system

Abstract

Provide a system. 【Solution means】 Voice recognition means for converting customer voice data into language data in real time, Natural language processing means for analyzing the language data and extracting the subject of the inquiry, Memory search means for searching for relevant information from the inquiry history and knowledge base, Generative machine learning means for generating an optimal answer based on the relevant information, Information presentation means for displaying the generated answer on the display device of the person in charge in real time, Information storage means for recording the information after response as event information, Input means mounted on a smart device for receiving questions from customers in real time, A system including.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In conventional telephone reception services, there has been a problem that the response to inquiries from customers tends to be delayed, resulting in a decrease in customer satisfaction. The process of manually searching for and presenting the necessary information by the person in charge is inefficient and time-consuming, so customers have to wait. In addition, since it depends on the experience and skills of the person in charge, there may be variations in the response. To solve these problems, the introduction of a new system that enables quick and consistent responses is required.

Means for Solving the Problems

[0005] This invention provides speech recognition and natural language processing means that convert customer speech into text in real time, analyze the text, and automatically extract the subject of the inquiry. Furthermore, a database search means searches the inquiry history and knowledge base, and a generation AI means generates the optimal answer based on the relevant information. This generated answer is displayed in real time on the employee's terminal via an information presentation means, supporting rapid interaction with the customer. In addition, after the entire response process is completed, a storage means is provided to record the response details as incident information, which can be used to improve future operations. These means enable rapid and efficient customer response and improve customer satisfaction.

[0006] "Voice recognition means" refers to technology that converts voice data from customers into text data in real time.

[0007] "Natural language processing means" refers to processing technology that analyzes text data and automatically extracts the subject and content of a query.

[0008] A "database search method" is a technology that searches query history and knowledge bases to retrieve relevant information.

[0009] "Generative AI means" refers to artificial intelligence technology that generates the optimal answer to a customer inquiry based on information obtained from a database.

[0010] "Information presentation means" refers to technology that displays generated responses on the employee's terminal in real time, supporting communication with customers.

[0011] "Memory management" refers to the technology of recording and storing information and results obtained after an incident as incident information. [Brief explanation of the drawing]

[0012] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0013] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

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

[0015] In the following embodiments, a labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0016] In the following embodiments, a labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, a labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0018] In the following embodiments, a labeled communication I / F (Interface) is an interface including a communication processor and an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark), etc.

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0020] [First Embodiment]

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

[0022] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0023] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0025] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

[0027] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0029] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0033] The system of the present invention is designed to streamline customer service and improve service quality. Its embodiments are described in detail below.

[0034] This system primarily consists of a server, terminals, and users. When a user contacts the service desk via telephone, the server uses speech recognition to convert the user's voice into text data in real time. This converted text data is then analyzed by the server using natural language processing to extract the subject and sentiment of the inquiry.

[0035] Next, the server uses a database search mechanism to retrieve past inquiry history and knowledge base information based on the extracted information. Using this retrieved information, the server utilizes a generation AI mechanism to generate the optimal answer. This answer includes information to address the user's specific question and is displayed on the terminal in real time through an information display mechanism.

[0036] The terminal receives this generated information and displays it on the screen, helping staff members respond to users quickly. By obtaining the answers users need, the efficiency of counter services is significantly improved. After the interaction is complete, a record of the interaction performed by the staff member on the terminal is saved to the server by a storage device and stored in the database as incident information.

[0037] As a concrete example, consider a scenario where a user asks a question about their account password. The server transcribes the user's voice into text and analyzes it to identify the topic related to "password reset." It searches the database for past reset procedures and uses AI generation to create the optimal procedure. This procedure is displayed on the staff member's terminal, allowing the staff member to immediately guide the user through the process.

[0038] This allows the system to respond quickly and accurately, thereby increasing customer satisfaction.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] When a user calls the service desk and communicates their inquiry verbally, the server captures the audio. Using speech recognition technology, this audio data is converted into text data in real time.

[0042] Step 2:

[0043] After receiving the text data, the server performs analysis using natural language processing techniques. It identifies the subject of the inquiry, simultaneously conducts customer sentiment analysis, and extracts relevant information.

[0044] Step 3:

[0045] Based on the extracted information, the server uses database search tools to retrieve past query history and knowledge base information. This information includes previous similar cases and related solutions.

[0046] Step 4:

[0047] The server uses AI generation tools to create the optimal answer based on the acquired information. The prediction algorithm also calculates possible next questions and prepares additional information.

[0048] Step 5:

[0049] The generated response and additional information are sent from the server to the terminal. The terminal receives this information and displays it on the screen in real time. This information display allows the person in charge to respond to the user quickly.

[0050] Step 6:

[0051] After the interaction with the user is complete, the details and results of the interaction are confirmed on the terminal and sent to the server. The server then uses a storage device to record this information as incident data in a database.

[0052] (Example 1)

[0053] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0054] In customer service, the challenge lies in providing prompt and accurate responses to improve customer satisfaction. In particular, efficient handling of inquiries via telephone and seamless information provision by staff are essential. Furthermore, it is necessary to effectively utilize past inquiry history and knowledge bases while addressing the diverse emotions of customers.

[0055] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0056] In this invention, the server includes speech recognition means for converting customer voice data into text data in real time, natural language processing means for analyzing the text data to extract the subject and sentiment of the inquiry, database search means for searching for relevant information from inquiry history and a knowledge base, generation modeling means for generating the optimal answer based on the relevant information, information presentation means for displaying the generated answer on a display device in real time, and storage means for recording post-handling information as situational information. This makes it possible to provide quick and accurate responses to customer inquiries and significantly improve the efficiency of customer service.

[0057] "Speech recognition means" refers to technology for converting speech data into text data in real time.

[0058] "Natural language processing means" refers to techniques that analyze text data and extract the subject and sentiment of an inquiry from it.

[0059] A "database search method" is a technology used to search query history and knowledge bases to identify relevant information.

[0060] "Generative modeling means" refers to artificial intelligence technology used to generate the optimal response based on acquired information.

[0061] "Information presentation means" refers to technology for displaying generated responses on a display device in real time.

[0062] A "memory device" is a technology that records information after an interaction as situational information and stores it for future reference.

[0063] The system of this invention is designed to streamline customer service by integrating means of speech recognition, natural language processing, database retrieval, generative modeling, information presentation, and memory. Specific embodiments utilizing each means are described below.

[0064] Speech recognition means

[0065] The server captures voice data from users in real time and converts it into text data using speech recognition technology. This process utilizes a speech recognition API (e.g., general-purpose speech recognition software) to enable accurate text conversion.

[0066] Natural language processing means

[0067] The server analyzes the converted text data using natural language processing software (e.g., a general-purpose natural language processing library). This analysis extracts the subject and sentiment of the query.

[0068] Database search means

[0069] The server uses the extracted information to search past query history and knowledge bases through the database. This search allows for the rapid collection of relevant information. A database management system (e.g., general-purpose database management software) assists in this process.

[0070] Generative modeling means

[0071] The server uses a generative AI model to create the optimal response to provide to the customer based on the collected information. This process utilizes generative modeling techniques (e.g., general-purpose generative AI models).

[0072] Information presentation means

[0073] The terminal displays the generated responses in real time, allowing staff to respond to users quickly. This enables effective customer service.

[0074] storage means

[0075] The response results recorded on the terminal are efficiently collected and stored on the server. This allows them to be managed as situational information that can be referenced at a later date.

[0076] Specific example

[0077] Let's take an example where a user inquires about their account password over the phone. The server converts the user's voice into text, and analysis identifies the topic as "password reset." The server searches its database for relevant procedures and uses a generative AI model to generate the reset procedure. This procedure is displayed on the terminal, allowing the staff to immediately guide the user through it.

[0078] Example of a prompt

[0079] "User: I forgot my account password. How can I reset it?"

[0080] This system is expected to expedite responses to customer inquiries and significantly improve service quality.

[0081] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0082] Step 1:

[0083] Receiving voice input

[0084] The user contacts the service desk via telephone. The server receives the voice data from the user in real time. The voice data, as input, is passed to the server in its original format.

[0085] Step 2:

[0086] Converting audio data to text

[0087] The server converts the received audio data into text data using speech recognition. Specifically, it analyzes the audio waveform using a speech recognition API, applies a language model, and converts the content into text format. This text data is then passed to the next step as output.

[0088] Step 3:

[0089] Text data analysis

[0090] The server analyzes the generated text data using natural language processing (NLP) tools. It uses NLP software to extract the subject and sentiment of the query from the input text data. The resulting subject and sentiment information is then output, allowing the server to proceed to the next step.

[0091] Step 4:

[0092] Search for related information

[0093] The server uses database search tools to retrieve past query history and knowledge base information based on the extracted subject information. The input consists of the subject information from the analysis results and the contents of the database, and the output is information related to the query. This related information is then passed on to the next step.

[0094] Step 5:

[0095] Generating answers using AI

[0096] The server uses a generative AI model based on the acquired relevant information to generate the optimal answer. Relevant information is provided to the AI ​​model as input and prompts, and a constructed answer is generated as output. This answer is then passed on to the next step.

[0097] Step 6:

[0098] Display the answer

[0099] The server generates a response, which is then sent to the terminal via an information display device. The terminal displays this response on the staff member's screen in real time. The outputted response is then presented to the staff member for user support.

[0100] Step 7:

[0101] Record of the results of the response

[0102] The information handled by the staff member on the terminal is sent to the server, and the server records this information as status data in a database using its storage device. The entered information is recorded exactly as it is and saved for future reference.

[0103] (Application Example 1)

[0104] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0105] There is a need to improve the real-time and efficiency of customer service, as well as the quality of service by taking customer emotions into consideration. Traditional methods have been problematic because they make it difficult to provide information to respond quickly and appropriately to a large number of customer inquiries. Furthermore, if information after an interaction is not properly managed, it is less likely to be used to improve future interactions, resulting in decreased customer satisfaction.

[0106] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0107] In this invention, the server includes a speech recognition means that converts customer voice data into language data in real time; a natural language processing means that analyzes the language data to extract the subject of the inquiry; a memory retrieval means that searches for relevant information from the inquiry history and knowledge base; a generation machine learning means that generates the optimal answer based on the relevant information; an information presentation means that displays the generated answer in real time on the employee's display device; an information storage means that records the information after the response as event information; and an input means that is mounted on a smart device and receives questions from customers in real time. This enables real-time and efficient customer service and high-quality service that responds to customer emotions.

[0108] "Customer voice data" refers to voice information emitted by customers, and is typically digital data of sound acquired during conversational interactions.

[0109] "Real-time" means that information processing and responses occur simultaneously with the event in which they occur, resulting in a state with extremely little delay.

[0110] "Language data" refers to text-based data obtained by converting audio data into characters, and it forms the basis for analysis using natural language processing.

[0111] "Speech recognition means" refers to a technology or device for converting speech data into text data, and typically includes a microphone and speech recognition software.

[0112] "Natural language processing methods" refer to technologies and algorithms used to analyze language data and extract its subject matter and emotions, and are used to understand the structure, meaning, and context of language.

[0113] A "memory retrieval system" is a system for retrieving relevant information from past inquiry history and knowledge bases, and generally utilizes a database management system.

[0114] "Generative machine learning methods" are algorithms and models used to generate optimal answers based on searched relevant information, and are also known as generative AI.

[0115] "Information presentation means" refers to a system for displaying the generated response on the operator's display device, which usually includes a display or monitor.

[0116] "Information storage means" refers to devices and technologies that properly store information after a transaction and use it for future reference, and databases and storage services often fall into this category.

[0117] A "smart device" is a general term for advanced electronic devices equipped with computers and communication functions, and possessing various functions including voice input.

[0118] "Input means" refers to methods or devices for taking user instructions or data into the system, and includes voice recognition devices and touch panels.

[0119] The system for implementing this invention consists of a server, terminals, and users to efficiently handle customer interactions. The server uses speech recognition means to convert customer voice data into language data in real time. For this purpose, Google® Speech-to-Text is used as the speech recognition API. The converted language data is analyzed by a natural language processing engine. Here, spaCy or AllenNLP is used to extract the subject and sentiment of the inquiry.

[0120] The server then searches its query history and knowledge base through memory retrieval mechanisms. This typically involves using a database management system such as MongoDB. Based on the relevant information, it then generates the optimal answer using machine learning tools. Generative AI models such as OpenAI® GPT-4® are used for this generation.

