms to analyze the context of the text and generate a translation appropriate for that context. For example, the text "Please check the delivery status of your item" is determined to be in a business context.

[0452] The server then invokes a translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used to output a translation optimized for the context.

[0453] Based on the generated translation, the server uses a response generation module to create an appropriate response, which then suggests a specific action to the user. For example, in response to the query "Please check the delivery status of your product," the server generates the response "The delivery status of your product has been confirmed and it is on its way."

[0454] The generated response is sent from the server to the terminal, which displays it to the user, who can check the translation and the response in real time.

[0455] Furthermore, all conversations are saved by a chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[0456] Hardware and Software Configuration

[0457] Hardware:

[0458] User device (smartphone or PC)

[0459] Server (Cloud server or on-premise server)

[0460] software:

[0461] Natural Language Processing Algorithms

[0462] Google Translation API (translation engine)

[0463] Python 3.x

[0464] The Requests library (for sending HTTP requests)

[0465] The server analyzes and translates the input text to quickly and accurately respond to user inquiries, generating and displaying a response, enabling users to receive smooth support even for inquiries in multiple languages.

[0466] Examples and prompts

[0467] For example, if a user types "Please check the delivery status of your product," the system analyzes the context, determines it is a business context, and calls the translation engine to translate it into English. The final response displayed to the user is "The delivery status of your product has been confirmed and it is on its way."

[0468] An example of an input prompt for the generative AI model is as follows:

[0469] text

[0470] If a user types "Please check the delivery status of my item," analyze the context, determine that it is a business context, call the translation engine to translate it into English, and generate an appropriate response to display to the user.

[0471] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0472] Step 1:

[0473] A user inputs text using a terminal. For example, the user inputs "Please check the delivery status of the product." This input text is received by the terminal as an instruction from the user. The received text is prepared to be sent to the server as is.

[0474] Step 2:

[0475] The terminal sends the input text to the server. The text sent from the terminal is received by the server. The received text proceeds to the next processing stage.

[0476] Step 3:

[0477] The server passes the received text to the context analysis module, which uses natural language processing algorithms to analyze the context of the text. For example, the text "Please check the delivery status of your item" is determined to be a business context. The input is text, and the output is context information for the text.

[0478] Step 4:

[0479] The server invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used here to output a context-optimized translation (e.g., "Could you please confirm the delivery status of the product?"). The input is context information and text, and the output is the translated text.

[0480] Step 5:

[0481] The server uses a response generation module to create an appropriate response based on the generated translation. This module uses an AI-based response algorithm to generate a response that suggests a specific action for the user. For example: "The delivery status of your product has been confirmed and it is on its way." The input is the translation, and the output is a response to the user.

[0482] Step 6:

[0483] The generated response is sent from the server to the device. The device receives the response and displays it to the user. The user can see the translated response in real time. The input is the response from the server, and the output is the text displayed to the user.

[0484] Step 7:

[0485] All exchanges are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. This allows past exchanges to be searched and used for future communication. The input is the exchange data, and the output is the saved chat history.

[0486] 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.

[0487] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, enabling more effective and natural communication. Specifically, it is composed of the following multiple modules:

[0488] A means of receiving input text

[0489] A means of analyzing the context of received text

[0490] Emotion analysis means to recognize user emotions

[0491] A means of generating optimal translations based on analyzed context

[0492] A means of tailoring responses based on perceived emotions

[0493] A means of generating a response based on the generated translation

[0494] A means of displaying the generated response

[0495] A way to preserve past interactions

[0496] Receiving text input

[0497] A user inputs text through the terminal. For example, if the user inputs "Please check the meeting schedule," the terminal receives this text and prepares to send it to the server.

[0498] Context and Sentiment Analysis

[0499] After the server receives the text data sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. At the same time, the sentiment analysis module identifies the user's sentiment (positive, negative, neutral) from the text. For example, the text "Please confirm the meeting schedule" is in a business context and the sentiment is determined to be slightly negative.

[0500] Translation Generation

[0501] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for the business context while also taking negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0502] Response Generation and Adjustment

[0503] Based on the translation generated by the translation engine module, the server uses the response generation module to generate an appropriate response, taking into account the results of sentiment analysis and adjusting it based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0504] Viewing the response

[0505] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the tailored response in real time.

[0506] Save chat history

[0507] All interactions are recorded and saved by the server's chat history management module. This module records detailed information about each chat, such as the date and time, content, context analysis results, sentiment analysis results, translation results, and final responses, in a database. The saved data can be used for future communications.

[0508] As a concrete example of its implementation, when a user inputs something like "Please prepare the materials" during the preparation stage of a business meeting, the system will analyze the user's emotions and provide an appropriately tailored translation and response, resulting in more useful communication for the user. This exchange is also saved on the server for future reference.

[0509] The processing flow will be explained below.

[0510] Step 1:

[0511] A user inputs text through a terminal. For example, the user inputs "Please check the meeting schedule."

[0512] Step 2:

[0513] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[0514] Step 3:

[0515] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[0516] Step 4:

[0517] The server's context analysis module analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the meeting schedule" would be classified as a business context.

[0518] Step 5:

[0519] Based on the results of the context analysis, the server calls the sentiment analysis module. This module determines the user's sentiment from the text. It uses a natural language processing algorithm to classify it as positive, negative, or neutral. For example, the phrase "please" (please) detects an overall negative sentiment.

[0520] Step 6:

[0521] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for business contexts and takes negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0522] Step 7:

[0523] The server receives the translated text generated by the translation engine module and passes it to the response generation module. The response generation module creates a specific response for the user based on the translated text. The response generation module also takes into account the results of sentiment analysis and adjusts based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0524] Step 8:

[0525] The server sends the generated response to the terminal.

[0526] Step 9:

[0527] The terminal receives the response sent from the server and displays it to the user, allowing the user to see the adjusted response in real time.

[0528] Step 10:

[0529] The server records and stores all interactions using a chat history management module. The stored data includes details such as date and time, input text, context analysis results, sentiment analysis results, translation results, and final responses. Past interactions are made available for future communication.

[0530] Example 2

[0531] 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."

[0532] While conventional natural language processing systems can analyze the context of input text and generate translations, it has been difficult to realize communication that takes into account the user's emotions. Therefore, there is a need for a system that can generate natural and effective responses based on the user's emotions.

[0533] 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.

[0534] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the user's emotions, means for generating an optimal translation based on the analyzed context and emotions, means for generating a response based on the generated translation, means for displaying the generated response, and means for saving past exchanges, thereby enabling natural communication that takes into account the user's emotions.

[0535] The "means for receiving input text" is a means for receiving text data input by a user through a terminal and transmitting it for subsequent processing.

[0536] The "means for analyzing the context of received text" refers to a means for extracting keywords and phrases from the received text data and determining the context and category (business, casual, formal, etc.).

[0537] The "means for analyzing user emotions" refers to a means for identifying user emotions (positive, negative, neutral) from received text data using a natural language processing algorithm.

[0538] "Means for generating optimal translations based on analyzed context and sentiment" refers to means for generating appropriate translations based on the results of contextual and sentiment analysis. This is done using a generative AI model.

[0539] The "means for generating a response based on the generated translation" refers to a means for creating an appropriate response using the generated translation, taking into account the results of sentiment analysis.

[0540] The "means for displaying the generated response" is a means for transmitting the generated response text to the terminal and displaying it to the user.

[0541] "Means for storing past interactions" refers to a means for recording and storing details of all interactions (date and time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database.

[0542] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, thereby realizing more effective and natural communication. This system is composed of the following multiple modules.

[0543] System Components

[0544] 1. A means of receiving input text

[0545] When a user inputs text through the terminal, for example, "Please check the meeting schedule," the terminal receives this text and transmits it to the server through the network.

[0546] 2. Means of analyzing the context of received text

[0547] The server receives the text data sent from the device and passes it to a context analysis module, which uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine its context.

[0548] 3. Means of analyzing user emotions

[0549] The server uses a sentiment analysis module to identify the user's sentiment (positive, negative, neutral) from the text. Sentiment analysis algorithms (e.g., VADER and TextBlob) are used to precisely analyze the sentiment of the input text.

[0550] 4. A means to generate optimal translations based on analyzed context and sentiment

[0551] The server then calls the translation engine module based on the analyzed context and sentiment information. Using a generative AI model (e.g., GPT-3), it generates a translation that best fits the context and sentiment. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0552] 5. A means of generating a response based on the generated translation

[0553] The server uses the response generation module to generate an appropriate response based on the generated translation. Based on the results of sentiment analysis, the response is fine-tuned based on sentiment. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0554] 6. A way to display the generated response

[0555] The generated response is sent from the server to the terminal, which displays the received response on its screen so that the user can check it in real time.

[0556] 7. A way to preserve past interactions

[0557] All interactions are recorded and stored by the chat history management module on the server. Details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) are stored in a database (e.g., MySQL or MongoDB) for future reference and analysis.

[0558] Specific examples

[0559] If a user types "Please prepare the materials" while preparing for a business meeting, the system will generate an appropriate response through the following steps. First, the device sends the input text to the server, which performs contextual and sentiment analysis. Contextual analysis determines the context as "business," and sentiment analysis identifies it as "somewhat urgent." The translation engine translates it as "Please prepare the materials urgently," and the response generation module generates the reply, "The materials will be prepared and sent to you shortly. We appreciate your prompt request." The device displays this response to the user, and the details of the exchange are stored on the server.

[0560] Prompt Sentence Examples

[0561] "Please check the meeting schedule, and I'd also like to discuss a new project."

[0562] "Please prepare the materials."

[0563] In this way, the present invention is a system that realizes natural and effective communication for users.

[0564] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0565] The flow of this system's program processing

[0566] Step 1:

[0567] A user inputs text through a terminal. The input text (e.g., "Please check the meeting schedule.") is received by the terminal and temporarily stored in memory. The terminal then converts the received text into network packets and prepares them to be sent to the server. The input is text data, and the output is network packets for transmission.

[0568] Step 2:

[0569] The server receives text data sent from the device via the network. The server converts the received data into an appropriate format for analysis and passes it to the context analysis and sentiment analysis modules. The input is network packets, and the output is a parseable data structure (e.g., JSON format).

[0570] Step 3:

[0571] The server's context analysis module uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine their context. The server then passes the extracted keywords, phrases, and context categories to the next processing step. The input is the text data, and the output is the context analysis results.