[0121] The generated answers are displayed in real time on the employee's terminal via an information display device. Smart devices, such as smart glasses or head-mounted displays, receive customer questions and function as input devices. This allows the terminal to provide answers instantly, enabling rapid customer service.

[0122] Furthermore, information after the response is stored on a server using an information storage device and added to a database as incident information. This entire process supports the smooth and accurate progress of customer support.

[0123] As a concrete example, consider a scenario where a customer asks, "What is the return policy for this product?" The server analyzes the voice data, extracts relevant information about the policy, generates a detailed explanation using a generative AI model, and displays it on the device.

[0124] Examples of prompt statements include:

[0125] "Customer question: 'What is the return policy for this product?' Based on this question, please briefly explain the detailed return policy."

[0126] These are some examples.

[0127] This system will enable real-time, high-quality customer service at stores.

[0128] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0129] Step 1:

[0130] The user inputs a question by voice through a smart device. This voice is transmitted to the server as digital voice data.

[0131] Step 2:

[0132] The server uses speech recognition to convert the received audio data into text data. In this process, a speech recognition API is used to process the audio data into readable language data.

[0133] Step 3:

[0134] The server uses natural language processing (NLP) tools to analyze text data and extract the subject and sentiment of the query. Specifically, it uses an NLP engine to analyze sentence structure and emotional tone to determine the subject matter.

[0135] Step 4:

[0136] The server uses memory retrieval means to search for relevant information from query history and knowledge bases based on the extracted subject. A database management system is used to extract relevant information and past answers.

[0137] Step 5:

[0138] The server uses generative machine learning to generate the best possible answers based on the searched information. It leverages generative AI models to generate natural-sounding sentences based on the information, creating appropriate answers to the user's questions.

[0139] Step 6:

[0140] The generated responses are transmitted in real time to the terminal's display device via an information presentation system and provided to the user. The terminal visually displays the received information, allowing the user to confirm it immediately.

[0141] Step 7:

[0142] The server stores the information after handling a request in a database using an information storage device, which is then used for later analysis and reference. This stored data is then used for processing subsequent queries.

[0143] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0144] This invention realizes a system that further improves the quality of customer service by combining an emotion engine. The embodiments are described in detail below.

[0145] This system primarily consists of a server that receives voice input from users, a terminal that displays response information, and an emotion engine that performs emotion recognition. When a user connects to the service desk via telephone, the server uses speech recognition to convert the user's voice into text data. This text data is analyzed by natural language processing to extract the subject of the inquiry. At this stage, the server uses the emotion engine to recognize the user's emotions from the voice data in real time. The recognized emotions are then reflected in responses designed to enhance customer satisfaction.

[0146] The server further uses database search tools to search past inquiry history and knowledge bases, collecting relevant information. Based on this, a generation AI tool generates an optimal response that takes into account the user's emotional state. This response is transmitted to the terminal in real time via an information display tool and displayed to the customer service representative. Based on the displayed information, the representative can respond flexibly according to the customer's situation.

[0147] As a concrete example, consider a case where a user contacts the system feeling anxious because their product hasn't arrived. In this case, the server uses an emotion engine to recognize the user's anxiety from their voice and quickly generates a reassuring tone and information using AI. Subsequently, a response including a prompt delivery status check and a plan of action is displayed on the terminal. The staff member can then use this to take appropriate action to reassure the user.

[0148] In this way, by utilizing the emotion engine, this system can respond flexibly and accurately to diverse and emotional customer interactions, dramatically improving the quality of customer service.

[0149] The following describes the processing flow.

[0150] Step 1:

[0151] The user contacts the service desk by phone and verbally explains their question or problem. The server captures this audio and converts it into text data using speech recognition technology.

[0152] Step 2:

[0153] The server receives text data and uses natural language processing to analyze the subject of the query. This analysis is then used to prepare for more detailed information retrieval.

[0154] Step 3:

[0155] The server activates an emotion engine to analyze the user's emotions from the voice data. This analysis helps determine what emotions the user is experiencing, such as dissatisfaction, worry, or impatience.

[0156] Step 4:

[0157] Based on the acquired subject and sentiment data, the server uses database search methods to search the query history and knowledge base to extract relevant solutions and information.

[0158] Step 5:

[0159] The server uses AI generation tools to create the optimal response, taking into account the extracted information and the user's emotions. This response includes emotionally sensitive language and countermeasures.

[0160] Step 6:

[0161] The generated responses are sent from the server to the terminal in real time. The terminal receives this information and displays it on the screen for the person in charge to review.

[0162] Step 7:

[0163] After the conversation with the user ends, the terminal summarizes the details of the interaction and records them as incident information on the server. This record is stored in a database and used for future improvements and analysis.

[0164] (Example 2)

[0165] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0166] In customer service, it is essential to accurately recognize customer emotions from their voice and provide appropriate responses quickly based on those emotions. However, conventional systems struggle to respond while considering customer emotions, which can lead to a decline in the quality of service. In particular, the inability to provide real-time responses that respond to customer emotions can lead to a decrease in customer satisfaction.

[0167] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0168] In this invention, the server includes means for converting customer voice information into text information in real time, means for analyzing the text information to extract the subject of the communication, and means for recognizing the customer's emotions from the voice information. This makes it possible to generate and provide an appropriate response in real time that corresponds to the customer's emotions.

[0169] "Speech recognition means" refers to technology that converts speech information into text information in real time.

[0170] "Natural language processing means" refers to technologies for analyzing textual information and extracting the main topic of a message.

[0171] "Emotion analysis means" refers to technology that recognizes customer emotions in real time from voice information.

[0172] A "database search method" is a technology that searches for relevant information from past contact history and knowledge bases.

[0173] A "generative AI model" is an artificial intelligence technology that generates the optimal response by taking into account the customer's emotional state.

[0174] "Information presentation means" refers to a technology that displays the generated responses on the worker's terminal in real time.

[0175] A "memory device" is a technology that records and retains information after a response as case information.

[0176] This invention is a system that generates responses in real time based on customer voice information, thereby improving the quality of customer service. The system mainly consists of a server, terminals, and users.

[0177] The server uses speech recognition technology to receive voice input. The speech recognition software used is a common speech recognition API. This converts the user's voice information into text in real time. The converted text is then parsed using a natural language processing library to extract the subject of the communication. A wide range of open-source natural language processing libraries are available.

[0178] Subsequently, the server uses emotion analysis tools to recognize the customer's emotions from the audio information. This process utilizes emotion analysis software to determine the customer's emotions in real time. The recognized emotion information is used to generate responses that enhance customer satisfaction.

[0179] Furthermore, the server searches the customer's past inquiry history and knowledge base using a database management system. For example, a widely used relational database management system is used for the database system.

[0180] The generative AI model generates the optimal response within the server, taking into account the user's emotional state and the content of their inquiry. An advanced language model is applied to this generative AI model. This model generates natural-sounding answers based on the context and the customer's situation.

[0181] The generated response is transmitted to the terminal via an information display device. The terminal displays this information to the worker in real time, helping them to provide the best possible service to the customer.

[0182] For example, if a user contacts the support center because "the product hasn't arrived," the system sends information to the AI ​​model in the form of a prompt message such as, "The user is feeling anxious about the delay in product delivery. Please suggest measures to alleviate this anxiety." This allows the support staff to quickly provide reassurance to the user.

[0183] Thus, by combining speech recognition, emotion analysis technology, and a generative AI model, the system of the present invention enables flexible and accurate responses tailored to customer emotions, dramatically improving the quality of customer service.

[0184] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0185] Step 1:

[0186] The user connects to the service desk via telephone and makes an inquiry in voice format. The input here is the user's voice, which the server receives. The user's voice is the first data input for the system.

[0187] Step 2:

[0188] The server uses speech recognition to convert the received user's voice into text information in real time. Specifically, speech recognition software analyzes the speech waveform and generates corresponding text data. Here, the input is the audio signal, and the output is the text information.

[0189] Step 3:

[0190] The server analyzes the converted character data using natural language processing tools. Specifically, it uses a natural language processing library to extract the subject and intent of the customer's inquiry from the string. The input here is text data, and the output is the extracted subject and intent information.

[0191] Step 4:

[0192] The server uses emotion analysis tools to recognize customer emotions from voice and text data. Specifically, emotion analysis software evaluates features such as language and tone of voice to identify emotions. Input is voice and text, and output is emotion information.

[0193] Step 5:

[0194] The server utilizes database search capabilities to retrieve past customer inquiry history and knowledge base data. Here, the database system executes queries to collect relevant information. The input is the extracted subject, and the output is the related historical data.

[0195] Step 6:

[0196] The server uses a generative AI model to generate the optimal response, taking into account the customer's emotional state and the content of their inquiry. Specifically, the generative AI model generates a contextually natural response based on the prompt text. The inputs are the subject, emotional information, and historical data, and the output is the generated response.

[0197] Step 7:

[0198] The server sends the generated response to the terminal via an information display device. The terminal displays this information on the operator's screen in real time. The input here is the generated response data, and the output is the display on the terminal screen. The person in charge uses this information to take appropriate action for the customer.

[0199] (Application Example 2)

[0200] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0201] In modern e-commerce websites, simply presenting product information when customers obtain it via voice can lead to decreased customer satisfaction. Furthermore, the lack of personalized responses that take customer emotions into account makes it difficult to provide prompt and effective customer support.

[0202] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0203] In this invention, the server includes speech recognition means for converting customer voice information into text information in real time, natural language processing means for analyzing the text information and extracting the subject of the inquiry, and emotion recognition means for detecting the customer's emotional state and generating a response based on that emotional state. This makes it possible to provide customized information that takes into account the customer's emotions when they make an inquiry by voice.

[0204] "Customer voice information" refers to voice data obtained from users, and includes information such as the customer's intentions and emotions.

[0205] "Textual information" refers to text data converted from audio data by speech recognition technology.

[0206] "Speech recognition means" refers to technologies and devices that convert speech data into text data in real time.

[0207] "Natural language processing means" refers to techniques for analyzing text data and extracting the subject of a query.

[0208] "Information retrieval means" refers to technologies and devices for searching for relevant information from inquiry history or knowledge bases.

[0209] "Generative AI means" refers to artificial intelligence technology used to generate the optimal answer based on relevant information.

[0210] "Information presentation means" refers to technologies or devices that display generated answers on the employee's terminal in real time.

[0211] "Storage methods" refer to technologies and devices for recording information after a response as event information.

[0212] "Emotion recognition means" refers to technology that detects a customer's emotional state from voice information and generates a response based on that.

[0213] This application example is an embodiment of a system that improves customer support using customer voice information. The system consists of voice recognition means, natural language processing means, information retrieval means, generation AI means, information presentation means, storage means, and emotion recognition means. These means are combined to analyze customer voice input and generate appropriate responses.

[0214] First, when a user makes a voice inquiry to the service using their smartphone, the server converts the voice data into text using speech recognition technology. By using software such as the Google Speech-to-Text API for speech recognition, it is possible to convert voice data into text with high accuracy.

[0215] Next, the server analyzes the text information using natural language processing techniques to extract the subject of the query. In this step, natural language processing technology using TENSORFLOW® is employed to accurately grasp the user's intent.

[0216] Subsequently, the emotion recognition system uses emotion analysis software such as IBM Watson® Tone Analyzer to detect the customer's emotions and passes them to the generative AI system. The generative AI system uses OpenAI's GPT model to generate answers that enable personalized responses that take into account the customer's emotional state.

[0217] The generated responses are displayed realistically and in real time on the user's device through an information presentation system. This allows the user to receive responses that match their emotions. Finally, all response information is recorded as event information using a storage system.

[0218] For example, if a customer asks, "I want this product right now, do you have it in stock?", the system recognizes the sentiment as "hopeful," quickly searches for inventory information, and provides a customized response such as, "It is currently out of stock in our online store, but we can check the stock at your nearest store."

[0219] Examples of prompts to input into a generative AI model:

[0220] "User input: Do you have it in stock? Emotion: Hopeful. Answer:"

[0221] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0222] Step 1:

[0223] The user makes a voice inquiry using their smartphone. This voice data is sent to the server. The input here is the user's voice data. The server receives this voice data and sends it to the next processing step.

[0224] Step 2:

[0225] The server converts audio data into text using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to convert speech to text. The input here is audio data, and the output is text. The server then passes this text to the next step.

[0226] Step 3:

[0227] The server analyzes the received text information using natural language processing techniques to extract the subject of the query. Natural language processing techniques using TensorFlow are employed here. The input is text information, and the output is parsed data including the topic.

[0228] Step 4:

[0229] The server passes the analyzed data to an emotion recognition system to detect the customer's emotional state. Specifically, it uses IBM Watson Tone Analyzer to analyze the sentiment of textual information and identify the emotional state. The input is the analyzed data, and the output is the emotional state.

[0230] Step 5:

[0231] The server uses generative AI to generate the optimal response, taking into account the emotional state. This process uses OpenAI's GPT model to create personalized responses tailored to the customer's emotions. The input is the emotional state and the subject of the inquiry, and the output is the customized response.

[0232] Step 6:

[0233] The server displays the generated response on the user's terminal using an information display mechanism. The response is displayed in real time, allowing the user to review it. The input is the customized response, and the output is the information displayed on the terminal.

[0234] Step 7:

[0235] The server records all interactions using storage methods and saves them as event information in a database. This provides a history of interactions that can be referenced later. The input consists of the response and data of its processing, while the output is the saved event information.

[0236] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0237] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0238] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0239] [Second Embodiment]

[0240] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0241] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0242] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0243] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0244] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0245] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0246] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0247] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0248] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0249] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0250] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0251] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0252] The system of the present invention is designed to streamline customer service and improve service quality. Its embodiments are described in detail below.

[0253] This system primarily consists of a server, terminals, and users. When a user contacts the service desk via telephone, the server uses speech recognition to convert the user's voice into text data in real time. This converted text data is then analyzed by the server using natural language processing to extract the subject and sentiment of the inquiry.

[0254] Next, the server uses a database search mechanism to retrieve past inquiry history and knowledge base information based on the extracted information. Using this retrieved information, the server utilizes a generation AI mechanism to generate the optimal answer. This answer includes information to address the user's specific question and is displayed on the terminal in real time through an information display mechanism.

[0255] The terminal receives this generated information and displays it on the screen, helping staff members respond to users quickly. By obtaining the answers users need, the efficiency of counter services is significantly improved. After the interaction is complete, a record of the interaction performed by the staff member on the terminal is saved to the server by a storage device and stored in the database as incident information.

[0256] As a concrete example, consider a scenario where a user asks a question about their account password. The server transcribes the user's voice into text and analyzes it to identify the topic related to "password reset." It searches the database for past reset procedures and uses AI generation to create the optimal procedure. This procedure is displayed on the staff member's terminal, allowing the staff member to immediately guide the user through the process.

[0257] This allows the system to respond quickly and accurately, thereby increasing customer satisfaction.

[0258] The following describes the processing flow.

[0259] Step 1:

[0260] When a user calls the service desk and communicates their inquiry verbally, the server captures the audio. Using speech recognition technology, this audio data is converted into text data in real time.

[0261] Step 2:

[0262] After receiving the text data, the server performs analysis using natural language processing techniques. It identifies the subject of the inquiry, simultaneously conducts customer sentiment analysis, and extracts relevant information.

[0263] Step 3:

[0264] Based on the extracted information, the server uses database search tools to retrieve past query history and knowledge base information. This information includes previous similar cases and related solutions.

[0265] Step 4:

[0266] The server uses AI generation tools to create the optimal answer based on the acquired information. The prediction algorithm also calculates possible next questions and prepares additional information.

[0267] Step 5:

[0268] The generated response and additional information are sent from the server to the terminal. The terminal receives this information and displays it on the screen in real time. This information display allows the person in charge to respond to the user quickly.

[0269] Step 6:

[0270] After the interaction with the user is complete, the details and results of the interaction are confirmed on the terminal and sent to the server. The server then uses a storage device to record this information as incident data in a database.

[0271] (Example 1)

[0272] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0273] In customer service, the challenge lies in providing prompt and accurate responses to improve customer satisfaction. In particular, efficient handling of inquiries via telephone and seamless information provision by staff are essential. Furthermore, it is necessary to effectively utilize past inquiry history and knowledge bases while addressing the diverse emotions of customers.

[0274] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0275] In this invention, the server includes speech recognition means for converting customer voice data into text data in real time, natural language processing means for analyzing the text data to extract the subject and sentiment of the inquiry, database search means for searching for relevant information from inquiry history and a knowledge base, generation modeling means for generating the optimal answer based on the relevant information, information presentation means for displaying the generated answer on a display device in real time, and storage means for recording post-handling information as situational information. This makes it possible to provide quick and accurate responses to customer inquiries and significantly improve the efficiency of customer service.

[0276] "Speech recognition means" refers to technology for converting speech data into text data in real time.

[0277] "Natural language processing means" refers to techniques that analyze text data and extract the subject and sentiment of an inquiry from it.

[0278] A "database search method" is a technology used to search query history and knowledge bases to identify relevant information.

[0279] "Generative modeling means" refers to artificial intelligence technology used to generate the optimal response based on acquired information.

[0280] "Information presentation means" refers to technology for displaying generated responses on a display device in real time.

[0281] The "memory means" is a technology that records the information after correspondence as situation information and stores it for future reference.

[0282] The system of this invention is designed to integrate various means such as speech recognition, natural language processing, database search, generation modeling, information presentation, and memory to improve customer response efficiency. Specific embodiments using each means are shown below.

[0283] Speech recognition means

[0284] The server captures voice data from the user in real time and converts it into text data using speech recognition technology. For this process, a speech recognition API (e.g., general-purpose speech recognition software) is used to enable accurate text conversion.

[0285] Natural language processing means

[0286] The server analyzes the converted text data through natural language processing software (e.g., general natural language processing library). Through this analysis, the subject and sentiment of the inquiry are extracted.

[0287] Database search means

[0288] The server searches the database for past inquiry histories and knowledge bases based on the extracted information. Through this search, relevant information is quickly collected. A database management system (e.g., general-purpose database management software) supports this.

[0289] Generation modeling means

[0290] The server creates an optimal answer to provide to the customer using a generation AI model based on the collected information. In this process, generation modeling technology (e.g., general-purpose generation AI model) is utilized.

[0291] Information presentation means

[0292] The terminal displays the generated responses in real time, allowing staff to respond to users quickly. This enables effective customer service.

[0293] storage means

[0294] The response results recorded on the terminal are efficiently collected and stored on the server. This allows them to be managed as situational information that can be referenced at a later date.

[0295] Specific example

[0296] Let's take an example where a user inquires about their account password over the phone. The server converts the user's voice into text, and analysis identifies the topic as "password reset." The server searches its database for relevant procedures and uses a generative AI model to generate the reset procedure. This procedure is displayed on the terminal, allowing the staff to immediately guide the user through it.

[0297] Example of a prompt

[0298] "User: I forgot my account password. How can I reset it?"

[0299] This system is expected to expedite responses to customer inquiries and significantly improve service quality.

[0300] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0301] Step 1:

[0302] Receiving voice input

[0303] The user contacts the service desk via telephone. The server receives the voice data from the user in real time. The voice data, as input, is passed to the server in its original format.

[0304] Step 2:

[0305] Text conversion of voice data

[0306] The server converts the received voice data into text data using voice recognition means. Specifically, it analyzes the voice waveform using a voice recognition API and applies a language model to convert its content into text format. This text data is passed as output to the next step.

[0307] Step 3:

[0308] Analysis of text data

[0309] The server analyzes the generated text data using natural language processing means. For the input text data, an operation is performed to extract the subject and sentiment of the inquiry using natural language processing software. The subject and sentiment information as the analysis result proceeds as output to the next step.

[0310] Step 4:

[0311] Search for relevant information

[0312] The server searches the past inquiry history and knowledge base using database search means based on the extracted subject information. The input includes the subject information of the analysis result and the content of the database, and relevant information related to the inquiry is obtained as output. This relevant information is passed to the next step.

[0313] Step 5:

[0314] Creation of an answer by generative AI

[0315] The server uses a generative AI model based on the obtained relevant information to generate an optimal answer. The relevant information is given to the AI model as a prompt for the input information, and an answer constructed as output is generated. This answer is passed to the next step.

[0316] Step 6:

[0317] Display the answer

[0318] The server generates a response, which is then sent to the terminal via an information display device. The terminal displays this response on the staff member's screen in real time. The outputted response is then presented to the staff member for user support.

[0319] Step 7:

[0320] Record of the results of the response

[0321] The information handled by the staff member on the terminal is sent to the server, and the server records this information as status data in a database using its storage device. The entered information is recorded exactly as it is and saved for future reference.

[0322] (Application Example 1)

[0323] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0324] There is a need to improve the real-time and efficiency of customer service, as well as the quality of service by taking customer emotions into consideration. Traditional methods have been problematic because they make it difficult to provide information to respond quickly and appropriately to a large number of customer inquiries. Furthermore, if information after an interaction is not properly managed, it is less likely to be used to improve future interactions, resulting in decreased customer satisfaction.

[0325] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0326] In this invention, the server includes a speech recognition means that converts customer voice data into language data in real time; a natural language processing means that analyzes the language data to extract the subject of the inquiry; a memory retrieval means that searches for relevant information from the inquiry history and knowledge base; a generation machine learning means that generates the optimal answer based on the relevant information; an information presentation means that displays the generated answer in real time on the employee's display device; an information storage means that records the information after the response as event information; and an input means that is mounted on a smart device and receives questions from customers in real time. This enables real-time and efficient customer service and high-quality service that responds to customer emotions.

[0327] "Customer voice data" refers to voice information emitted by customers, and is typically digital data of sound acquired during conversational interactions.

[0328] "Real-time" means that information processing and responses occur simultaneously with the event in which they occur, resulting in a state with extremely little delay.

[0329] "Language data" refers to text-based data obtained by converting audio data into characters, and it forms the basis for analysis using natural language processing.

[0330] "Speech recognition means" refers to a technology or device for converting speech data into text data, and typically includes a microphone and speech recognition software.

[0331] "Natural language processing methods" refer to technologies and algorithms used to analyze language data and extract its subject matter and emotions, and are used to understand the structure, meaning, and context of language.

[0332] A "memory retrieval system" is a system for retrieving relevant information from past inquiry history and knowledge bases, and generally utilizes a database management system.

[0333] "Generative machine learning methods" are algorithms and models used to generate optimal answers based on searched relevant information, and are also known as generative AI.

[0334] "Information presentation means" refers to a system for displaying the generated response on the operator's display device, which usually includes a display or monitor.

[0335] "Information storage means" refers to devices and technologies that properly store information after a transaction and use it for future reference, and databases and storage services often fall into this category.

[0336] A "smart device" is a general term for advanced electronic devices equipped with computers and communication functions, and possessing various functions including voice input.

[0337] "Input means" refers to methods or devices for taking user instructions or data into the system, and includes voice recognition devices and touch panels.

[0338] The system for implementing this invention consists of a server, terminals, and users to efficiently handle customer interactions. The server uses speech recognition means to convert customer voice data into language data in real time. For this purpose, Google Speech-to-Text or similar speech recognition APIs are used. The converted language data is analyzed by a natural language processing engine. Here, spaCy or AllenNLP is used to extract the subject and sentiment of the inquiry.

[0339] The server then searches its query history and knowledge base through memory retrieval mechanisms. This typically involves using a database management system such as MongoDB. Based on the relevant information, it then generates the optimal answer using machine learning tools. Generative AI models such as OpenAI GPT-4 are used for this generation.

[0340] The generated answers are displayed in real time on the employee's terminal via an information display device. Smart devices, such as smart glasses or head-mounted displays, receive customer questions and function as input devices. This allows the terminal to provide answers instantly, enabling rapid customer service.

[0341] Furthermore, information after the response is stored on a server using an information storage device and added to a database as incident information. This entire process supports the smooth and accurate progress of customer support.

[0342] As a concrete example, consider a scenario where a customer asks, "What is the return policy for this product?" The server analyzes the voice data, extracts relevant information about the policy, generates a detailed explanation using a generative AI model, and displays it on the device.

[0343] Examples of prompt statements include:

[0344] "Customer question: 'What is the return policy for this product?' Based on this question, please briefly explain the detailed return policy."

[0345] These are some examples.

[0346] This system will enable real-time, high-quality customer service at stores.

[0347] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0348] Step 1:

[0349] The user inputs a question by voice through a smart device. This voice is transmitted to the server as digital voice data.

[0350] Step 2:

[0351] The server uses speech recognition to convert the received audio data into text data. In this process, a speech recognition API is used to process the audio data into readable language data.