[0572] Step 4:

[0573] The server's sentiment analysis module uses a natural language processing algorithm (e.g., VADER or TextBlob) to identify the user's sentiment from the text. The server passes the result of the sentiment analysis (positive, negative, or neutral) to the next processing step. The input is the text data, and the output is the sentiment analysis result.

[0574] Step 5:

[0575] Based on the analyzed context and sentiment information, the server invokes the translation engine module and generates the optimal translation using a generative AI model (e.g., GPT-3). The generated translation (e.g., "Could you please confirm the meeting schedule? Thank you for your understanding.") is passed to the next step. The input is the results of context analysis and sentiment analysis, and the output is the optimized translation.

[0576] Step 6:

[0577] The server uses a response generation module to generate an appropriate response based on the generated translation. Based on the results of sentiment analysis, the response is fine-tuned based on sentiment (e.g., "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."). The input is the translation, and the output is the generated response.

[0578] Step 7:

[0579] The generated response is sent from the server to the terminal. The server converts the response into a network packet and sends it to the terminal. The input is the generated response, and the output is the network packet for transmission.

[0580] Step 8:

[0581] The terminal receives the response sent from the server and displays it on the screen. The input is the network packet of the response, and the output is the displayed text that the user can see. This allows the user to see the adjusted response in real time.

[0582] Step 9:

[0583] All interactions are recorded and stored by the server's chat history management module. The server stores details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database (e.g., MySQL or MongoDB) for future reference and analysis. The input is chat details data, and the output is stored database entries.

[0584] Through the above process, this system realizes natural communication that takes into account the user's emotions and context.

[0585] (Application example 2)

[0586] 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."

[0587] To ensure smooth communication with customers in physical stores, it is necessary to respond quickly and appropriately to their questions and requests. However, it is not easy to accurately understand the customer's emotions and context and respond based on them. In particular, it is difficult to respond appropriately according to the customer's emotional state, which can lead to a decrease in customer satisfaction and an unsatisfactory experience.

[0588] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0589] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the emotion of the user, means for generating an optimal translation based on the analyzed context, means for adjusting a response based on the recognized emotion, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past interactions, and means for supporting customer service in a physical store. This allows for the rapid generation of appropriate responses that take into account the emotion and context of the customer, thereby improving the efficiency of customer service in a physical store and customer satisfaction.

[0590] The "means for receiving input text" is a function for receiving text input from a user and sending the data to a server for analysis.

[0591] The "means for analyzing the context of received text" is a function for analyzing received text data and understanding the context according to the meaning and use of the text.

[0592] "Emotion analysis means for recognizing user emotions" is a function for analyzing and identifying user emotions (positive, negative, neutral, etc.) based on keywords and phrases in the text.

[0593] The "means for generating optimal translations based on analyzed context" is a function for generating appropriate and natural translations according to the context and sentiment.

[0594] The "means for adjusting a response based on a recognized emotion" is a function for taking into account the emotional state of the user and adjusting the response so that an optimized response is generated.

[0595] The "means for generating a response based on the generated translation" is a function for generating an appropriate response based on the translated text.

[0596] The "means for displaying the generated response" is a function for displaying the generated response in a form that is easy for the user to see.

[0597] "Means for storing past interactions" is a function that records all interactions as data and stores them for future reference.

[0598] "Means to support customer service in physical stores" refers to support functions that facilitate smooth communication with customers in physical stores.

[0599] This invention is a system for facilitating customer service in brick-and-mortar stores. The system receives user input text, analyzes context and sentiment to generate an optimal translation, and provides a series of functions to display a response based on that. Specific embodiments are described below.

[0600] System configuration

[0601] The system mainly consists of the following elements:

[0602] A means of receiving input text

[0603] A means of analyzing the context of received text

[0604] Emotion analysis means to recognize user emotions

[0605] A means of generating optimal translations based on analyzed context

[0606] A means of tailoring responses based on perceived emotions

[0607] A means of generating a response based on the generated translation

[0608] A means of displaying the generated response

[0609] A way to preserve past interactions

[0610] A means to support customer service in physical stores

[0611] System operation details

[0612] Receiving text input

[0613] A user inputs text using a terminal (e.g., a smartphone) in a physical store. For example, if the user inputs "Please check product inventory," the terminal receives this text and prepares it for transmission to the server.

[0614] Context and Sentiment Analysis

[0615] When the server receives text from a device, it first passes it to a context analysis module, which uses natural language processing algorithms (for example, Python's Hugging Face library) to extract keywords and phrases from the text and analyze its context. At the same time, a sentiment analysis module identifies the user's sentiment from the text.

[0616] For example, the text "Please check product availability" is a request to check product availability, and the sentiment is judged to be slightly negative.

[0617] Translation generation and adjustment

[0618] Based on the analyzed context and sentiment information, the server calls the translation engine module, which generates a translation that best suits the context and sentiment, such as "Could you please confirm the stock availability? Thank you."

[0619] The system then tailors the response based on the recognized sentiment. For example, for text with a negative sentiment like "Please check the product availability," it generates a sentiment-sensitive response. In this case, it generates a response like "The stock availability has been checked. Thank you for your patience."

[0620] View and save responses

[0621] The generated response is sent from the server to the terminal, which then displays it to the user, allowing the user to see the response tailored to the situation in real time.

[0622] All interactions are recorded and saved by the server's chat history management module, which records detailed information such as the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses in a database for future reference and analysis.

[0623] Examples of concrete examples and prompts

[0624] For example, consider the following exchange:

[0625] User Input:

[0626] "Please check product availability."

[0627] System response:

[0628] "The stock availability has been checked. Thank you for your patience."

[0629] Example prompt sentence:

[0630] "A user types, 'Check product availability.' Use contextual analysis and sentiment identification to generate an appropriate response and translation."

[0631] Contextual analysis results: Check product availability

[0632] Emotion identification result: Negative

[0633] Translation to generate: The stock availability has been checked.

[0634] Generate response: We have confirmed the product is in stock. Thank you.

[0635] This system will enable the rapid generation of appropriate responses that take into account customer emotions and context, improving the efficiency of customer service in physical stores and customer satisfaction.

[0636] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0637] Step 1:

[0638] The user inputs text. The user uses a device such as a smartphone in a physical store to input a question or request about a product. An example of input text is "Please check product availability." The device receives this text.

[0639] Input: User text input

[0640] Output: Received text data

[0641] Step 2:

[0642] The terminal sends the received text to the server. At this time, the text data is safely delivered to the server via the network. The server receives this text data.

[0643] Input: Text data sent from the terminal

[0644] Output: Text data received by the server

[0645] Step 3:

[0646] The server passes the received text data to a context analysis module, which uses a natural language processing algorithm (e.g., Python's Hugging Face library) to extract keywords and phrases from the text. For example, it extracts "product inventory" and determines that the context is a business context.

[0647] Input: Text data received by the server

[0648] Output: Context information (e.g. product availability, business context)

[0649] Step 4:

[0650] The server simultaneously passes the text to a sentiment analysis module, which identifies the user's sentiment from the text. For example, a request like "Please check product availability" is identified as a slightly negative sentiment. This analysis is performed using a generative AI model.

[0651] Input: Text data received by the server

[0652] Output: Emotional information (e.g., negative)

[0653] Step 5:

[0654] The server passes the results of the contextual and sentiment analysis to the translation engine module, which generates the optimal translation based on the analysis results. For example, the generated text might be "Can you please check the stock availability? Thank you." A generative AI model is also used at this stage.

[0655] Input: Contextual and emotional information

[0656] Output:Translated text

[0657] Step 6:

[0658] The translated text is passed to a response generation module, which adjusts the response based on the sentiment analysis results. For example, if the sentiment is negative, a more considerate response such as "The stock availability has been checked. Thank you for your patience" will be generated.

[0659] Input: Translation text and sentiment information

[0660] Output: Adjusted response text

[0661] Step 7:

[0662] The server sends the generated response text to the terminal, which displays it to the user, who can see the appropriately tailored response in real time.

[0663] Input: Reply text sent from the server

[0664] Output: The response text displayed to the user

[0665] Step 8:

[0666] All interactions are recorded and saved by the chat history management module on the server, including the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses. The saved data is available for future reference and analysis.

[0667] Input: Various generated data (date, time, content, analysis results, final response, etc.)

[0668] Output: Saved chat history

[0669] 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.

[0670] 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.

[0671] 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.

[0672] [Third embodiment]

[0673] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0674] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0675] 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).

[0676] 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.

[0677] 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.

[0678] 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).

[0679] 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. 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.

[0680] 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.

[0681] 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.

[0682] 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.

[0683] 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.

[0684] 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."

[0685] The natural language processing system of the present invention is composed of multiple modules, specifically including a means for receiving input text, a means for analyzing the context of the received text, a means for generating an optimal translation based on the analyzed context, a means for generating a response based on the generated translation, a means for displaying the generated response, and a means for saving past exchanges.

[0686] Receiving text input

[0687] A user inputs text through the terminal. For example, if the user inputs "Please check the schedule for business meetings," the terminal receives this text and prepares to send it to the server.

[0688] Contextual analysis

[0689] After the server receives the text sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to analyze the text and classify the context as business, casual, or formal. To do this, it performs keyword extraction and phrase analysis. For example, the text "Please confirm the business meeting schedule" is determined to be a business context.

[0690] Translation Generation

[0691] The server then invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. Using a generative AI model, a phrase optimized for the context is output. For example, "Please confirm the business meeting schedule" is translated to "Could you please confirm the meeting schedule?"

[0692] Response Generation

[0693] Based on the generated translation, the server uses the response generation module to create an appropriate response, which then suggests a specific action to the user. For example, after receiving the translation result, the response generated is "The meeting schedule is confirmed for tomorrow at 3 PM."

[0694] Viewing the response

[0695] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the translation and the response in real time.

[0696] Save chat history

[0697] All conversations are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[0698] As a concrete example of implementation, if someone inputs "Please send the meeting materials" during the preparation stage of a business meeting, the system will perform appropriate context analysis and translation, and ultimately display a reply with an English translation of "Please send the meeting documents." This exchange is also saved on the server for future reference.

[0699] The processing flow will be explained below.

[0700] Step 1:

[0701] The user inputs text through the terminal. For example, the user inputs "Please check the schedule for business meetings."

[0702] Step 2:

[0703] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[0704] Step 3:

[0705] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[0706] Step 4:

[0707] The context analysis module installed on the server analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the schedule for the business meeting" is classified as a business context.

[0708] Step 5:

[0709] The server calls the translation engine module based on the results of the context analysis. The translation engine module generates a translation that is optimal for the business context. It uses a generative AI model to select the most appropriate expression. For example, "Please confirm the business meeting schedule." is output as "Could you please confirm the meeting schedule?"