[0352] Step 3:

[0353] The server uses natural language processing (NLP) tools to analyze text data and extract the subject and sentiment of the query. Specifically, it uses an NLP engine to analyze sentence structure and emotional tone to determine the subject matter.

[0354] Step 4:

[0355] The server uses memory retrieval means to search for relevant information from query history and knowledge bases based on the extracted subject. A database management system is used to extract relevant information and past answers.

[0356] Step 5:

[0357] The server uses generative machine learning to generate the best possible answers based on the searched information. It leverages generative AI models to generate natural-sounding sentences based on the information, creating appropriate answers to the user's questions.

[0358] Step 6:

[0359] The generated responses are transmitted in real time to the terminal's display device via an information presentation system and provided to the user. The terminal visually displays the received information, allowing the user to confirm it immediately.

[0360] Step 7:

[0361] The server stores the information after handling a request in a database using an information storage device, which is then used for later analysis and reference. This stored data is then used for processing subsequent queries.

[0362] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0363] This invention realizes a system that further improves the quality of customer service by combining an emotion engine. The embodiments are described in detail below.

[0364] This system primarily consists of a server that receives voice input from users, a terminal that displays response information, and an emotion engine that performs emotion recognition. When a user connects to the service desk via telephone, the server uses speech recognition to convert the user's voice into text data. This text data is analyzed by natural language processing to extract the subject of the inquiry. At this stage, the server uses the emotion engine to recognize the user's emotions from the voice data in real time. The recognized emotions are then reflected in responses designed to enhance customer satisfaction.

[0365] The server further uses database search tools to search past inquiry history and knowledge bases, collecting relevant information. Based on this, a generation AI tool generates an optimal response that takes into account the user's emotional state. This response is transmitted to the terminal in real time via an information display tool and displayed to the customer service representative. Based on the displayed information, the representative can respond flexibly according to the customer's situation.

[0366] As a concrete example, consider a case where a user contacts the system feeling anxious because their product hasn't arrived. In this case, the server uses an emotion engine to recognize the user's anxiety from their voice and quickly generates a reassuring tone and information using AI. Subsequently, a response including a prompt delivery status check and a plan of action is displayed on the terminal. The staff member can then use this to take appropriate action to reassure the user.

[0367] In this way, by utilizing the emotion engine, this system can respond flexibly and accurately to diverse and emotional customer interactions, dramatically improving the quality of customer service.

[0368] The following describes the processing flow.

[0369] Step 1:

[0370] The user contacts the service desk by phone and verbally explains their question or problem. The server captures this audio and converts it into text data using speech recognition technology.

[0371] Step 2:

[0372] The server receives text data and uses natural language processing to analyze the subject of the query. This analysis is then used to prepare for more detailed information retrieval.

[0373] Step 3:

[0374] The server activates an emotion engine to analyze the user's emotions from the voice data. This analysis helps determine what emotions the user is experiencing, such as dissatisfaction, worry, or impatience.

[0375] Step 4:

[0376] Based on the acquired subject and sentiment data, the server uses database search methods to search the query history and knowledge base to extract relevant solutions and information.

[0377] Step 5:

[0378] The server uses AI generation tools to create the optimal response, taking into account the extracted information and the user's emotions. This response includes emotionally sensitive language and countermeasures.

[0379] Step 6:

[0380] The generated responses are sent from the server to the terminal in real time. The terminal receives this information and displays it on the screen for the person in charge to review.

[0381] Step 7:

[0382] After the conversation with the user ends, the terminal summarizes the details of the interaction and records them as incident information on the server. This record is stored in a database and used for future improvements and analysis.

[0383] (Example 2)

[0384] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0385] In customer service, it is essential to accurately recognize customer emotions from their voice and provide appropriate responses quickly based on those emotions. However, conventional systems struggle to respond while considering customer emotions, which can lead to a decline in the quality of service. In particular, the inability to provide real-time responses that respond to customer emotions can lead to a decrease in customer satisfaction.

[0386] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0387] In this invention, the server includes means for converting customer voice information into text information in real time, means for analyzing the text information to extract the subject of the communication, and means for recognizing the customer's emotions from the voice information. This makes it possible to generate and provide an appropriate response in real time that corresponds to the customer's emotions.

[0388] "Speech recognition means" refers to technology that converts speech information into text information in real time.

[0389] "Natural language processing means" refers to technologies for analyzing textual information and extracting the main topic of a message.

[0390] "Emotion analysis means" refers to technology that recognizes customer emotions in real time from voice information.

[0391] A "database search method" is a technology that searches for relevant information from past contact history and knowledge bases.

[0392] A "generative AI model" is an artificial intelligence technology that generates the optimal response by taking into account the customer's emotional state.

[0393] "Information presentation means" refers to a technology that displays the generated responses on the worker's terminal in real time.

[0394] A "memory device" is a technology that records and retains information after a response as case information.

[0395] This invention is a system that generates responses in real time based on customer voice information, thereby improving the quality of customer service. The system mainly consists of a server, terminals, and users.

[0396] The server uses speech recognition technology to receive voice input. The speech recognition software used is a common speech recognition API. This converts the user's voice information into text in real time. The converted text is then parsed using a natural language processing library to extract the subject of the communication. A wide range of open-source natural language processing libraries are available.

[0397] Subsequently, the server uses emotion analysis tools to recognize the customer's emotions from the audio information. This process utilizes emotion analysis software to determine the customer's emotions in real time. The recognized emotion information is used to generate responses that enhance customer satisfaction.

[0398] Furthermore, the server searches the customer's past inquiry history and knowledge base using a database management system. For example, a widely used relational database management system is used for the database system.

[0399] The generative AI model generates the optimal response within the server, taking into account the user's emotional state and the content of their inquiry. An advanced language model is applied to this generative AI model. This model generates natural-sounding answers based on the context and the customer's situation.

[0400] The generated response is transmitted to the terminal via an information display device. The terminal displays this information to the worker in real time, helping them to provide the best possible service to the customer.

[0401] For example, if a user contacts the support center because "the product hasn't arrived," the system sends information to the AI ​​model in the form of a prompt message such as, "The user is feeling anxious about the delay in product delivery. Please suggest measures to alleviate this anxiety." This allows the support staff to quickly provide reassurance to the user.

[0402] Thus, by combining speech recognition, emotion analysis technology, and a generative AI model, the system of the present invention enables flexible and accurate responses tailored to customer emotions, dramatically improving the quality of customer service.

[0403] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0404] Step 1:

[0405] The user connects to the service desk via telephone and makes an inquiry in voice format. The input here is the user's voice, which the server receives. The user's voice is the first data input for the system.

[0406] Step 2:

[0407] The server uses speech recognition to convert the received user's voice into text information in real time. Specifically, speech recognition software analyzes the speech waveform and generates corresponding text data. Here, the input is the audio signal, and the output is the text information.

[0408] Step 3:

[0409] The server analyzes the converted character data using natural language processing tools. Specifically, it uses a natural language processing library to extract the subject and intent of the customer's inquiry from the string. The input here is text data, and the output is the extracted subject and intent information.

[0410] Step 4:

[0411] The server uses emotion analysis tools to recognize customer emotions from voice and text data. Specifically, emotion analysis software evaluates features such as language and tone of voice to identify emotions. Input is voice and text, and output is emotion information.

[0412] Step 5:

[0413] The server utilizes database search capabilities to retrieve past customer inquiry history and knowledge base data. Here, the database system executes queries to collect relevant information. The input is the extracted subject, and the output is the related historical data.

[0414] Step 6:

[0415] The server uses a generative AI model to generate the optimal response, taking into account the customer's emotional state and the content of their inquiry. Specifically, the generative AI model generates a contextually natural response based on the prompt text. The inputs are the subject, emotional information, and historical data, and the output is the generated response.

[0416] Step 7:

[0417] The server sends the generated response to the terminal via an information display device. The terminal displays this information on the operator's screen in real time. The input here is the generated response data, and the output is the display on the terminal screen. The person in charge uses this information to take appropriate action for the customer.

[0418] (Application Example 2)

[0419] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0420] In modern e-commerce websites, simply presenting product information when customers obtain it via voice can lead to decreased customer satisfaction. Furthermore, the lack of personalized responses that take customer emotions into account makes it difficult to provide prompt and effective customer support.

[0421] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0422] In this invention, the server includes speech recognition means for converting customer voice information into text information in real time, natural language processing means for analyzing the text information and extracting the subject of the inquiry, and emotion recognition means for detecting the customer's emotional state and generating a response based on that emotional state. This makes it possible to provide customized information that takes into account the customer's emotions when they make an inquiry by voice.

[0423] "Customer voice information" refers to voice data obtained from users, and includes information such as the customer's intentions and emotions.

[0424] "Textual information" refers to text data converted from audio data by speech recognition technology.

[0425] "Speech recognition means" refers to technologies and devices that convert speech data into text data in real time.

[0426] "Natural language processing means" refers to techniques for analyzing text data and extracting the subject of a query.

[0427] "Information retrieval means" refers to technologies and devices for searching for relevant information from inquiry history or knowledge bases.

[0428] "Generative AI means" refers to artificial intelligence technology used to generate the optimal answer based on relevant information.

[0429] "Information presentation means" refers to technologies or devices that display generated answers on the employee's terminal in real time.

[0430] "Storage methods" refer to technologies and devices for recording information after a response as event information.

[0431] "Emotion recognition means" refers to technology that detects a customer's emotional state from voice information and generates a response based on that.

[0432] This application example is an embodiment of a system that improves customer support using customer voice information. The system consists of voice recognition means, natural language processing means, information retrieval means, generation AI means, information presentation means, storage means, and emotion recognition means. These means are combined to analyze customer voice input and generate appropriate responses.

[0433] First, when a user makes a voice inquiry to the service using their smartphone, the server converts the voice data into text using speech recognition technology. By using software such as the Google Speech-to-Text API for speech recognition, it is possible to convert voice data into text with high accuracy.

[0434] Next, the server analyzes the text information using natural language processing techniques to extract the subject of the query. In this step, natural language processing techniques using TensorFlow are employed to accurately grasp the user's intent.

[0435] Subsequently, the emotion recognition system uses emotion analysis software such as IBM Watson Tone Analyzer to detect the customer's emotions and passes them to the generative AI system. The generative AI system uses OpenAI's GPT model to generate answers that enable personalized responses that take into account the customer's emotional state.

[0436] The generated responses are displayed realistically and in real time on the user's device through an information presentation system. This allows the user to receive responses that match their emotions. Finally, all response information is recorded as event information using a storage system.

[0437] For example, if a customer asks, "I want this product right now, do you have it in stock?", the system recognizes the sentiment as "hopeful," quickly searches for inventory information, and provides a customized response such as, "It is currently out of stock in our online store, but we can check the stock at your nearest store."

[0438] Examples of prompts to input into a generative AI model:

[0439] "User input: Do you have it in stock? Emotion: Hopeful. Answer:"

[0440] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0441] Step 1:

[0442] The user makes a voice inquiry using their smartphone. This voice data is sent to the server. The input here is the user's voice data. The server receives this voice data and sends it to the next processing step.

[0443] Step 2:

[0444] The server converts audio data into text using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to convert speech to text. The input here is audio data, and the output is text. The server then passes this text to the next step.

[0445] Step 3:

[0446] The server analyzes the received text information using natural language processing techniques to extract the subject of the query. Natural language processing techniques using TensorFlow are employed here. The input is text information, and the output is parsed data including the topic.

[0447] Step 4:

[0448] The server passes the analyzed data to an emotion recognition system to detect the customer's emotional state. Specifically, it uses IBM Watson Tone Analyzer to analyze the sentiment of textual information and identify the emotional state. The input is the analyzed data, and the output is the emotional state.

[0449] Step 5:

[0450] The server uses generative AI to generate the optimal response, taking into account the emotional state. This process uses OpenAI's GPT model to create personalized responses tailored to the customer's emotions. The input is the emotional state and the subject of the inquiry, and the output is the customized response.

[0451] Step 6:

[0452] The server displays the generated response on the user's terminal using an information display mechanism. The response is displayed in real time, allowing the user to review it. The input is the customized response, and the output is the information displayed on the terminal.

[0453] Step 7:

[0454] The server records all interactions using storage methods and saves them as event information in a database. This provides a history of interactions that can be referenced later. The input consists of the response and data of its processing, while the output is the saved event information.

[0455] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0456] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0457] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0458] [Third Embodiment]

[0459] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0460] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0461] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0462] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0463] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0464] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0465] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0466] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0467] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0468] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0469] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0470] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0471] The system of the present invention is designed to streamline customer service and improve service quality. Its embodiments are described in detail below.

[0472] This system primarily consists of a server, terminals, and users. When a user contacts the service desk via telephone, the server uses speech recognition to convert the user's voice into text data in real time. This converted text data is then analyzed by the server using natural language processing to extract the subject and sentiment of the inquiry.

[0473] Next, the server uses a database search mechanism to retrieve past inquiry history and knowledge base information based on the extracted information. Using this retrieved information, the server utilizes a generation AI mechanism to generate the optimal answer. This answer includes information to address the user's specific question and is displayed on the terminal in real time through an information display mechanism.

[0474] The terminal receives this generated information and displays it on the screen, helping staff members respond to users quickly. By obtaining the answers users need, the efficiency of counter services is significantly improved. After the interaction is complete, a record of the interaction performed by the staff member on the terminal is saved to the server by a storage device and stored in the database as incident information.

[0475] As a concrete example, consider a scenario where a user asks a question about their account password. The server transcribes the user's voice into text and analyzes it to identify the topic related to "password reset." It searches the database for past reset procedures and uses AI generation to create the optimal procedure. This procedure is displayed on the staff member's terminal, allowing the staff member to immediately guide the user through the process.

[0476] This allows the system to respond quickly and accurately, thereby increasing customer satisfaction.

[0477] The following describes the processing flow.

[0478] Step 1:

[0479] When a user calls the service desk and communicates their inquiry verbally, the server captures the audio. Using speech recognition technology, this audio data is converted into text data in real time.

[0480] Step 2:

[0481] After receiving the text data, the server performs analysis using natural language processing techniques. It identifies the subject of the inquiry, simultaneously conducts customer sentiment analysis, and extracts relevant information.

[0482] Step 3:

[0483] Based on the extracted information, the server uses database search tools to retrieve past query history and knowledge base information. This information includes previous similar cases and related solutions.

[0484] Step 4:

[0485] The server uses AI generation tools to create the optimal answer based on the acquired information. The prediction algorithm also calculates possible next questions and prepares additional information.

[0486] Step 5:

[0487] The generated response and additional information are sent from the server to the terminal. The terminal receives this information and displays it on the screen in real time. This information display allows the person in charge to respond to the user quickly.

[0488] Step 6:

[0489] After the interaction with the user is complete, the details and results of the interaction are confirmed on the terminal and sent to the server. The server then uses a storage device to record this information as incident data in a database.

[0490] (Example 1)

[0491] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0492] In customer service, the challenge lies in providing prompt and accurate responses to improve customer satisfaction. In particular, efficient handling of inquiries via telephone and seamless information provision by staff are essential. Furthermore, it is necessary to effectively utilize past inquiry history and knowledge bases while addressing the diverse emotions of customers.

[0493] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0494] In this invention, the server includes speech recognition means for converting customer voice data into text data in real time, natural language processing means for analyzing the text data to extract the subject and sentiment of the inquiry, database search means for searching for relevant information from inquiry history and a knowledge base, generation modeling means for generating the optimal answer based on the relevant information, information presentation means for displaying the generated answer on a display device in real time, and storage means for recording post-handling information as situational information. This makes it possible to provide quick and accurate responses to customer inquiries and significantly improve the efficiency of customer service.

[0495] "Speech recognition means" refers to technology for converting speech data into text data in real time.

[0496] "Natural language processing means" refers to techniques that analyze text data and extract the subject and sentiment of an inquiry from it.

[0497] A "database search method" is a technology used to search query history and knowledge bases to identify relevant information.

[0498] "Generative modeling means" refers to artificial intelligence technology used to generate the optimal response based on acquired information.

[0499] "Information presentation means" refers to technology for displaying generated responses on a display device in real time.

[0500] A "memory device" is a technology that records information after an interaction as situational information and stores it for future reference.

[0501] The system of this invention is designed to streamline customer service by integrating means of speech recognition, natural language processing, database retrieval, generative modeling, information presentation, and memory. Specific embodiments utilizing each means are described below.

[0502] Speech recognition means

[0503] The server captures voice data from users in real time and converts it into text data using speech recognition technology. This process utilizes a speech recognition API (e.g., general-purpose speech recognition software) to enable accurate text conversion.

[0504] Natural language processing means

[0505] The server analyzes the converted text data using natural language processing software (e.g., a general-purpose natural language processing library). This analysis extracts the subject and sentiment of the query.

[0506] Database search means

[0507] The server uses the extracted information to search past query history and knowledge bases through the database. This search allows for the rapid collection of relevant information. A database management system (e.g., general-purpose database management software) assists in this process.

[0508] Generative modeling means

[0509] The server uses a generative AI model to create the optimal response to provide to the customer based on the collected information. This process utilizes generative modeling techniques (e.g., general-purpose generative AI models).

[0510] Information presentation means

[0511] The terminal displays the generated responses in real time, allowing staff to respond to users quickly. This enables effective customer service.

[0512] storage means

[0513] The response results recorded on the terminal are efficiently collected and stored on the server. This allows them to be managed as situational information that can be referenced at a later date.

[0514] Specific example

[0515] Let's take an example where a user inquires about their account password over the phone. The server converts the user's voice into text, and analysis identifies the topic as "password reset." The server searches its database for relevant procedures and uses a generative AI model to generate the reset procedure. This procedure is displayed on the terminal, allowing the staff to immediately guide the user through it.

[0516] Example of a prompt

[0517] "User: I forgot my account password. How can I reset it?"

[0518] This system is expected to expedite responses to customer inquiries and significantly improve service quality.

[0519] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0520] Step 1:

[0521] Receiving voice input

[0522] The user contacts the service desk via telephone. The server receives the voice data from the user in real time. The voice data, as input, is passed to the server in its original format.

[0523] Step 2:

[0524] Converting audio data to text

[0525] The server converts the received audio data into text data using speech recognition. Specifically, it analyzes the audio waveform using a speech recognition API, applies a language model, and converts the content into text format. This text data is then passed to the next step as output.

[0526] Step 3:

[0527] Text data analysis

[0528] The server analyzes the generated text data using natural language processing (NLP) tools. It uses NLP software to extract the subject and sentiment of the query from the input text data. The resulting subject and sentiment information is then output, allowing the server to proceed to the next step.

[0529] Step 4:

[0530] Search for related information

[0531] The server uses database search tools to retrieve past query history and knowledge base information based on the extracted subject information. The input consists of the subject information from the analysis results and the contents of the database, and the output is information related to the query. This related information is then passed on to the next step.

[0532] Step 5:

[0533] Generating answers using AI

[0534] The server uses a generative AI model based on the acquired relevant information to generate the optimal answer. Relevant information is provided to the AI ​​model as input and prompts, and a constructed answer is generated as output. This answer is then passed on to the next step.

[0535] Step 6:

[0536] Display the answer

[0537] The server generates a response, which is then sent to the terminal via an information display device. The terminal displays this response on the staff member's screen in real time. The outputted response is then presented to the staff member for user support.

[0538] Step 7:

[0539] Record of the results of the response

[0540] The information handled by the staff member on the terminal is sent to the server, and the server records this information as status data in a database using its storage device. The entered information is recorded exactly as it is and saved for future reference.

[0541] (Application Example 1)

[0542] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0543] There is a need to improve the real-time and efficiency of customer service, as well as the quality of service by taking customer emotions into consideration. Traditional methods have been problematic because they make it difficult to provide information to respond quickly and appropriately to a large number of customer inquiries. Furthermore, if information after an interaction is not properly managed, it is less likely to be used to improve future interactions, resulting in decreased customer satisfaction.

[0544] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0545] In this invention, the server includes a speech recognition means that converts customer voice data into language data in real time; a natural language processing means that analyzes the language data to extract the subject of the inquiry; a memory retrieval means that searches for relevant information from the inquiry history and knowledge base; a generation machine learning means that generates the optimal answer based on the relevant information; an information presentation means that displays the generated answer in real time on the employee's display device; an information storage means that records the information after the response as event information; and an input means that is mounted on a smart device and receives questions from customers in real time. This enables real-time and efficient customer service and high-quality service that responds to customer emotions.

[0546] "Customer voice data" refers to voice information emitted by customers, and is typically digital data of sound acquired during conversational interactions.

[0547] "Real-time" means that information processing and responses occur simultaneously with the event in which they occur, resulting in a state with extremely little delay.

[0548] "Language data" refers to text-based data obtained by converting audio data into characters, and it forms the basis for analysis using natural language processing.

[0549] "Speech recognition means" refers to a technology or device for converting speech data into text data, and typically includes a microphone and speech recognition software.

[0550] "Natural language processing methods" refer to technologies and algorithms used to analyze language data and extract its subject matter and emotions, and are used to understand the structure, meaning, and context of language.

[0551] A "memory retrieval system" is a system for retrieving relevant information from past inquiry history and knowledge bases, and generally utilizes a database management system.

[0552] "Generative machine learning methods" are algorithms and models used to generate optimal answers based on searched relevant information, and are also known as generative AI.

[0553] "Information presentation means" refers to a system for displaying the generated response on the operator's display device, which usually includes a display or monitor.

[0554] "Information storage means" refers to devices and technologies that properly store information after a transaction and use it for future reference, and databases and storage services often fall into this category.

[0555] A "smart device" is a general term for advanced electronic devices equipped with computers and communication functions, and possessing various functions including voice input.

[0556] "Input means" refers to methods or devices for taking user instructions or data into the system, and includes voice recognition devices and touch panels.

[0557] The system for implementing this invention consists of a server, terminals, and users to efficiently handle customer interactions. The server uses speech recognition means to convert customer voice data into language data in real time. For this purpose, Google Speech-to-Text or similar speech recognition APIs are used. The converted language data is analyzed by a natural language processing engine. Here, spaCy or AllenNLP is used to extract the subject and sentiment of the inquiry.

[0558] The server then searches its query history and knowledge base through memory retrieval mechanisms. This typically involves using a database management system such as MongoDB. Based on the relevant information, it then generates the optimal answer using machine learning tools. Generative AI models such as OpenAI GPT-4 are used for this generation.

[0559] The generated answers are displayed in real time on the employee's terminal via an information display device. Smart devices, such as smart glasses or head-mounted displays, receive customer questions and function as input devices. This allows the terminal to provide answers instantly, enabling rapid customer service.

[0560] Furthermore, information after the response is stored on a server using an information storage device and added to a database as incident information. This entire process supports the smooth and accurate progress of customer support.

[0561] As a concrete example, consider a scenario where a customer asks, "What is the return policy for this product?" The server analyzes the voice data, extracts relevant information about the policy, generates a detailed explanation using a generative AI model, and displays it on the device.

[0562] Examples of prompt statements include:

[0563] "Customer question: 'What is the return policy for this product?' Based on this question, please briefly explain the detailed return policy."

[0564] These are some examples.

[0565] This system will enable real-time, high-quality customer service at stores.

[0566] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0567] Step 1:

[0568] The user inputs a question by voice through a smart device. This voice is transmitted to the server as digital voice data.

[0569] Step 2:

[0570] The server uses speech recognition to convert the received audio data into text data. In this process, a speech recognition API is used to process the audio data into readable language data.

[0571] Step 3:

[0572] The server uses natural language processing (NLP) tools to analyze text data and extract the subject and sentiment of the query. Specifically, it uses an NLP engine to analyze sentence structure and emotional tone to determine the subject matter.

[0573] Step 4:

[0574] The server uses memory retrieval means to search for relevant information from query history and knowledge bases based on the extracted subject. A database management system is used to extract relevant information and past answers.

[0575] Step 5:

[0576] The server uses generative machine learning to generate the best possible answers based on the searched information. It leverages generative AI models to generate natural-sounding sentences based on the information, creating appropriate answers to the user's questions.

[0577] Step 6:

[0578] The generated responses are transmitted in real time to the terminal's display device via an information presentation system and provided to the user. The terminal visually displays the received information, allowing the user to confirm it immediately.

[0579] Step 7:

[0580] The server stores the information after handling a request in a database using an information storage device, which is then used for later analysis and reference. This stored data is then used for processing subsequent queries.

[0581] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0582] This invention realizes a system that further improves the quality of customer service by combining an emotion engine. The embodiments are described in detail below.