[0710] Step 6:

[0711] The server receives the translated text generated by the translation engine module and passes it to the response generation module, which then creates a specific response for the user based on the translated text. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM."

[0712] Step 7:

[0713] The server sends the generated response to the terminal.

[0714] Step 8:

[0715] The terminal receives the response sent from the server and displays it to the user, allowing the user to check the response in real time.

[0716] Step 9:

[0717] The server records and stores all conversations using a chat history management module. The stored data includes details such as date and time, input text, analysis results, translation results, and final responses. Past conversations are made available for future communication.

[0718] Example 1

[0719] 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."

[0720] Conventional natural language processing systems have difficulty properly analyzing the context of text and generating optimal translations and responses. They also lack the ability to efficiently store and later search the history of interactions with users. As a result, there are problems with reduced accuracy and efficiency of communication.

[0721] 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.

[0722] In this invention, the server includes means for receiving text entered by a user through a terminal, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response on the terminal, and means for saving past exchanges. This makes it possible to appropriately analyze the context of the text and generate optimal translations and responses. Furthermore, the history of exchanges with the user can be efficiently saved and later searched for and used.

[0723] "User" refers to an individual or organization that uses the system by inputting text through a terminal.

[0724] "Terminal" refers to a computer device or personal digital assistant that a user uses to enter and send text.

[0725] "Text" refers to a character string or sentence that a user inputs through a terminal.

[0726] "Server" refers to a computer system that processes text received from a terminal and performs analysis, translation, and response generation.

[0727] The "context analysis means" refers to a mechanism that has an algorithm that analyzes received text and classifies the context of the text into business, casual, formal, etc.

[0728] "Translation generation means" refers to a mechanism that has an algorithm that generates an optimal translation based on analyzed context information.

[0729] "Generative AI model" refers to an algorithmic model that uses artificial intelligence techniques to generate translations and responses that are optimized for contextual information.

[0730] A "prompt" refers to a command or question input to a generative AI model.

[0731] The "response generation means" refers to a mechanism having an algorithm that automatically creates a response to the user based on the generated translation.

[0732] The "chat history management means" refers to a database system that stores the history of interactions with users and manages it so that it can be searched and referenced later.

[0733] The natural language processing system of the present invention is composed of multiple modules. In this system, the user, terminal, and server each play their respective roles and operate smoothly to analyze input text, translate, generate responses, and manage history.

[0734] First, a user inputs text through a device such as a PC or smartphone. For example, if the user inputs "Please confirm the business meeting schedule," the device packages the input text in JSON format and sends it to the server as an HTTP POST request. The device used can be a general computer or a mobile information terminal, and the software can be React or Vue.js.

[0735] Next, the server receives this HTTP request sent from the device. The server passes the received text data to a context analysis module. This module uses a natural language processing library (e.g., spaCy or NLTK) to tokenize the text and extract keywords, then classifies the context as business, casual, or formal. For example, "Please confirm the schedule for the business meeting" is considered a business context.

[0736] The server then invokes a translation engine module based on the context analysis. This module generates the optimal translation based on the analyzed context information. Specifically, a generative AI model (e.g., OpenAI GPT-3) is used to output a phrase optimized for the context. For example, "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?" An example of this prompt is "Translate this business-related text to English: Please confirm the business meeting schedule."

[0737] The server then uses a response generation module to generate an appropriate response based on the translated text. The response generation module generates a response that indicates a specific action based on the translated text. For example, in response to the translated text "Could you please confirm the meeting schedule?", the server generates the response "The meeting schedule is confirmed for tomorrow at 3 PM."

[0738] The generated response is sent from the server to the device, which then displays it to the user using a front-end framework such as React or Vue.js, providing the response to the user in real time.

[0739] Finally, the server stores all conversations through a chat history management module, which records details of each chat, such as the date and time, content, translation results, and responses, in a database (e.g., MySQL or PostgreSQL), efficiently storing past conversations so that future communications can be easily accessed and utilized.

[0740] As a concrete example of implementation, when preparing for a business meeting, a user may input "Please send the meeting materials." The system will perform appropriate context analysis and translation, and ultimately display a response to the user with an English translation: "Please send the meeting documents." All of this communication is also stored on the server for future reference.

[0741] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0742] Step 1:

[0743] The user inputs text through the terminal. Specifically, the user enters "Please confirm the business meeting schedule" into the text input field on the terminal and clicks the send button. The terminal packages this input text in JSON format and sends it to the server as an HTTP POST request. The input is the text from the user, and the output is an HTTP request to the server.

[0744] Step 2:

[0745] The server receives an HTTP request sent from a terminal. The server parses the received JSON formatted text data and passes it to the context analysis module. The input is the content of the HTTP request, and the output is the input data for the context analysis module.

[0746] Step 3:

[0747] The context analysis module analyzes the received text data using a natural language processing library (e.g., spaCy or NLTK). Specifically, it tokenizes the text, extracts keywords, and classifies the context as business, casual, or formal. The input is the text to be analyzed, and the output is context information. For example, "Please confirm the schedule for the business meeting" is classified as a business context.

[0748] Step 4:

[0749] The server calls the translation engine module based on the context information obtained from the context analysis module. The translation engine module uses a generative AI model (e.g., OpenAI GPT-3) to generate a translation optimized for the context information. The input is the context information and the original text, and the output is the translation. "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?"

[0750] Step 5:

[0751] The server uses a response generation module to create an appropriate response based on the generated translation. The response generation module generates a response indicating a specific action based on the translated text. The input is the translation and the output is the response. For example, in response to the question, "Could you please confirm the meeting schedule?", the response generated is, "The meeting schedule is confirmed for tomorrow at 3 PM."

[0752] Step 6:

[0753] The server sends the generated response to the terminal. The terminal receives it and displays it on the user interface. The input is the response from the server, and the output is the response displayed on the user interface. A front-end framework such as React or Vue.js is used for display.

[0754] Step 7:

[0755] All conversations are saved in the chat history management module on the server. The chat history management module records detailed information such as the date and time of each chat, content, translation results, and responses in a database (e.g., MySQL or PostgreSQL). The input is chat details, and the output is historical data saved in the database. This allows for later search and reference.

[0756] (Application example 1)

[0757] 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."

[0758] The purpose of this invention is to provide a customer support system that can quickly and accurately automatically generate translations and responses to multilingual inquiries from users. In particular, it solves the problem of efficiently responding to inquiries from a variety of users by performing context analysis, generating appropriate translations, and then generating responses based on those translations.

[0759] 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.

[0760] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past exchanges, means for receiving, analyzing, and translating user input in multiple languages ​​to generate a response, and means for displaying the generated translation and response in a chat format and maintaining a history, thereby enabling a prompt and accurate response to user inquiries.

[0761] "Entered text" means textual information entered by a user using a device.

[0762] "Context" is background information that helps understand the meaning and intent of the input text.

[0763] A "translation" is a text that is converted from an input text into a different language.

[0764] A "response" is a response provided by the system to a user's inquiry or request.

[0765] "Display means" means a device or software that displays the generated translation or response in a user-friendly format.

[0766] "Storage means" is a function for recording past interactions and data so that they can be retrieved later.

[0767] "Contextual analysis means" is an analytical function that understands the background and meaning of input text and performs appropriate classification and interpretation.

[0768] "Multilingual input" refers to a user entering text in multiple different languages.

[0769] A "natural language processing algorithm" is a set of procedures and calculations that allow a computer to understand and process human language.

[0770] A "generative AI model" is a trained model that uses artificial intelligence to analyze text and generate appropriate translations and responses.

[0771] "Chat format" is a format that displays alternating dialogue between the user and the system.

[0772] "History retention" is a function that records past inquiries and responses so that they can be accessed later.

[0773] The present invention has as its main object to provide a multilingual customer support system. Detailed embodiments of the invention will be described below.

[0774] This system implements a series of processes that receive input from users, analyze the context, generate appropriate translations, and return responses. The system consists of a server and a user terminal.

[0775] Program Overview

[0776] First, the user inputs text using the terminal. For example, if the user inputs "Please check the delivery status of the product," the terminal prepares to send this text to the server.

[0777] When the server receives the text from the device, it passes it to a context analysis module. This module uses natural language processing algorithms to analyze the context of the text and generate a translation appropriate for that context. For example, the text "Please check the delivery status of your item" is determined to be in a business context.

[0778] The server then invokes a translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used to output a translation optimized for the context.

[0779] Based on the generated translation, the server uses a response generation module to create an appropriate response, which then suggests a specific action to the user. For example, in response to the query "Please check the delivery status of your product," the server generates the response "The delivery status of your product has been confirmed and it is on its way."

[0780] The generated response is sent from the server to the terminal, which displays it to the user, who can check the translation and the response in real time.

[0781] Furthermore, all conversations are saved by a chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[0782] Hardware and Software Configuration

[0783] Hardware:

[0784] User device (smartphone or PC)

[0785] Server (Cloud server or on-premise server)

[0786] software:

[0787] Natural Language Processing Algorithms

[0788] Google Translation API (translation engine)

[0789] Python 3.x

[0790] The Requests library (for sending HTTP requests)

[0791] The server analyzes and translates the input text to quickly and accurately respond to user inquiries, generating and displaying a response, enabling users to receive smooth support even for inquiries in multiple languages.

[0792] Examples and prompts

[0793] For example, if a user types "Please check the delivery status of your product," the system analyzes the context, determines it is a business context, and calls the translation engine to translate it into English. The final response displayed to the user is "The delivery status of your product has been confirmed and it is on its way."

[0794] An example of an input prompt for the generative AI model is as follows:

[0795] text

[0796] If a user types "Please check the delivery status of my item," analyze the context, determine that it is a business context, call the translation engine to translate it into English, and generate an appropriate response to display to the user.

[0797] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0798] Step 1:

[0799] A user inputs text using a terminal. For example, the user inputs "Please check the delivery status of the product." This input text is received by the terminal as an instruction from the user. The received text is prepared to be sent to the server as is.

[0800] Step 2:

[0801] The terminal sends the input text to the server. The text sent from the terminal is received by the server. The received text proceeds to the next processing stage.

[0802] Step 3:

[0803] The server passes the received text to the context analysis module, which uses natural language processing algorithms to analyze the context of the text. For example, the text "Please check the delivery status of your item" is determined to be a business context. The input is text, and the output is context information for the text.

[0804] Step 4:

[0805] The server invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used here to output a context-optimized translation (e.g., "Could you please confirm the delivery status of the product?"). The input is context information and text, and the output is the translated text.

[0806] Step 5:

[0807] The server uses a response generation module to create an appropriate response based on the generated translation. This module uses an AI-based response algorithm to generate a response that suggests a specific action for the user. For example: "The delivery status of your product has been confirmed and it is on its way." The input is the translation, and the output is a response to the user.

[0808] Step 6:

[0809] The generated response is sent from the server to the device. The device receives the response and displays it to the user. The user can see the translated response in real time. The input is the response from the server, and the output is the text displayed to the user.

[0810] Step 7:

[0811] All exchanges are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. This allows past exchanges to be searched and used for future communication. The input is the exchange data, and the output is the saved chat history.

[0812] 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.

[0813] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, enabling more effective and natural communication. Specifically, it is composed of the following multiple modules:

[0814] A means of receiving input text

[0815] A means of analyzing the context of received text

[0816] Emotion analysis means to recognize user emotions

[0817] A means of generating optimal translations based on analyzed context

[0818] A means of tailoring responses based on perceived emotions

[0819] A means of generating a response based on the generated translation

[0820] A means of displaying the generated response

[0821] A way to preserve past interactions

[0822] Receiving text input

[0823] A user inputs text through the terminal. For example, if the user inputs "Please check the meeting schedule," the terminal receives this text and prepares to send it to the server.

[0824] Context and Sentiment Analysis

[0825] After the server receives the text data sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. At the same time, the sentiment analysis module identifies the user's sentiment (positive, negative, neutral) from the text. For example, the text "Please confirm the meeting schedule" is in a business context and the sentiment is determined to be slightly negative.

[0826] Translation Generation

[0827] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for the business context while also taking negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0828] Response Generation and Adjustment

[0829] Based on the translation generated by the translation engine module, the server uses the response generation module to generate an appropriate response, taking into account the results of sentiment analysis and adjusting it based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0830] Viewing the response

[0831] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the tailored response in real time.

[0832] Save chat history

[0833] All interactions are recorded and saved by the server's chat history management module. This module records detailed information about each chat, such as the date and time, content, context analysis results, sentiment analysis results, translation results, and final responses, in a database. The saved data can be used for future communications.

[0834] As a concrete example of its implementation, when a user inputs something like "Please prepare the materials" during the preparation stage of a business meeting, the system will analyze the user's emotions and provide an appropriately tailored translation and response, resulting in more useful communication for the user. This exchange is also saved on the server for future reference.

[0835] The processing flow will be explained below.

[0836] Step 1:

[0837] A user inputs text through a terminal. For example, the user inputs "Please check the meeting schedule."

[0838] Step 2:

[0839] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[0840] Step 3:

[0841] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[0842] Step 4:

[0843] The server's context analysis module analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the meeting schedule" would be classified as a business context.

[0844] Step 5:

[0845] Based on the results of the context analysis, the server calls the sentiment analysis module. This module determines the user's sentiment from the text. It uses a natural language processing algorithm to classify it as positive, negative, or neutral. For example, the phrase "please" (please) detects an overall negative sentiment.

[0846] Step 6:

[0847] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for business contexts and takes negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0848] Step 7:

[0849] The server receives the translated text generated by the translation engine module and passes it to the response generation module. The response generation module creates a specific response for the user based on the translated text. The response generation module also takes into account the results of sentiment analysis and adjusts based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0850] Step 8:

[0851] The server sends the generated response to the terminal.

[0852] Step 9:

[0853] The terminal receives the response sent from the server and displays it to the user, allowing the user to see the adjusted response in real time.

[0854] Step 10:

[0855] The server records and stores all interactions using a chat history management module. The stored data includes details such as date and time, input text, context analysis results, sentiment analysis results, translation results, and final responses. Past interactions are made available for future communication.

[0856] Example 2

[0857] 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."

[0858] While conventional natural language processing systems can analyze the context of input text and generate translations, it has been difficult to realize communication that takes into account the user's emotions. Therefore, there is a need for a system that can generate natural and effective responses based on the user's emotions.

[0859] 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.

[0860] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the user's emotions, means for generating an optimal translation based on the analyzed context and emotions, means for generating a response based on the generated translation, means for displaying the generated response, and means for saving past exchanges, thereby enabling natural communication that takes into account the user's emotions.

[0861] The "means for receiving input text" is a means for receiving text data input by a user through a terminal and transmitting it for subsequent processing.

[0862] The "means for analyzing the context of received text" refers to a means for extracting keywords and phrases from the received text data and determining the context and category (business, casual, formal, etc.).

[0863] The "means for analyzing user emotions" refers to a means for identifying user emotions (positive, negative, neutral) from received text data using a natural language processing algorithm.

[0864] "Means for generating optimal translations based on analyzed context and sentiment" refers to means for generating appropriate translations based on the results of contextual and sentiment analysis. This is done using a generative AI model.

[0865] The "means for generating a response based on the generated translation" refers to a means for creating an appropriate response using the generated translation, taking into account the results of sentiment analysis.

[0866] The "means for displaying the generated response" is a means for transmitting the generated response text to the terminal and displaying it to the user.

[0867] "Means for storing past interactions" refers to a means for recording and storing details of all interactions (date and time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database.

[0868] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, thereby realizing more effective and natural communication. This system is composed of the following multiple modules.

[0869] System Components

[0870] 1. A means of receiving input text

[0871] When a user inputs text through the terminal, for example, "Please check the meeting schedule," the terminal receives this text and transmits it to the server through the network.

[0872] 2. Means of analyzing the context of received text

[0873] The server receives the text data sent from the device and passes it to a context analysis module, which uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine its context.

[0874] 3. Means of analyzing user emotions

[0875] The server uses a sentiment analysis module to identify the user's sentiment (positive, negative, neutral) from the text. Sentiment analysis algorithms (e.g., VADER and TextBlob) are used to precisely analyze the sentiment of the input text.

[0876] 4. A means to generate optimal translations based on analyzed context and sentiment

[0877] The server then calls the translation engine module based on the analyzed context and sentiment information. Using a generative AI model (e.g., GPT-3), it generates a translation that best fits the context and sentiment. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[0878] 5. A means of generating a response based on the generated translation

[0879] The server uses the response generation module to generate an appropriate response based on the translation. Based on the results of sentiment analysis, the response is fine-tuned based on the sentiment. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[0880] 6. A way to display the generated response

[0881] The generated response is sent from the server to the terminal, which displays the received response on its screen so that the user can check it in real time.

[0882] 7. A way to preserve past interactions

[0883] All interactions are recorded and stored by the chat history management module on the server. Details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) are stored in a database (e.g., MySQL or MongoDB) for future reference and analysis.

[0884] Specific examples

[0885] If a user types "Please prepare the materials" while preparing for a business meeting, the system will generate an appropriate response through the following steps. First, the device sends the input text to the server, which performs contextual and sentiment analysis. Contextual analysis determines the context as "business," and sentiment analysis identifies it as "somewhat urgent." The translation engine translates it as "Please prepare the materials urgently," and the response generation module generates the reply, "The materials will be prepared and sent to you shortly. We appreciate your prompt request." The device displays this response to the user, and the details of the exchange are stored on the server.

[0886] Prompt Sentence Examples

[0887] "Please check the meeting schedule, and I'd also like to discuss a new project."

[0888] "Please prepare the materials."

[0889] In this way, the present invention is a system that realizes natural and effective communication for users.

[0890] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0891] The flow of this system's program processing

[0892] Step 1:

[0893] A user inputs text through a terminal. The input text (e.g., "Please check the meeting schedule.") is received by the terminal and temporarily stored in memory. The terminal then converts the received text into network packets and prepares them to be sent to the server. The input is text data, and the output is network packets for transmission.

[0894] Step 2:

[0895] The server receives text data sent from the device via the network. The server converts the received data into an appropriate format for analysis and passes it to the context analysis and sentiment analysis modules. The input is network packets, and the output is a parseable data structure (e.g., JSON format).

[0896] Step 3:

[0897] The server's context analysis module uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine their context. The server then passes the extracted keywords, phrases, and context categories to the next processing step. The input is the text data, and the output is the context analysis results.

[0898] Step 4:

[0899] The server's sentiment analysis module uses a natural language processing algorithm (e.g., VADER or TextBlob) to identify the user's sentiment from the text. The server passes the result of the sentiment analysis (positive, negative, or neutral) to the next processing step. The input is the text data, and the output is the sentiment analysis result.

[0900] Step 5:

[0901] Based on the analyzed context and sentiment information, the server invokes the translation engine module and generates the optimal translation using a generative AI model (e.g., GPT-3). The generated translation (e.g., "Could you please confirm the meeting schedule? Thank you for your understanding.") is passed to the next step. The input is the results of context analysis and sentiment analysis, and the output is the optimized translation.

[0902] Step 6:

[0903] The server uses a response generation module to generate an appropriate response based on the generated translation. Based on the results of sentiment analysis, the response is fine-tuned based on sentiment (e.g., "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."). The input is the translation, and the output is the generated response.

[0904] Step 7:

[0905] The generated response is sent from the server to the terminal. The server converts the response into a network packet and sends it to the terminal. The input is the generated response, and the output is the network packet for transmission.

[0906] Step 8:

[0907] The terminal receives the response sent from the server and displays it on the screen. The input is the network packet of the response, and the output is the displayed text that the user can see. This allows the user to see the adjusted response in real time.

[0908] Step 9:

[0909] All interactions are recorded and stored by the server's chat history management module. The server stores details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database (e.g., MySQL or MongoDB) for future reference and analysis. The input is chat details data, and the output is stored database entries.

[0910] Through the above process, this system realizes natural communication that takes into account the user's emotions and context.

[0911] (Application example 2)

[0912] 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."

[0913] To ensure smooth communication with customers in physical stores, it is necessary to respond quickly and appropriately to their questions and requests. However, it is not easy to accurately understand the customer's emotions and context and respond based on them. In particular, it is difficult to respond appropriately according to the customer's emotional state, which can lead to a decrease in customer satisfaction and an unsatisfactory experience.

[0914] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0915] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the emotion of the user, means for generating an optimal translation based on the analyzed context, means for adjusting a response based on the recognized emotion, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past interactions, and means for supporting customer service in a physical store. This allows for the rapid generation of appropriate responses that take into account the emotion and context of the customer, thereby improving the efficiency of customer service in a physical store and customer satisfaction.

[0916] The "means for receiving input text" is a function for receiving text input from a user and sending the data to a server for analysis.

[0917] The "means for analyzing the context of received text" is a function for analyzing received text data and understanding the context according to the meaning and use of the text.