[0583] This system primarily consists of a server that receives voice input from users, a terminal that displays response information, and an emotion engine that performs emotion recognition. When a user connects to the service desk via telephone, the server uses speech recognition to convert the user's voice into text data. This text data is analyzed by natural language processing to extract the subject of the inquiry. At this stage, the server uses the emotion engine to recognize the user's emotions from the voice data in real time. The recognized emotions are then reflected in responses designed to enhance customer satisfaction.

[0584] The server further uses database search tools to search past inquiry history and knowledge bases, collecting relevant information. Based on this, a generation AI tool generates an optimal response that takes into account the user's emotional state. This response is transmitted to the terminal in real time via an information display tool and displayed to the customer service representative. Based on the displayed information, the representative can respond flexibly according to the customer's situation.

[0585] As a concrete example, consider a case where a user contacts the system feeling anxious because their product hasn't arrived. In this case, the server uses an emotion engine to recognize the user's anxiety from their voice and quickly generates a reassuring tone and information using AI. Subsequently, a response including a prompt delivery status check and a plan of action is displayed on the terminal. The staff member can then use this to take appropriate action to reassure the user.

[0586] In this way, by utilizing the emotion engine, this system can respond flexibly and accurately to diverse and emotional customer interactions, dramatically improving the quality of customer service.

[0587] The following describes the processing flow.

[0588] Step 1:

[0589] The user contacts the service desk by phone and verbally explains their question or problem. The server captures this audio and converts it into text data using speech recognition technology.

[0590] Step 2:

[0591] The server receives text data and uses natural language processing to analyze the subject of the query. This analysis is then used to prepare for more detailed information retrieval.

[0592] Step 3:

[0593] The server activates an emotion engine to analyze the user's emotions from the voice data. This analysis helps determine what emotions the user is experiencing, such as dissatisfaction, worry, or impatience.

[0594] Step 4:

[0595] Based on the acquired subject and sentiment data, the server uses database search methods to search the query history and knowledge base to extract relevant solutions and information.

[0596] Step 5:

[0597] The server uses AI generation tools to create the optimal response, taking into account the extracted information and the user's emotions. This response includes emotionally sensitive language and countermeasures.

[0598] Step 6:

[0599] The generated responses are sent from the server to the terminal in real time. The terminal receives this information and displays it on the screen for the person in charge to review.

[0600] Step 7:

[0601] After the conversation with the user ends, the terminal summarizes the details of the interaction and records them as incident information on the server. This record is stored in a database and used for future improvements and analysis.

[0602] (Example 2)

[0603] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0604] In customer service, it is essential to accurately recognize customer emotions from their voice and provide appropriate responses quickly based on those emotions. However, conventional systems struggle to respond while considering customer emotions, which can lead to a decline in the quality of service. In particular, the inability to provide real-time responses that respond to customer emotions can lead to a decrease in customer satisfaction.

[0605] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0606] In this invention, the server includes means for converting customer voice information into text information in real time, means for analyzing the text information to extract the subject of the communication, and means for recognizing the customer's emotions from the voice information. This makes it possible to generate and provide an appropriate response in real time that corresponds to the customer's emotions.

[0607] "Speech recognition means" refers to technology that converts speech information into text information in real time.

[0608] "Natural language processing means" refers to technologies for analyzing textual information and extracting the main topic of a message.

[0609] "Emotion analysis means" refers to technology that recognizes customer emotions in real time from voice information.

[0610] A "database search method" is a technology that searches for relevant information from past contact history and knowledge bases.

[0611] A "generative AI model" is an artificial intelligence technology that generates the optimal response by taking into account the customer's emotional state.

[0612] "Information presentation means" refers to a technology that displays the generated responses on the worker's terminal in real time.

[0613] A "memory device" is a technology that records and retains information after a response as case information.

[0614] This invention is a system that generates responses in real time based on customer voice information, thereby improving the quality of customer service. The system mainly consists of a server, terminals, and users.

[0615] The server uses speech recognition technology to receive voice input. The speech recognition software used is a common speech recognition API. This converts the user's voice information into text in real time. The converted text is then parsed using a natural language processing library to extract the subject of the communication. A wide range of open-source natural language processing libraries are available.

[0616] Subsequently, the server uses emotion analysis tools to recognize the customer's emotions from the audio information. This process utilizes emotion analysis software to determine the customer's emotions in real time. The recognized emotion information is used to generate responses that enhance customer satisfaction.

[0617] Furthermore, the server searches the customer's past inquiry history and knowledge base using a database management system. For example, a widely used relational database management system is used for the database system.

[0618] The generative AI model generates the optimal response within the server, taking into account the user's emotional state and the content of their inquiry. An advanced language model is applied to this generative AI model. This model generates natural-sounding answers based on the context and the customer's situation.

[0619] The generated response is transmitted to the terminal via an information display device. The terminal displays this information to the worker in real time, helping them to provide the best possible service to the customer.

[0620] For example, if a user contacts the support center because "the product hasn't arrived," the system sends information to the AI ​​model in the form of a prompt message such as, "The user is feeling anxious about the delay in product delivery. Please suggest measures to alleviate this anxiety." This allows the support staff to quickly provide reassurance to the user.

[0621] Thus, by combining speech recognition, emotion analysis technology, and a generative AI model, the system of the present invention enables flexible and accurate responses tailored to customer emotions, dramatically improving the quality of customer service.

[0622] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0623] Step 1:

[0624] The user connects to the service desk via telephone and makes an inquiry in voice format. The input here is the user's voice, which the server receives. The user's voice is the first data input for the system.

[0625] Step 2:

[0626] The server uses speech recognition to convert the received user's voice into text information in real time. Specifically, speech recognition software analyzes the speech waveform and generates corresponding text data. Here, the input is the audio signal, and the output is the text information.

[0627] Step 3:

[0628] The server analyzes the converted character data using natural language processing tools. Specifically, it uses a natural language processing library to extract the subject and intent of the customer's inquiry from the string. The input here is text data, and the output is the extracted subject and intent information.

[0629] Step 4:

[0630] The server uses emotion analysis tools to recognize customer emotions from voice and text data. Specifically, emotion analysis software evaluates features such as language and tone of voice to identify emotions. Input is voice and text, and output is emotion information.

[0631] Step 5:

[0632] The server utilizes database search capabilities to retrieve past customer inquiry history and knowledge base data. Here, the database system executes queries to collect relevant information. The input is the extracted subject, and the output is the related historical data.

[0633] Step 6:

[0634] The server uses a generative AI model to generate the optimal response, taking into account the customer's emotional state and the content of their inquiry. Specifically, the generative AI model generates a contextually natural response based on the prompt text. The inputs are the subject, emotional information, and historical data, and the output is the generated response.

[0635] Step 7:

[0636] The server sends the generated response to the terminal via an information display device. The terminal displays this information on the operator's screen in real time. The input here is the generated response data, and the output is the display on the terminal screen. The person in charge uses this information to take appropriate action for the customer.

[0637] (Application Example 2)

[0638] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0639] In modern e-commerce websites, simply presenting product information when customers obtain it via voice can lead to decreased customer satisfaction. Furthermore, the lack of personalized responses that take customer emotions into account makes it difficult to provide prompt and effective customer support.

[0640] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0641] In this invention, the server includes speech recognition means for converting customer voice information into text information in real time, natural language processing means for analyzing the text information and extracting the subject of the inquiry, and emotion recognition means for detecting the customer's emotional state and generating a response based on that emotional state. This makes it possible to provide customized information that takes into account the customer's emotions when they make an inquiry by voice.

[0642] "Customer voice information" refers to voice data obtained from users, and includes information such as the customer's intentions and emotions.

[0643] "Textual information" refers to text data converted from audio data by speech recognition technology.

[0644] "Speech recognition means" refers to technologies and devices that convert speech data into text data in real time.

[0645] "Natural language processing means" refers to techniques for analyzing text data and extracting the subject of a query.

[0646] "Information retrieval means" refers to technologies and devices for searching for relevant information from inquiry history or knowledge bases.

[0647] "Generative AI means" refers to artificial intelligence technology used to generate the optimal answer based on relevant information.

[0648] "Information presentation means" refers to technologies or devices that display generated answers on the employee's terminal in real time.

[0649] "Storage methods" refer to technologies and devices for recording information after a response as event information.

[0650] "Emotion recognition means" refers to technology that detects a customer's emotional state from voice information and generates a response based on that.

[0651] This application example is an embodiment of a system that improves customer support using customer voice information. The system consists of voice recognition means, natural language processing means, information retrieval means, generation AI means, information presentation means, storage means, and emotion recognition means. These means are combined to analyze customer voice input and generate appropriate responses.

[0652] First, when a user makes a voice inquiry to the service using their smartphone, the server converts the voice data into text using speech recognition technology. By using software such as the Google Speech-to-Text API for speech recognition, it is possible to convert voice data into text with high accuracy.

[0653] Next, the server analyzes the text information using natural language processing techniques to extract the subject of the query. In this step, natural language processing techniques using TensorFlow are employed to accurately grasp the user's intent.

[0654] Subsequently, the emotion recognition system uses emotion analysis software such as IBM Watson Tone Analyzer to detect the customer's emotions and passes them to the generative AI system. The generative AI system uses OpenAI's GPT model to generate answers that enable personalized responses that take into account the customer's emotional state.

[0655] The generated responses are displayed realistically and in real time on the user's device through an information presentation system. This allows the user to receive responses that match their emotions. Finally, all response information is recorded as event information using a storage system.

[0656] For example, if a customer asks, "I want this product right now, do you have it in stock?", the system recognizes the sentiment as "hopeful," quickly searches for inventory information, and provides a customized response such as, "It is currently out of stock in our online store, but we can check the stock at your nearest store."

[0657] Examples of prompts to input into a generative AI model:

[0658] "User input: Do you have it in stock? Emotion: Hopeful. Answer:"

[0659] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0660] Step 1:

[0661] The user makes a voice inquiry using their smartphone. This voice data is sent to the server. The input here is the user's voice data. The server receives this voice data and sends it to the next processing step.

[0662] Step 2:

[0663] The server converts audio data into text using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to convert speech to text. The input here is audio data, and the output is text. The server then passes this text to the next step.

[0664] Step 3:

[0665] The server analyzes the received text information using natural language processing techniques to extract the subject of the query. Natural language processing techniques using TensorFlow are employed here. The input is text information, and the output is parsed data including the topic.

[0666] Step 4:

[0667] The server passes the analyzed data to an emotion recognition system to detect the customer's emotional state. Specifically, it uses IBM Watson Tone Analyzer to analyze the sentiment of textual information and identify the emotional state. The input is the analyzed data, and the output is the emotional state.

[0668] Step 5:

[0669] The server uses generative AI to generate the optimal response, taking into account the emotional state. This process uses OpenAI's GPT model to create personalized responses tailored to the customer's emotions. The input is the emotional state and the subject of the inquiry, and the output is the customized response.

[0670] Step 6:

[0671] The server displays the generated response on the user's terminal using an information display mechanism. The response is displayed in real time, allowing the user to review it. The input is the customized response, and the output is the information displayed on the terminal.

[0672] Step 7:

[0673] The server records all interactions using storage methods and saves them as event information in a database. This provides a history of interactions that can be referenced later. The input consists of the response and data of its processing, while the output is the saved event information.

[0674] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0675] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0676] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0677] [Fourth Embodiment]

[0678] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0679] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0680] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0681] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0682] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0683] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0684] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0685] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0686] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0687] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0688] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0689] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0690] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0691] The system of the present invention is designed to streamline customer service and improve service quality. Its embodiments are described in detail below.

[0692] This system primarily consists of a server, terminals, and users. When a user contacts the service desk via telephone, the server uses speech recognition to convert the user's voice into text data in real time. This converted text data is then analyzed by the server using natural language processing to extract the subject and sentiment of the inquiry.

[0693] Next, the server uses a database search mechanism to retrieve past inquiry history and knowledge base information based on the extracted information. Using this retrieved information, the server utilizes a generation AI mechanism to generate the optimal answer. This answer includes information to address the user's specific question and is displayed on the terminal in real time through an information display mechanism.

[0694] The terminal receives this generated information and displays it on the screen, helping staff members respond to users quickly. By obtaining the answers users need, the efficiency of counter services is significantly improved. After the interaction is complete, a record of the interaction performed by the staff member on the terminal is saved to the server by a storage device and stored in the database as incident information.

[0695] As a concrete example, consider a scenario where a user asks a question about their account password. The server transcribes the user's voice into text and analyzes it to identify the topic related to "password reset." It searches the database for past reset procedures and uses AI generation to create the optimal procedure. This procedure is displayed on the staff member's terminal, allowing the staff member to immediately guide the user through the process.

[0696] This allows the system to respond quickly and accurately, thereby increasing customer satisfaction.

[0697] The following describes the processing flow.

[0698] Step 1:

[0699] When a user calls the service desk and communicates their inquiry verbally, the server captures the audio. Using speech recognition technology, this audio data is converted into text data in real time.

[0700] Step 2:

[0701] After receiving the text data, the server performs analysis using natural language processing techniques. It identifies the subject of the inquiry, simultaneously conducts customer sentiment analysis, and extracts relevant information.

[0702] Step 3:

[0703] Based on the extracted information, the server uses database search tools to retrieve past query history and knowledge base information. This information includes previous similar cases and related solutions.

[0704] Step 4:

[0705] The server uses AI generation tools to create the optimal answer based on the acquired information. The prediction algorithm also calculates possible next questions and prepares additional information.

[0706] Step 5:

[0707] The generated response and additional information are sent from the server to the terminal. The terminal receives this information and displays it on the screen in real time. This information display allows the person in charge to respond to the user quickly.

[0708] Step 6:

[0709] After the interaction with the user is complete, the details and results of the interaction are confirmed on the terminal and sent to the server. The server then uses a storage device to record this information as incident data in a database.

[0710] (Example 1)

[0711] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0712] In customer service, the challenge lies in providing prompt and accurate responses to improve customer satisfaction. In particular, efficient handling of inquiries via telephone and seamless information provision by staff are essential. Furthermore, it is necessary to effectively utilize past inquiry history and knowledge bases while addressing the diverse emotions of customers.

[0713] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0714] In this invention, the server includes speech recognition means for converting customer voice data into text data in real time, natural language processing means for analyzing the text data to extract the subject and sentiment of the inquiry, database search means for searching for relevant information from inquiry history and a knowledge base, generation modeling means for generating the optimal answer based on the relevant information, information presentation means for displaying the generated answer on a display device in real time, and storage means for recording post-handling information as situational information. This makes it possible to provide quick and accurate responses to customer inquiries and significantly improve the efficiency of customer service.

[0715] "Speech recognition means" refers to technology for converting speech data into text data in real time.

[0716] "Natural language processing means" refers to techniques that analyze text data and extract the subject and sentiment of an inquiry from it.

[0717] A "database search method" is a technology used to search query history and knowledge bases to identify relevant information.

[0718] "Generative modeling means" refers to artificial intelligence technology used to generate the optimal response based on acquired information.

[0719] "Information presentation means" refers to technology for displaying generated responses on a display device in real time.

[0720] A "memory device" is a technology that records information after an interaction as situational information and stores it for future reference.

[0721] The system of this invention is designed to streamline customer service by integrating means of speech recognition, natural language processing, database retrieval, generative modeling, information presentation, and memory. Specific embodiments utilizing each means are described below.

[0722] Speech recognition means

[0723] The server captures voice data from users in real time and converts it into text data using speech recognition technology. This process utilizes a speech recognition API (e.g., general-purpose speech recognition software) to enable accurate text conversion.

[0724] Natural language processing means

[0725] The server analyzes the converted text data using natural language processing software (e.g., a general-purpose natural language processing library). This analysis extracts the subject and sentiment of the query.

[0726] Database search means

[0727] The server uses the extracted information to search past query history and knowledge bases through the database. This search allows for the rapid collection of relevant information. A database management system (e.g., general-purpose database management software) assists in this process.

[0728] Generative modeling means

[0729] The server uses a generative AI model to create the optimal response to provide to the customer based on the collected information. This process utilizes generative modeling techniques (e.g., general-purpose generative AI models).

[0730] Information presentation means

[0731] The terminal displays the generated responses in real time, allowing staff to respond to users quickly. This enables effective customer service.

[0732] storage means

[0733] The response results recorded on the terminal are efficiently collected and stored on the server. This allows them to be managed as situational information that can be referenced at a later date.

[0734] Specific example

[0735] Let's take an example where a user inquires about their account password over the phone. The server converts the user's voice into text, and analysis identifies the topic as "password reset." The server searches its database for relevant procedures and uses a generative AI model to generate the reset procedure. This procedure is displayed on the terminal, allowing the staff to immediately guide the user through it.

[0736] Example of a prompt

[0737] "User: I forgot my account password. How can I reset it?"

[0738] This system is expected to expedite responses to customer inquiries and significantly improve service quality.

[0739] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0740] Step 1:

[0741] Receiving voice input

[0742] The user contacts the service desk via telephone. The server receives the voice data from the user in real time. The voice data, as input, is passed to the server in its original format.

[0743] Step 2:

[0744] Converting audio data to text

[0745] The server converts the received audio data into text data using speech recognition. Specifically, it analyzes the audio waveform using a speech recognition API, applies a language model, and converts the content into text format. This text data is then passed to the next step as output.

[0746] Step 3:

[0747] Text data analysis

[0748] The server analyzes the generated text data using natural language processing (NLP) tools. It uses NLP software to extract the subject and sentiment of the query from the input text data. The resulting subject and sentiment information is then output, allowing the server to proceed to the next step.

[0749] Step 4:

[0750] Search for related information

[0751] The server uses database search tools to retrieve past query history and knowledge base information based on the extracted subject information. The input consists of the subject information from the analysis results and the contents of the database, and the output is information related to the query. This related information is then passed on to the next step.

[0752] Step 5:

[0753] Generating answers using AI

[0754] The server uses a generative AI model based on the acquired relevant information to generate the optimal answer. Relevant information is provided to the AI ​​model as input and prompts, and a constructed answer is generated as output. This answer is then passed on to the next step.

[0755] Step 6:

[0756] Display the answer

[0757] The server generates a response, which is then sent to the terminal via an information display device. The terminal displays this response on the staff member's screen in real time. The outputted response is then presented to the staff member for user support.

[0758] Step 7:

[0759] Record of the results of the response

[0760] The information handled by the staff member on the terminal is sent to the server, and the server records this information as status data in a database using its storage device. The entered information is recorded exactly as it is and saved for future reference.

[0761] (Application Example 1)

[0762] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0763] There is a need to improve the real-time and efficiency of customer service, as well as the quality of service by taking customer emotions into consideration. Traditional methods have been problematic because they make it difficult to provide information to respond quickly and appropriately to a large number of customer inquiries. Furthermore, if information after an interaction is not properly managed, it is less likely to be used to improve future interactions, resulting in decreased customer satisfaction.

[0764] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0765] In this invention, the server includes a speech recognition means that converts customer voice data into language data in real time; a natural language processing means that analyzes the language data to extract the subject of the inquiry; a memory retrieval means that searches for relevant information from the inquiry history and knowledge base; a generation machine learning means that generates the optimal answer based on the relevant information; an information presentation means that displays the generated answer in real time on the employee's display device; an information storage means that records the information after the response as event information; and an input means that is mounted on a smart device and receives questions from customers in real time. This enables real-time and efficient customer service and high-quality service that responds to customer emotions.

[0766] "Customer voice data" refers to voice information emitted by customers, and is typically digital data of sound acquired during conversational interactions.

[0767] "Real-time" means that information processing and responses occur simultaneously with the event in which they occur, resulting in a state with extremely little delay.

[0768] "Language data" refers to text-based data obtained by converting audio data into characters, and it forms the basis for analysis using natural language processing.

[0769] "Speech recognition means" refers to a technology or device for converting speech data into text data, and typically includes a microphone and speech recognition software.

[0770] "Natural language processing methods" refer to technologies and algorithms used to analyze language data and extract its subject matter and emotions, and are used to understand the structure, meaning, and context of language.

[0771] A "memory retrieval system" is a system for retrieving relevant information from past inquiry history and knowledge bases, and generally utilizes a database management system.

[0772] "Generative machine learning methods" are algorithms and models used to generate optimal answers based on searched relevant information, and are also known as generative AI.

[0773] "Information presentation means" refers to a system for displaying the generated response on the operator's display device, which usually includes a display or monitor.

[0774] "Information storage means" refers to devices and technologies that properly store information after a transaction and use it for future reference, and databases and storage services often fall into this category.

[0775] A "smart device" is a general term for advanced electronic devices equipped with computers and communication functions, and possessing various functions including voice input.

[0776] "Input means" refers to methods or devices for taking user instructions or data into the system, and includes voice recognition devices and touch panels.

[0777] The system for implementing this invention consists of a server, terminals, and users to efficiently handle customer interactions. The server uses speech recognition means to convert customer voice data into language data in real time. For this purpose, Google Speech-to-Text or similar speech recognition APIs are used. The converted language data is analyzed by a natural language processing engine. Here, spaCy or AllenNLP is used to extract the subject and sentiment of the inquiry.

[0778] The server then searches its query history and knowledge base through memory retrieval mechanisms. This typically involves using a database management system such as MongoDB. Based on the relevant information, it then generates the optimal answer using machine learning tools. Generative AI models such as OpenAI GPT-4 are used for this generation.

[0779] The generated answers are displayed in real time on the employee's terminal via an information display device. Smart devices, such as smart glasses or head-mounted displays, receive customer questions and function as input devices. This allows the terminal to provide answers instantly, enabling rapid customer service.

[0780] Furthermore, information after the response is stored on a server using an information storage device and added to a database as incident information. This entire process supports the smooth and accurate progress of customer support.

[0781] As a concrete example, consider a scenario where a customer asks, "What is the return policy for this product?" The server analyzes the voice data, extracts relevant information about the policy, generates a detailed explanation using a generative AI model, and displays it on the device.

[0782] Examples of prompt statements include:

[0783] "Customer question: 'What is the return policy for this product?' Based on this question, please briefly explain the detailed return policy."

[0784] These are some examples.

[0785] This system will enable real-time, high-quality customer service at stores.

[0786] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0787] Step 1:

[0788] The user inputs a question by voice through a smart device. This voice is transmitted to the server as digital voice data.

[0789] Step 2:

[0790] The server uses speech recognition to convert the received audio data into text data. In this process, a speech recognition API is used to process the audio data into readable language data.

[0791] Step 3:

[0792] The server uses natural language processing (NLP) tools to analyze text data and extract the subject and sentiment of the query. Specifically, it uses an NLP engine to analyze sentence structure and emotional tone to determine the subject matter.

[0793] Step 4:

[0794] The server uses memory retrieval means to search for relevant information from query history and knowledge bases based on the extracted subject. A database management system is used to extract relevant information and past answers.

[0795] Step 5:

[0796] The server uses generative machine learning to generate the best possible answers based on the searched information. It leverages generative AI models to generate natural-sounding sentences based on the information, creating appropriate answers to the user's questions.

[0797] Step 6:

[0798] The generated responses are transmitted in real time to the terminal's display device via an information presentation system and provided to the user. The terminal visually displays the received information, allowing the user to confirm it immediately.

[0799] Step 7:

[0800] The server stores the information after handling a request in a database using an information storage device, which is then used for later analysis and reference. This stored data is then used for processing subsequent queries.

[0801] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0802] This invention realizes a system that further improves the quality of customer service by combining an emotion engine. The embodiments are described in detail below.

[0803] This system primarily consists of a server that receives voice input from users, a terminal that displays response information, and an emotion engine that performs emotion recognition. When a user connects to the service desk via telephone, the server uses speech recognition to convert the user's voice into text data. This text data is analyzed by natural language processing to extract the subject of the inquiry. At this stage, the server uses the emotion engine to recognize the user's emotions from the voice data in real time. The recognized emotions are then reflected in responses designed to enhance customer satisfaction.

[0804] The server further uses database search tools to search past inquiry history and knowledge bases, collecting relevant information. Based on this, a generation AI tool generates an optimal response that takes into account the user's emotional state. This response is transmitted to the terminal in real time via an information display tool and displayed to the customer service representative. Based on the displayed information, the representative can respond flexibly according to the customer's situation.

[0805] As a concrete example, consider a case where a user contacts the system feeling anxious because their product hasn't arrived. In this case, the server uses an emotion engine to recognize the user's anxiety from their voice and quickly generates a reassuring tone and information using AI. Subsequently, a response including a prompt delivery status check and a plan of action is displayed on the terminal. The staff member can then use this to take appropriate action to reassure the user.

[0806] In this way, by utilizing the emotion engine, this system can respond flexibly and accurately to diverse and emotional customer interactions, dramatically improving the quality of customer service.