[0918] "Emotion analysis means for recognizing user emotions" is a function for analyzing and identifying user emotions (positive, negative, neutral, etc.) based on keywords and phrases in the text.

[0919] The "means for generating optimal translations based on analyzed context" is a function for generating appropriate and natural translations according to the context and sentiment.

[0920] The "means for adjusting a response based on a recognized emotion" is a function for taking into account the emotional state of the user and adjusting the response so that an optimized response is generated.

[0921] The "means for generating a response based on the generated translation" is a function for generating an appropriate response based on the translated text.

[0922] The "means for displaying the generated response" is a function for displaying the generated response in a form that is easy for the user to see.

[0923] "Means for storing past interactions" is a function that records all interactions as data and stores them for future reference.

[0924] "Means to support customer service in physical stores" refers to support functions that facilitate smooth communication with customers in physical stores.

[0925] This invention is a system for facilitating customer service in brick-and-mortar stores. The system receives user input text, analyzes context and sentiment to generate an optimal translation, and provides a series of functions to display a response based on that. Specific embodiments are described below.

[0926] System configuration

[0927] The system mainly consists of the following elements:

[0928] A means of receiving input text

[0929] A means of analyzing the context of received text

[0930] Emotion analysis means to recognize user emotions

[0931] A means of generating optimal translations based on analyzed context

[0932] A means of tailoring responses based on perceived emotions

[0933] A means of generating a response based on the generated translation

[0934] A means of displaying the generated response

[0935] A way to preserve past interactions

[0936] A means to support customer service in physical stores

[0937] System operation details

[0938] Receiving text input

[0939] A user inputs text using a terminal (e.g., a smartphone) in a physical store. For example, if the user inputs "Please check product inventory," the terminal receives this text and prepares it for transmission to the server.

[0940] Context and Sentiment Analysis

[0941] When the server receives text from a device, it first passes it to a context analysis module, which uses natural language processing algorithms (for example, Python's Hugging Face library) to extract keywords and phrases from the text and analyze its context. At the same time, a sentiment analysis module identifies the user's sentiment from the text.

[0942] For example, the text "Please check product availability" is a request to check product availability, and the sentiment is judged to be slightly negative.

[0943] Translation generation and adjustment

[0944] Based on the analyzed context and sentiment information, the server calls the translation engine module, which generates a translation that best suits the context and sentiment, such as "Could you please confirm the stock availability? Thank you."

[0945] The system then tailors the response based on the recognized sentiment. For example, for text with a negative sentiment like "Please check the product availability," it generates a sentiment-sensitive response. In this case, it generates a response like "The stock availability has been checked. Thank you for your patience."

[0946] View and save responses

[0947] The generated response is sent from the server to the terminal, which then displays it to the user, allowing the user to see the response tailored to the situation in real time.

[0948] All interactions are recorded and saved by the server's chat history management module, which records detailed information such as the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses in a database for future reference and analysis.

[0949] Examples of concrete examples and prompts

[0950] For example, consider the following exchange:

[0951] User Input:

[0952] "Please check product availability."

[0953] System response:

[0954] "The stock availability has been checked. Thank you for your patience."

[0955] Example prompt sentence:

[0956] "A user types, 'Check product availability.' Use contextual analysis and sentiment identification to generate an appropriate response and translation."

[0957] Contextual analysis results: Check product availability

[0958] Emotion identification result: Negative

[0959] Translation to generate: The stock availability has been checked.

[0960] Generate response: We have confirmed the product is in stock. Thank you.

[0961] This system will enable the rapid generation of appropriate responses that take into account customer emotions and context, improving customer service efficiency and customer satisfaction in physical stores.

[0962] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0963] Step 1:

[0964] The user inputs text. The user uses a device such as a smartphone in a physical store to input a question or request about a product. An example of input text is "Please check product availability." The device receives this text.

[0965] Input: User text input

[0966] Output: Received text data

[0967] Step 2:

[0968] The terminal sends the received text to the server. At this time, the text data is safely delivered to the server via the network. The server receives this text data.

[0969] Input: Text data sent from the terminal

[0970] Output: Text data received by the server

[0971] Step 3:

[0972] The server passes the received text data to a context analysis module, which uses a natural language processing algorithm (e.g., Python's Hugging Face library) to extract keywords and phrases from the text. For example, it extracts "product inventory" and determines that the context is a business context.

[0973] Input: Text data received by the server

[0974] Output: Context information (e.g. product availability, business context)

[0975] Step 4:

[0976] The server simultaneously passes the text to a sentiment analysis module, which identifies the user's sentiment from the text. For example, a request like "Please check product availability" is identified as a slightly negative sentiment. This analysis is performed using a generative AI model.

[0977] Input: Text data received by the server

[0978] Output: Emotional information (e.g., negative)

[0979] Step 5:

[0980] The server passes the results of the contextual and sentiment analysis to the translation engine module, which generates the optimal translation based on the analysis results. For example, the generated text might be "Can you please check the stock availability? Thank you." A generative AI model is also used at this stage.

[0981] Input: Contextual and emotional information

[0982] Output:Translated text

[0983] Step 6:

[0984] The translated text is passed to a response generation module, which adjusts the response based on the sentiment analysis results. For example, if the sentiment is negative, a more considerate response such as "The stock availability has been checked. Thank you for your patience" will be generated.

[0985] Input: Translation text and sentiment information

[0986] Output: Adjusted response text

[0987] Step 7:

[0988] The server sends the generated response text to the terminal, which displays it to the user, who can see the appropriately tailored response in real time.

[0989] Input: Reply text sent from the server

[0990] Output: The response text displayed to the user

[0991] Step 8:

[0992] All interactions are recorded and saved by the chat history management module on the server, including the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses. The saved data is available for future reference and analysis.

[0993] Input: Various generated data (date, time, content, analysis results, final response, etc.)

[0994] Output: Saved chat history

[0995] 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.

[0996] 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.

[0997] 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.

[0998] [Fourth embodiment]

[0999] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1000] 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.

[1001] 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).

[1002] 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.

[1003] 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.

[1004] 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).

[1005] 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.

[1006] 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.

[1007] 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.

[1008] 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.

[1009] 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.

[1010] 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.

[1011] 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."

[1012] The natural language processing system of the present invention is composed of multiple modules, specifically including a means for receiving input text, a means for analyzing the context of the received text, a means for generating an optimal translation based on the analyzed context, a means for generating a response based on the generated translation, a means for displaying the generated response, and a means for saving past exchanges.

[1013] Receiving text input

[1014] A user inputs text through the terminal. For example, if the user inputs "Please check the schedule for business meetings," the terminal receives this text and prepares to send it to the server.

[1015] Contextual analysis

[1016] After the server receives the text sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to analyze the text and classify the context as business, casual, or formal. To do this, it performs keyword extraction and phrase analysis. For example, the text "Please confirm the business meeting schedule" is determined to be a business context.

[1017] Translation Generation

[1018] The server then invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. Using a generative AI model, a phrase optimized for the context is output. For example, "Please confirm the business meeting schedule" is translated to "Could you please confirm the meeting schedule?"

[1019] Response Generation

[1020] Based on the generated translation, the server uses the response generation module to create an appropriate response, which then suggests a specific action to the user. For example, after receiving the translation result, the response generated is "The meeting schedule is confirmed for tomorrow at 3 PM."

[1021] Viewing the response

[1022] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the translation and the response in real time.

[1023] Save chat history

[1024] All conversations are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[1025] As a concrete example of implementation, if someone inputs "Please send the meeting materials" during the preparation stage of a business meeting, the system will perform appropriate context analysis and translation, and ultimately display a reply with an English translation of "Please send the meeting documents." This exchange is also saved on the server for future reference.

[1026] The processing flow will be explained below.

[1027] Step 1:

[1028] The user inputs text through the terminal. For example, the user inputs "Please check the schedule for business meetings."

[1029] Step 2:

[1030] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[1031] Step 3:

[1032] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[1033] Step 4:

[1034] The context analysis module installed on the server analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the schedule for the business meeting" is classified as a business context.

[1035] Step 5:

[1036] The server calls the translation engine module based on the results of the context analysis. The translation engine module generates a translation that is optimal for the business context. It uses a generative AI model to select the most appropriate expression. For example, "Please confirm the business meeting schedule." is output as "Could you please confirm the meeting schedule?"

[1037] Step 6:

[1038] The server receives the translated text generated by the translation engine module and passes it to the response generation module, which then creates a specific response for the user based on the translated text. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM."

[1039] Step 7:

[1040] The server sends the generated response to the terminal.

[1041] Step 8:

[1042] The terminal receives the response sent from the server and displays it to the user, allowing the user to check the response in real time.

[1043] Step 9:

[1044] The server records and stores all conversations using a chat history management module. The stored data includes details such as date and time, input text, analysis results, translation results, and final responses. Past conversations are made available for future communication.

[1045] Example 1

[1046] 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."

[1047] Conventional natural language processing systems have difficulty properly analyzing the context of text and generating optimal translations and responses. They also lack the ability to efficiently store and later search the history of interactions with users. As a result, there are problems with reduced accuracy and efficiency of communication.

[1048] 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.

[1049] In this invention, the server includes means for receiving text entered by a user through a terminal, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response on the terminal, and means for saving past exchanges. This makes it possible to appropriately analyze the context of the text and generate optimal translations and responses. Furthermore, the history of exchanges with the user can be efficiently saved and later searched for and used.

[1050] "User" refers to an individual or organization that uses the system by inputting text through a terminal.

[1051] "Terminal" refers to a computer device or personal digital assistant that a user uses to enter and send text.

[1052] "Text" refers to a character string or sentence that a user inputs through a terminal.

[1053] "Server" refers to a computer system that processes text received from a terminal and performs analysis, translation, and response generation.

[1054] The "context analysis means" refers to a mechanism that has an algorithm that analyzes received text and classifies the context of the text into business, casual, formal, etc.

[1055] "Translation generation means" refers to a mechanism that has an algorithm that generates an optimal translation based on analyzed context information.

[1056] "Generative AI model" refers to an algorithmic model that uses artificial intelligence techniques to generate translations and responses that are optimized for contextual information.

[1057] A "prompt" refers to a command or question input to a generative AI model.

[1058] The "response generation means" refers to a mechanism having an algorithm that automatically creates a response to the user based on the generated translation.

[1059] The "chat history management means" refers to a database system that stores the history of interactions with users and manages it so that it can be searched and referenced later.

[1060] The natural language processing system of the present invention is composed of multiple modules. In this system, the user, terminal, and server each play their respective roles and operate smoothly to analyze input text, translate, generate responses, and manage history.

[1061] First, a user inputs text through a device such as a PC or smartphone. For example, if the user inputs "Please confirm the business meeting schedule," the device packages the input text in JSON format and sends it to the server as an HTTP POST request. The device used can be a general computer or a mobile information terminal, and the software can be React or Vue.js.

[1062] Next, the server receives this HTTP request sent from the device. The server passes the received text data to a context analysis module. This module uses a natural language processing library (e.g., spaCy or NLTK) to tokenize the text and extract keywords, then classifies the context as business, casual, or formal. For example, "Please confirm the schedule for the business meeting" is considered a business context.

[1063] The server then invokes a translation engine module based on the context analysis. This module generates the optimal translation based on the analyzed context information. Specifically, a generative AI model (e.g., OpenAI GPT-3) is used to output a phrase optimized for the context. For example, "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?" An example of this prompt is "Translate this business-related text to English: Please confirm the business meeting schedule."

[1064] The server then uses a response generation module to generate an appropriate response based on the translated text. The response generation module generates a response that indicates a specific action based on the translated text. For example, in response to the translated text "Could you please confirm the meeting schedule?", the server generates the response "The meeting schedule is confirmed for tomorrow at 3 PM."

[1065] The generated response is sent from the server to the device, which then displays it to the user using a front-end framework such as React or Vue.js, providing the response to the user in real time.

[1066] Finally, the server stores all conversations through a chat history management module, which records details of each chat, such as the date and time, content, translation results, and responses, in a database (e.g., MySQL or PostgreSQL), efficiently storing past conversations so that future communications can be easily accessed and utilized.

[1067] As a concrete example of implementation, when preparing for a business meeting, a user may input "Please send the meeting materials." The system will perform appropriate context analysis and translation, and ultimately display a response to the user with an English translation: "Please send the meeting documents." All of this communication is also stored on the server for future reference.

[1068] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1069] Step 1:

[1070] The user inputs text through the terminal. Specifically, the user enters "Please confirm the business meeting schedule" into the text input field on the terminal and clicks the send button. The terminal packages this input text in JSON format and sends it to the server as an HTTP POST request. The input is the text from the user, and the output is an HTTP request to the server.

[1071] Step 2:

[1072] The server receives an HTTP request sent from a terminal. The server parses the received JSON formatted text data and passes it to the context analysis module. The input is the content of the HTTP request, and the output is the input data for the context analysis module.

[1073] Step 3:

[1074] The context analysis module analyzes the received text data using a natural language processing library (e.g., spaCy or NLTK). Specifically, it tokenizes the text, extracts keywords, and classifies the context as business, casual, or formal. The input is the text to be analyzed, and the output is context information. For example, "Please confirm the schedule for the business meeting" is classified as a business context.

[1075] Step 4:

[1076] The server calls the translation engine module based on the context information obtained from the context analysis module. The translation engine module uses a generative AI model (e.g., OpenAI GPT-3) to generate a translation optimized for the context information. The input is the context information and the original text, and the output is the translation. "Please confirm the business meeting schedule." is translated to "Could you please confirm the meeting schedule?"

[1077] Step 5:

[1078] The server uses a response generation module to create an appropriate response based on the generated translation. The response generation module generates a response indicating a specific action based on the translated text. The input is the translation and the output is the response. For example, in response to the question, "Could you please confirm the meeting schedule?", the response generated is, "The meeting schedule is confirmed for tomorrow at 3 PM."

[1079] Step 6:

[1080] The server sends the generated response to the terminal. The terminal receives it and displays it on the user interface. The input is the response from the server, and the output is the response displayed on the user interface. A front-end framework such as React or Vue.js is used for display.

[1081] Step 7:

[1082] All conversations are saved in the chat history management module on the server. The chat history management module records detailed information such as the date and time of each chat, content, translation results, and responses in a database (e.g., MySQL or PostgreSQL). The input is chat details, and the output is historical data saved in the database. This allows for later search and reference.

[1083] (Application example 1)

[1084] 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."

[1085] The purpose of this invention is to provide a customer support system that can quickly and accurately automatically generate translations and responses to multilingual inquiries from users. In particular, it solves the problem of efficiently responding to inquiries from a variety of users by performing context analysis, generating appropriate translations, and then generating responses based on those translations.

[1086] 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.

[1087] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for generating an optimal translation based on the analyzed context, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past exchanges, means for receiving, analyzing, and translating user input in multiple languages ​​to generate a response, and means for displaying the generated translation and response in a chat format and maintaining a history, thereby enabling a prompt and accurate response to user inquiries.

[1088] "Entered text" means textual information entered by a user using a device.

[1089] "Context" is background information that helps understand the meaning and intent of the input text.

[1090] A "translation" is a text that is converted from an input text into a different language.

[1091] A "response" is a response provided by the system to a user's inquiry or request.

[1092] "Display means" means a device or software that displays the generated translation or response in a user-friendly format.

[1093] "Storage means" is a function for recording past interactions and data so that they can be retrieved later.

[1094] "Contextual analysis means" is an analytical function that understands the background and meaning of input text and performs appropriate classification and interpretation.

[1095] "Multilingual input" refers to a user entering text in multiple different languages.

[1096] A "natural language processing algorithm" is a set of procedures and calculations that allow a computer to understand and process human language.

[1097] A "generative AI model" is a trained model that uses artificial intelligence to analyze text and generate appropriate translations and responses.

[1098] "Chat format" is a format that displays alternating dialogue between the user and the system.

[1099] "History retention" is a function that records past inquiries and responses so that they can be accessed later.

[1100] The present invention has as its main object to provide a multilingual customer support system. Detailed embodiments of the invention will be described below.

[1101] This system implements a series of processes that receive input from users, analyze the context, generate appropriate translations, and return responses. The system consists of a server and a user terminal.

[1102] Program Overview

[1103] First, the user inputs text using the terminal. For example, if the user inputs "Please check the delivery status of the product," the terminal prepares to send this text to the server.

[1104] When the server receives the text from the device, it passes it to a context analysis module. This module uses natural language processing algorithms to analyze the context of the text and generate a translation appropriate for that context. For example, the text "Please check the delivery status of your item" is determined to be in a business context.

[1105] The server then invokes a translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used to output a translation optimized for the context.

[1106] Based on the generated translation, the server uses a response generation module to create an appropriate response, which then suggests a specific action to the user. For example, in response to the query "Please check the delivery status of your product," the server generates the response "The delivery status of your product has been confirmed and it is on its way."

[1107] The generated response is sent from the server to the terminal, which displays it to the user, who can check the translation and the response in real time.

[1108] Furthermore, all conversations are saved by a chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. Past conversations can be searched and used for future communication.

[1109] Hardware and Software Configuration

[1110] Hardware:

[1111] User device (smartphone or PC)

[1112] Server (Cloud server or on-premise server)

[1113] software:

[1114] Natural Language Processing Algorithms

[1115] Google Translation API (translation engine)

[1116] Python 3.x

[1117] The Requests library (for sending HTTP requests)

[1118] The server analyzes and translates the input text to quickly and accurately respond to user inquiries, generating and displaying a response, enabling users to receive smooth support even for inquiries in multiple languages.

[1119] Examples and prompts

[1120] For example, if a user types "Please check the delivery status of your product," the system analyzes the context, determines it is a business context, and calls the translation engine to translate it into English. The final response displayed to the user is "The delivery status of your product has been confirmed and it is on its way."

[1121] An example of an input prompt for the generative AI model is as follows:

[1122] text

[1123] If a user types "Please check the delivery status of my item," analyze the context, determine that it is a business context, call the translation engine to translate it into English, and generate an appropriate response to display to the user.

[1124] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1125] Step 1:

[1126] A user inputs text using a terminal. For example, the user inputs "Please check the delivery status of the product." This input text is received by the terminal as an instruction from the user. The received text is prepared to be sent to the server as is.

[1127] Step 2:

[1128] The terminal sends the input text to the server. The text sent from the terminal is received by the server. The received text proceeds to the next processing stage.

[1129] Step 3:

[1130] The server passes the received text to the context analysis module, which uses natural language processing algorithms to analyze the context of the text. For example, the text "Please check the delivery status of your item" is determined to be a business context. The input is text, and the output is context information for the text.

[1131] Step 4:

[1132] The server invokes the translation engine module based on the analyzed context information. This module generates a translation appropriate for the specified context. A generative AI model is used here to output a context-optimized translation (e.g., "Could you please confirm the delivery status of the product?"). The input is context information and text, and the output is the translated text.

[1133] Step 5:

[1134] The server uses a response generation module to create an appropriate response based on the generated translation. This module uses an AI-based response algorithm to generate a response that suggests a specific action for the user. For example: "The delivery status of your product has been confirmed and it is on its way." The input is the translation, and the output is a response to the user.

[1135] Step 6:

[1136] The generated response is sent from the server to the device. The device receives the response and displays it to the user. The user can see the translated response in real time. The input is the response from the server, and the output is the text displayed to the user.

[1137] Step 7:

[1138] All exchanges are saved by the chat history management module on the server. This module records detailed information about each chat, such as the date and time, content, translation results, and responses, in a database. This allows past exchanges to be searched and used for future communication. The input is the exchange data, and the output is the saved chat history.

[1139] 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.

[1140] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, enabling more effective and natural communication. Specifically, it is composed of the following multiple modules:

[1141] A means of receiving input text

[1142] A means of analyzing the context of received text

[1143] Emotion analysis means to recognize user emotions

[1144] A means of generating optimal translations based on analyzed context

[1145] A means of tailoring responses based on perceived emotions

[1146] A means of generating a response based on the generated translation

[1147] A means of displaying the generated response

[1148] A way to preserve past interactions

[1149] Receiving text input

[1150] A user inputs text through the terminal. For example, if the user inputs "Please check the meeting schedule," the terminal receives this text and prepares to send it to the server.

[1151] Context and Sentiment Analysis

[1152] After the server receives the text data sent from the device, it passes it to the context analysis module. This module uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. At the same time, the sentiment analysis module identifies the user's sentiment (positive, negative, neutral) from the text. For example, the text "Please confirm the meeting schedule" is in a business context and the sentiment is determined to be slightly negative.

[1153] Translation Generation

[1154] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for the business context while also taking negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[1155] Response Generation and Adjustment

[1156] Based on the translation generated by the translation engine module, the server uses the response generation module to generate an appropriate response, taking into account the results of sentiment analysis and adjusting it based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[1157] Viewing the response

[1158] The generated response is sent from the server to the terminal, which displays it to the user, allowing the user to see the tailored response in real time.

[1159] Save chat history

[1160] All interactions are recorded and saved by the server's chat history management module. This module records detailed information about each chat, such as the date and time, content, context analysis results, sentiment analysis results, translation results, and final responses, in a database. The saved data can be used for future communications.

[1161] As a concrete example of its implementation, when a user inputs something like "Please prepare the materials" during the preparation stage of a business meeting, the system will analyze the user's emotions and provide an appropriately tailored translation and response, resulting in more useful communication for the user. This exchange is also saved on the server for future reference.

[1162] The processing flow will be explained below.

[1163] Step 1:

[1164] A user inputs text through a terminal. For example, the user inputs "Please check the meeting schedule."

[1165] Step 2:

[1166] The terminal receives the input text, structures the text data, and prepares it for transmission to the server.

[1167] Step 3:

[1168] The server receives the text data sent from the terminal, temporarily stores the received text data, and passes it to the context analysis module.

[1169] Step 4:

[1170] The server's context analysis module analyzes the received text. It uses natural language processing algorithms to extract keywords and phrases from the text and determine its context. For example, the text "Please confirm the meeting schedule" would be classified as a business context.

[1171] Step 5:

[1172] Based on the results of the context analysis, the server calls the sentiment analysis module. This module determines the user's sentiment from the text. It uses a natural language processing algorithm to classify it as positive, negative, or neutral. For example, the phrase "please" (please) detects an overall negative sentiment.

[1173] Step 6:

[1174] The server then calls the translation engine module based on the analyzed context and sentiment information. This module generates a translation that is optimal for business contexts and takes negative sentiment into account. Using a generative AI model, the most appropriate phrase for the context and sentiment is output. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[1175] Step 7:

[1176] The server receives the translated text generated by the translation engine module and passes it to the response generation module. The response generation module creates a specific response for the user based on the translated text. The response generation module also takes into account the results of sentiment analysis and adjusts based on emotion. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[1177] Step 8:

[1178] The server sends the generated response to the terminal.

[1179] Step 9:

[1180] The terminal receives the response sent from the server and displays it to the user, allowing the user to see the adjusted response in real time.

[1181] Step 10:

[1182] The server records and stores all interactions using a chat history management module. The stored data includes details such as date and time, input text, context analysis results, sentiment analysis results, translation results, and final responses. Past interactions are made available for future communication.

[1183] Example 2

[1184] 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."

[1185] While conventional natural language processing systems can analyze the context of input text and generate translations, it has been difficult to realize communication that takes into account the user's emotions. Therefore, there is a need for a system that can generate natural and effective responses based on the user's emotions.

[1186] 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.

[1187] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the user's emotions, means for generating an optimal translation based on the analyzed context and emotions, means for generating a response based on the generated translation, means for displaying the generated response, and means for saving past exchanges, thereby enabling natural communication that takes into account the user's emotions.

[1188] The "means for receiving input text" is a means for receiving text data input by a user through a terminal and transmitting it for subsequent processing.

[1189] The "means for analyzing the context of received text" refers to a means for extracting keywords and phrases from the received text data and determining the context and category (business, casual, formal, etc.).

[1190] The "means for analyzing user emotions" refers to a means for identifying user emotions (positive, negative, neutral) from received text data using a natural language processing algorithm.

[1191] "Means for generating optimal translations based on analyzed context and sentiment" refers to means for generating appropriate translations based on the results of contextual and sentiment analysis. This is done using a generative AI model.

[1192] The "means for generating a response based on the generated translation" refers to a means for creating an appropriate response using the generated translation, taking into account the results of sentiment analysis.

[1193] The "means for displaying the generated response" is a means for transmitting the generated response text to the terminal and displaying it to the user.

[1194] "Means for storing past interactions" refers to a means for recording and storing details of all interactions (date and time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database.

[1195] The natural language processing system of the present invention incorporates an emotion engine that recognizes the user's emotions, thereby realizing more effective and natural communication. This system is composed of the following multiple modules.

[1196] System Components

[1197] 1. A means of receiving input text

[1198] When a user inputs text through the terminal, for example, "Please check the meeting schedule," the terminal receives this text and transmits it to the server through the network.

[1199] 2. Means of analyzing the context of received text

[1200] The server receives the text data sent from the device and passes it to a context analysis module, which uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine its context.

[1201] 3. Means of analyzing user emotions

[1202] The server uses a sentiment analysis module to identify the user's sentiment (positive, negative, neutral) from the text. Sentiment analysis algorithms (e.g., VADER and TextBlob) are used to precisely analyze the sentiment of the input text.

[1203] 4. A means to generate optimal translations based on analyzed context and sentiment

[1204] The server then calls the translation engine module based on the analyzed context and sentiment information. Using a generative AI model (e.g., GPT-3), it generates a translation that best fits the context and sentiment. For example, "Please confirm the meeting schedule" is output as "Could you please confirm the meeting schedule? Thank you for your understanding."

[1205] 5. A means of generating a response based on the generated translation

[1206] The server uses the response generation module to generate an appropriate response based on the generated translation. Based on the results of sentiment analysis, the response is fine-tuned based on sentiment. For example, the response generated might be, "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."

[1207] 6. A way to display the generated response

[1208] The generated response is sent from the server to the terminal, which displays the received response on its screen so that the user can check it in real time.

[1209] 7. A way to preserve past interactions

[1210] All interactions are recorded and stored by the chat history management module on the server. Details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) are stored in a database (e.g., MySQL or MongoDB) for future reference and analysis.

[1211] Specific examples

[1212] If a user types "Please prepare the materials" while preparing for a business meeting, the system will generate an appropriate response through the following steps. First, the device sends the input text to the server, which performs contextual and sentiment analysis. Contextual analysis determines that the context is "business," and sentiment analysis identifies it as "somewhat urgent." The translation engine translates it as "Please prepare the materials urgently," and the response generation module generates the reply, "The materials will be prepared and sent to you shortly. We appreciate your prompt request." The device displays this response to the user, and the details of the exchange are stored on the server.

[1213] Prompt Sentence Examples

[1214] "Please check the meeting schedule, and I'd also like to discuss a new project."

[1215] "Please prepare the materials."

[1216] In this way, the present invention is a system that realizes natural and effective communication for users.

[1217] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1218] The flow of this system's program processing

[1219] Step 1:

[1220] A user inputs text through a terminal. The input text (e.g., "Please check the meeting schedule.") is received by the terminal and temporarily stored in memory. The terminal then converts the received text into network packets and prepares them to be sent to the server. The input is text data, and the output is network packets for transmission.

[1221] Step 2:

[1222] The server receives text data sent from the device via the network. The server converts the received data into an appropriate format for analysis and passes it to the context analysis and sentiment analysis modules. The input is network packets, and the output is a parseable data structure (e.g., JSON format).

[1223] Step 3:

[1224] The server's context analysis module uses natural language processing algorithms (e.g., SpaCy or BERT) to extract keywords and phrases from the text and determine their context. The server then passes the extracted keywords, phrases, and context categories to the next processing step. The input is the text data, and the output is the context analysis results.

[1225] Step 4:

[1226] The server's sentiment analysis module uses a natural language processing algorithm (e.g., VADER or TextBlob) to identify the user's sentiment from the text. The server passes the result of the sentiment analysis (positive, negative, or neutral) to the next processing step. The input is the text data, and the output is the sentiment analysis result.

[1227] Step 5:

[1228] Based on the analyzed context and sentiment information, the server invokes the translation engine module and generates the optimal translation using a generative AI model (e.g., GPT-3). The generated translation (e.g., "Could you please confirm the meeting schedule? Thank you for your understanding.") is passed to the next step. The input is the results of context analysis and sentiment analysis, and the output is the optimized translation.

[1229] Step 6:

[1230] The server uses a response generation module to generate an appropriate response based on the generated translation. Based on the results of sentiment analysis, the response is fine-tuned based on sentiment (e.g., "The meeting schedule is confirmed for tomorrow at 3 PM. We appreciate your patience."). The input is the translation, and the output is the generated response.

[1231] Step 7:

[1232] The generated response is sent from the server to the terminal. The server converts the response into a network packet and sends it to the terminal. The input is the generated response, and the output is the network packet for transmission.

[1233] Step 8:

[1234] The terminal receives the response sent from the server and displays it on the screen. The input is the network packet of the response, and the output is the displayed text that the user can see. This allows the user to see the adjusted response in real time.

[1235] Step 9:

[1236] All interactions are recorded and stored by the server's chat history management module. The server stores details of each chat (date, time, content, context analysis results, sentiment analysis results, translation results, final response, etc.) in a database (e.g., MySQL or MongoDB) for future reference and analysis. The input is chat details data, and the output is stored database entries.

[1237] Through the above process, this system realizes natural communication that takes into account the user's emotions and context.

[1238] (Application example 2)

[1239] 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."

[1240] To ensure smooth communication with customers in physical stores, it is necessary to respond quickly and appropriately to their questions and requests. However, it is not easy to accurately understand the customer's emotions and context and respond based on them. In particular, it is difficult to respond appropriately according to the customer's emotional state, which can lead to a decrease in customer satisfaction and an unsatisfactory experience.

[1241] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1242] In this invention, the server includes means for receiving input text, means for analyzing the context of the received text, means for analyzing the emotion of the user, means for generating an optimal translation based on the analyzed context, means for adjusting a response based on the recognized emotion, means for generating a response based on the generated translation, means for displaying the generated response, means for saving past interactions, and means for supporting customer service in a physical store. This allows for the rapid generation of appropriate responses that take into account the emotion and context of the customer, thereby improving the efficiency of customer service in a physical store and customer satisfaction.

[1243] The "means for receiving input text" is a function for receiving text input from a user and sending the data to a server for analysis.

[1244] The "means for analyzing the context of received text" is a function for analyzing received text data and understanding the context according to the meaning and use of the text.

[1245] "Emotion analysis means for recognizing user emotions" is a function for analyzing and identifying user emotions (positive, negative, neutral, etc.) based on keywords and phrases in the text.

[1246] The "means for generating optimal translations based on analyzed context" is a function for generating appropriate and natural translations according to the context and sentiment.

[1247] The "means for adjusting a response based on a recognized emotion" is a function for taking into account the emotional state of the user and adjusting the response so that an optimized response is generated.

[1248] The "means for generating a response based on the generated translation" is a function for generating an appropriate response based on the translated text.

[1249] The "means for displaying the generated response" is a function for displaying the generated response in a form that is easy for the user to see.

[1250] "Means for storing past interactions" is a function that records all interactions as data and stores them for future reference.

[1251] "Means to support customer service in physical stores" refers to support functions that facilitate smooth communication with customers in physical stores.

[1252] This invention is a system for facilitating customer service in brick-and-mortar stores. The system receives user input text, analyzes context and sentiment to generate an optimal translation, and provides a series of functions to display a response based on that. Specific embodiments are described below.

[1253] System configuration

[1254] The system mainly consists of the following elements:

[1255] A means of receiving input text

[1256] A means of analyzing the context of received text

[1257] Emotion analysis means to recognize user emotions

[1258] A means of generating optimal translations based on analyzed context

[1259] A means of tailoring responses based on perceived emotions

[1260] A means of generating a response based on the generated translation

[1261] A means of displaying the generated response

[1262] A way to preserve past interactions

[1263] A means to support customer service in physical stores

[1264] System operation details

[1265] Receiving text input

[1266] A user inputs text using a terminal (e.g., a smartphone) in a physical store. For example, if the user inputs "Please check product inventory," the terminal receives this text and prepares it for transmission to the server.

[1267] Context and Sentiment Analysis

[1268] When the server receives text from a device, it first passes it to a context analysis module, which uses natural language processing algorithms (for example, Python's Hugging Face library) to extract keywords and phrases from the text and analyze its context. At the same time, a sentiment analysis module identifies the user's sentiment from the text.

[1269] For example, the text "Please check product availability" is a request to check product availability, and the sentiment is judged to be slightly negative.

[1270] Translation generation and adjustment

[1271] Based on the analyzed context and sentiment information, the server calls the translation engine module, which generates a translation that best suits the context and sentiment, such as "Could you please confirm the stock availability? Thank you."

[1272] The system then tailors the response based on the recognized sentiment. For example, for text with a negative sentiment like "Please check the product availability," it generates a sentiment-sensitive response. In this case, it generates a response like "The stock availability has been checked. Thank you for your patience."

[1273] View and save responses

[1274] The generated response is sent from the server to the terminal, which then displays it to the user, allowing the user to see the response tailored to the situation in real time.

[1275] All interactions are recorded and saved by the server's chat history management module, which records detailed information such as the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses in a database for future reference and analysis.

[1276] Examples of concrete examples and prompts

[1277] For example, consider the following exchange:

[1278] User Input:

[1279] "Please check product availability."

[1280] System response:

[1281] "The stock availability has been checked. Thank you for your patience."

[1282] Example prompt sentence:

[1283] "A user types, 'Check product availability.' Use contextual analysis and sentiment identification to generate an appropriate response and translation."

[1284] Contextual analysis results: Check product availability

[1285] Emotion identification result: Negative

[1286] Translation to generate: The stock availability has been checked.

[1287] Generate response: We have confirmed the product is in stock. Thank you.

[1288] This system will enable the rapid generation of appropriate responses that take into account customer emotions and context, improving customer service efficiency and customer satisfaction in physical stores.

[1289] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1290] Step 1:

[1291] The user inputs text. The user uses a device such as a smartphone in a physical store to input a question or request about a product. An example of input text is "Please check product availability." The device receives this text.

[1292] Input: User text input

[1293] Output: Received text data

[1294] Step 2:

[1295] The terminal sends the received text to the server. At this time, the text data is safely delivered to the server via the network. The server receives this text data.

[1296] Input: Text data sent from the terminal

[1297] Output: Text data received by the server

[1298] Step 3:

[1299] The server passes the received text data to a context analysis module, which uses a natural language processing algorithm (e.g., Python's Hugging Face library) to extract keywords and phrases from the text. For example, it extracts "product inventory" and determines that the context is a business context.

[1300] Input: Text data received by the server

[1301] Output: Context information (e.g. product availability, business context)

[1302] Step 4:

[1303] The server simultaneously passes the text to a sentiment analysis module, which identifies the user's sentiment from the text. For example, a request like "Please check product availability" is identified as a slightly negative sentiment. This analysis is performed using a generative AI model.

[1304] Input: Text data received by the server

[1305] Output: Emotional information (e.g., negative)

[1306] Step 5:

[1307] The server passes the results of the contextual and sentiment analysis to the translation engine module, which generates the optimal translation based on the analysis results. For example, the generated text might be "Can you please check the stock availability? Thank you." A generative AI model is also used at this stage.

[1308] Input: Contextual and emotional information

[1309] Output:Translated text

[1310] Step 6:

[1311] The translated text is passed to a response generation module, which adjusts the response based on the sentiment analysis results. For example, if the sentiment is negative, a more considerate response such as "The stock availability has been checked. Thank you for your patience" will be generated.

[1312] Input: Translation text and sentiment information

[1313] Output: Adjusted response text

[1314] Step 7:

[1315] The server sends the generated response text to the terminal, which displays it to the user, who can see the appropriately tailored response in real time.

[1316] Input: Reply text sent from the server

[1317] Output: The response text displayed to the user

[1318] Step 8:

[1319] All interactions are recorded and saved by the chat history management module on the server, including the date and time of each chat, content, context analysis results, sentiment analysis results, translation results, and final responses. The saved data is available for future reference and analysis.

[1320] Input: Various generated data (date, time, content, analysis results, final response, etc.)

[1321] Output: Saved chat history

[1322] 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.

[1323] 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.

[1324] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1325] 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.

[1326] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1327] 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.

[1328] 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).

[1329] 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.

[1330] 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."

[1331] 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.

[1332] 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).

[1333] 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.

[1334] 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.

[1335] 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.

[1336] 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.

[1337] 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.

[1338] 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.

[1339] 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.

[1340] 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.

[1341] 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.

[1342] 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.

[1343] The following is further disclosed regarding the above embodiment.

[1344] (Claim 1)

[1345] means for receiving input text;

[1346] means for analyzing the context of the received text;

[1347] means for generating an optimal translation based on the analyzed context;

[1348] means for generating a response based on the generated translation;

[1349] a means for displaying the generated response;

[1350] A way to preserve past interactions,

[1351] A system including:

[1352] (Claim 2)

[1353] 2. The system of claim 1, wherein the contextual analysis means comprises an algorithm for classifying text as business, casual, or formal.

[1354] (Claim 3)

[1355] 10. The system of claim 1, wherein the generated translation is text that is optimized for the context using natural language processing algorithms.

[1356] "Example 1"

[1357] (Claim 1)

[1358] means for receiving text entered by a user through the terminal;

[1359] means for analyzing the context of the received text;

[1360] means for generating an optimal translation based on the analyzed context;

[1361] means for generating a response based on the generated translation;

[1362] means for displaying the generated response on a terminal;

[1363] A way to preserve past interactions,

[1364] A system including:

[1365] (Claim 2)

[1366] 2. The system of claim 1, wherein the contextual analysis means comprises an algorithm for classifying text as business, casual, or formal.

[1367] (Claim 3)

[1368] 10. The system of claim 1, wherein the generated translation is text that is optimized for the context using natural language processing algorithms.

[1369] (Claim 4)

[1370] 10. The system of claim 1, wherein the means for generating a translation comprises an algorithm that uses a generative AI model to generate a translation based on the prompt sentence.

[1371] (Claim 5)

[1372] 10. The system of claim 1, wherein the terminal includes means for packaging input text in JSON format and sending it to the server as an HTTP POST request.

[1373] (Claim 6)

[1374] 2. The system of claim 1, wherein the server includes means for passing the text data to a context analysis module, the context analysis module using a natural language processing library to tokenize the text and perform keyword extraction.

[1375] "Application Example 1"

[1376] (Claim 1)

[1377] means for receiving input text;

[1378] means for analyzing the context of the received text;

[1379] means for generating an optimal translation based on the analyzed context;

[1380] means for generating a response based on the generated translation;

[1381] a means for displaying the generated response;

[1382] A way to preserve past interactions,

[1383] a means for receiving, parsing, and translating user input in multiple languages ​​to generate a response;

[1384] A means to display the generated translations and replies in a chat format and maintain a history;

[1385] A system including:

[1386] (Claim 2)

[1387] 10. The system of claim 1, wherein the context analysis means includes an algorithm for classifying text as business, casual, or formal, and further processes input in multiple languages.

[1388] (Claim 3)

[1389] 10. The system of claim 1, wherein the generated translation is text that is context-optimized using natural language processing algorithms and uses a generative AI model to improve the quality of the translation.

[1390] "Example 2: Combining Emotion Engines"

[1391] (Claim 1)

[1392] means for receiving input text;

[1393] means for analyzing the context of the received text;

[1394] means for analyzing user emotions;

[1395] means for generating an optimal translation based on the analyzed context and sentiment;

[1396] means for generating a response based on the generated translation;

[1397] a means for displaying the generated response;

[1398] A way to preserve past interactions,

[1399] A system including:

[1400] (Claim 2)

[1401] 2. The system of claim 1, wherein the context analysis means comprises an algorithm for classifying text as business, casual, or formal, and the sentiment analysis means classifies the sentiment of the text as positive, negative, or neutral.

[1402] (Claim 3)

[1403] 10. The system of claim 1, wherein the generated translation is text that is optimized for context and sentiment using natural language processing algorithms.

[1404] "Application example 2 when combining emotion engines"

[1405] (Claim 1)

[1406] means for receiving input text;

[1407] means for analyzing the context of the received text;

[1408] emotion analysis means for recognizing the emotion of a user;

[1409] means for generating an optimal translation based on the analyzed context;

[1410] a means for tailoring responses based on the perceived emotion;

[1411] means for generating a response based on the generated translation;

[1412] a means for displaying the generated response;

[1413] A way to preserve past interactions,

[1414] A means to support customer service in physical stores,

[1415] A system including:

[1416] (Claim 2)

[1417] 2. The system of claim 1, wherein the contextual analysis means comprises an algorithm for classifying text as business, casual, or formal.

[1418] (Claim 3)

[1419] 10. The system of claim 1, wherein the generated translation is text that is context-optimized using natural language processing algorithms and includes responses that are tailored based on sentiment analysis results. [Explanation of symbols]

[1420] 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. means for receiving input text; means for analyzing the context of the received text; means for generating an optimal translation based on the analyzed context; means for generating a response based on the generated translation; a means for displaying the generated response; A way to preserve past interactions, A system including:

2. 2. The system of claim 1, wherein the contextual analysis means comprises an algorithm for classifying text as business, casual, or formal.

3. 10. The system of claim 1, wherein the generated translation is text that is optimized for the context using natural language processing algorithms.

Images