[0807] The following describes the processing flow.

[0808] Step 1:

[0809] The user contacts the service desk by phone and verbally explains their question or problem. The server captures this audio and converts it into text data using speech recognition technology.

[0810] Step 2:

[0811] The server receives text data and uses natural language processing to analyze the subject of the query. This analysis is then used to prepare for more detailed information retrieval.

[0812] Step 3:

[0813] The server activates an emotion engine to analyze the user's emotions from the voice data. This analysis helps determine what emotions the user is experiencing, such as dissatisfaction, worry, or impatience.

[0814] Step 4:

[0815] Based on the acquired subject and sentiment data, the server uses database search methods to search the query history and knowledge base to extract relevant solutions and information.

[0816] Step 5:

[0817] The server uses AI generation tools to create the optimal response, taking into account the extracted information and the user's emotions. This response includes emotionally sensitive language and countermeasures.

[0818] Step 6:

[0819] The generated responses are sent from the server to the terminal in real time. The terminal receives this information and displays it on the screen for the person in charge to review.

[0820] Step 7:

[0821] After the conversation with the user ends, the terminal summarizes the details of the interaction and records them as incident information on the server. This record is stored in a database and used for future improvements and analysis.

[0822] (Example 2)

[0823] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0824] In customer service, it is essential to accurately recognize customer emotions from their voice and provide appropriate responses quickly based on those emotions. However, conventional systems struggle to respond while considering customer emotions, which can lead to a decline in the quality of service. In particular, the inability to provide real-time responses that respond to customer emotions can lead to a decrease in customer satisfaction.

[0825] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0826] In this invention, the server includes means for converting customer voice information into text information in real time, means for analyzing the text information to extract the subject of the communication, and means for recognizing the customer's emotions from the voice information. This makes it possible to generate and provide an appropriate response in real time that corresponds to the customer's emotions.

[0827] "Speech recognition means" refers to technology that converts speech information into text information in real time.

[0828] "Natural language processing means" refers to technologies for analyzing textual information and extracting the main topic of a message.

[0829] "Emotion analysis means" refers to technology that recognizes customer emotions in real time from voice information.

[0830] A "database search method" is a technology that searches for relevant information from past contact history and knowledge bases.

[0831] A "generative AI model" is an artificial intelligence technology that generates the optimal response by taking into account the customer's emotional state.

[0832] "Information presentation means" refers to a technology that displays the generated responses on the worker's terminal in real time.

[0833] A "memory device" is a technology that records and retains information after a response as case information.

[0834] This invention is a system that generates responses in real time based on customer voice information, thereby improving the quality of customer service. The system mainly consists of a server, terminals, and users.

[0835] The server uses speech recognition technology to receive voice input. The speech recognition software used is a common speech recognition API. This converts the user's voice information into text in real time. The converted text is then parsed using a natural language processing library to extract the subject of the communication. A wide range of open-source natural language processing libraries are available.

[0836] Subsequently, the server uses emotion analysis tools to recognize the customer's emotions from the audio information. This process utilizes emotion analysis software to determine the customer's emotions in real time. The recognized emotion information is used to generate responses that enhance customer satisfaction.

[0837] Furthermore, the server searches the customer's past inquiry history and knowledge base using a database management system. For example, a widely used relational database management system is used for the database system.

[0838] The generative AI model generates the optimal response within the server, taking into account the user's emotional state and the content of their inquiry. An advanced language model is applied to this generative AI model. This model generates natural-sounding answers based on the context and the customer's situation.

[0839] The generated response is transmitted to the terminal via an information display device. The terminal displays this information to the worker in real time, helping them to provide the best possible service to the customer.

[0840] For example, if a user contacts the support center because "the product hasn't arrived," the system sends information to the AI ​​model in the form of a prompt message such as, "The user is feeling anxious about the delay in product delivery. Please suggest measures to alleviate this anxiety." This allows the support staff to quickly provide reassurance to the user.

[0841] Thus, by combining speech recognition, emotion analysis technology, and a generative AI model, the system of the present invention enables flexible and accurate responses tailored to customer emotions, dramatically improving the quality of customer service.

[0842] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0843] Step 1:

[0844] The user connects to the service desk via telephone and makes an inquiry in voice format. The input here is the user's voice, which the server receives. The user's voice is the first data input for the system.

[0845] Step 2:

[0846] The server uses speech recognition to convert the received user's voice into text information in real time. Specifically, speech recognition software analyzes the speech waveform and generates corresponding text data. Here, the input is the audio signal, and the output is the text information.

[0847] Step 3:

[0848] The server analyzes the converted character data using natural language processing tools. Specifically, it uses a natural language processing library to extract the subject and intent of the customer's inquiry from the string. The input here is text data, and the output is the extracted subject and intent information.

[0849] Step 4:

[0850] The server uses emotion analysis tools to recognize customer emotions from voice and text data. Specifically, emotion analysis software evaluates features such as language and tone of voice to identify emotions. Input is voice and text, and output is emotion information.

[0851] Step 5:

[0852] The server utilizes database search capabilities to retrieve past customer inquiry history and knowledge base data. Here, the database system executes queries to collect relevant information. The input is the extracted subject, and the output is the related historical data.

[0853] Step 6:

[0854] The server uses a generative AI model to generate the optimal response, taking into account the customer's emotional state and the content of their inquiry. Specifically, the generative AI model generates a contextually natural response based on the prompt text. The inputs are the subject, emotional information, and historical data, and the output is the generated response.

[0855] Step 7:

[0856] The server sends the generated response to the terminal via an information display device. The terminal displays this information on the operator's screen in real time. The input here is the generated response data, and the output is the display on the terminal screen. The person in charge uses this information to take appropriate action for the customer.

[0857] (Application Example 2)

[0858] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0859] In modern e-commerce websites, simply presenting product information when customers obtain it via voice can lead to decreased customer satisfaction. Furthermore, the lack of personalized responses that take customer emotions into account makes it difficult to provide prompt and effective customer support.

[0860] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0861] In this invention, the server includes speech recognition means for converting customer voice information into text information in real time, natural language processing means for analyzing the text information and extracting the subject of the inquiry, and emotion recognition means for detecting the customer's emotional state and generating a response based on that emotional state. This makes it possible to provide customized information that takes into account the customer's emotions when they make an inquiry by voice.

[0862] "Customer voice information" refers to voice data obtained from users, and includes information such as the customer's intentions and emotions.

[0863] "Textual information" refers to text data converted from audio data by speech recognition technology.

[0864] "Speech recognition means" refers to technologies and devices that convert speech data into text data in real time.

[0865] "Natural language processing means" refers to techniques for analyzing text data and extracting the subject of a query.

[0866] "Information retrieval means" refers to technologies and devices for searching for relevant information from inquiry history or knowledge bases.

[0867] "Generative AI means" refers to artificial intelligence technology used to generate the optimal answer based on relevant information.

[0868] "Information presentation means" refers to technologies or devices that display generated answers on the employee's terminal in real time.

[0869] "Storage methods" refer to technologies and devices for recording information after a response as event information.

[0870] "Emotion recognition means" refers to technology that detects a customer's emotional state from voice information and generates a response based on that.

[0871] This application example is an embodiment of a system that improves customer support using customer voice information. The system consists of voice recognition means, natural language processing means, information retrieval means, generation AI means, information presentation means, storage means, and emotion recognition means. These means are combined to analyze customer voice input and generate appropriate responses.

[0872] First, when a user makes a voice inquiry to the service using their smartphone, the server converts the voice data into text using speech recognition technology. By using software such as the Google Speech-to-Text API for speech recognition, it is possible to convert voice data into text with high accuracy.

[0873] Next, the server analyzes the text information using natural language processing techniques to extract the subject of the query. In this step, natural language processing techniques using TensorFlow are employed to accurately grasp the user's intent.

[0874] Subsequently, the emotion recognition system uses emotion analysis software such as IBM Watson Tone Analyzer to detect the customer's emotions and passes them to the generative AI system. The generative AI system uses OpenAI's GPT model to generate answers that enable personalized responses that take into account the customer's emotional state.

[0875] The generated responses are displayed realistically and in real time on the user's device through an information presentation system. This allows the user to receive responses that match their emotions. Finally, all response information is recorded as event information using a storage system.

[0876] For example, if a customer asks, "I want this product right now, do you have it in stock?", the system recognizes the sentiment as "hopeful," quickly searches for inventory information, and provides a customized response such as, "It is currently out of stock in our online store, but we can check the stock at your nearest store."

[0877] Examples of prompts to input into a generative AI model:

[0878] "User input: Do you have it in stock? Emotion: Hopeful. Answer:"

[0879] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0880] Step 1:

[0881] The user makes a voice inquiry using their smartphone. This voice data is sent to the server. The input here is the user's voice data. The server receives this voice data and sends it to the next processing step.

[0882] Step 2:

[0883] The server converts audio data into text using speech recognition technology. Specifically, it uses the Google Speech-to-Text API to convert speech to text. The input here is audio data, and the output is text. The server then passes this text to the next step.

[0884] Step 3:

[0885] The server analyzes the received text information using natural language processing techniques to extract the subject of the query. Natural language processing techniques using TensorFlow are employed here. The input is text information, and the output is parsed data including the topic.

[0886] Step 4:

[0887] The server passes the analyzed data to an emotion recognition system to detect the customer's emotional state. Specifically, it uses IBM Watson Tone Analyzer to analyze the sentiment of textual information and identify the emotional state. The input is the analyzed data, and the output is the emotional state.

[0888] Step 5:

[0889] The server uses generative AI to generate the optimal response, taking into account the emotional state. This process uses OpenAI's GPT model to create personalized responses tailored to the customer's emotions. The input is the emotional state and the subject of the inquiry, and the output is the customized response.

[0890] Step 6:

[0891] The server displays the generated response on the user's terminal using an information display mechanism. The response is displayed in real time, allowing the user to review it. The input is the customized response, and the output is the information displayed on the terminal.

[0892] Step 7:

[0893] The server records all interactions using storage methods and saves them as event information in a database. This provides a history of interactions that can be referenced later. The input consists of the response and data of its processing, while the output is the saved event information.

[0894] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0895] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0896] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0897] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0898] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0899] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0900] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0901] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0902] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0903] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0904] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0905] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0906] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0907] 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.

[0908] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0909] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0910] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0911] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0912] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0913] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0914] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0915] The following is further disclosed regarding the embodiments described above.

[0916] (Claim 1)

[0917] A speech recognition method that converts customer voice data into text data in real time,

[0918] A natural language processing means for analyzing the text data and extracting the subject of the query,

[0919] A database search method for retrieving relevant information from inquiry history and knowledge base,

[0920] A generation AI means that generates the optimal answer based on the relevant information,

[0921] An information presentation method that displays the generated response in real time on the employee's terminal,

[0922] A storage means for recording information after the response as incident information,

[0923] A system that includes this.

[0924] (Claim 2)

[0925] The system according to claim 1, wherein the generating AI means automatically prepares additional information for the next predicted customer question.

[0926] (Claim 3)

[0927] The system according to claim 1, wherein the speech recognition means performs emotion analysis from speech data and generates a response based on the customer's emotions.

[0928] "Example 1"

[0929] (Claim 1)

[0930] A speech recognition method that converts customer voice data into text data in real time,

[0931] A natural language processing means for analyzing the text data to extract the subject and sentiment of the inquiry,

[0932] A database search method for retrieving relevant information from inquiry history and knowledge bases,

[0933] A generation modeling means that generates the optimal answer based on the relevant information,

[0934] Information presentation means that displays the generated response on a display device in real time,

[0935] A storage means for recording information after the response as situational information,

[0936] A system that includes this.

[0937] (Claim 2)

[0938] The system according to claim 1, wherein the generation modeling means automatically prepares additional information for the next predicted customer question.

[0939] (Claim 3)

[0940] The system according to claim 1, wherein the speech recognition means performs emotion analysis from speech data and generates a response based on the customer's emotions.

[0941] "Application Example 1"

[0942] (Claim 1)

[0943] A speech recognition method that converts customer voice data into language data in real time,

[0944] A natural language processing means for analyzing the language data and extracting the subject of the query,

[0945] A memory retrieval method that searches for related information from inquiry history and knowledge bases,

[0946] A generative machine learning means that generates the optimal answer based on the relevant information,

[0947] An information presentation means that displays the generated response in real time on the person in charge's display device,

[0948] Information storage means for recording information after a response as event information,

[0949] An input method installed in a smart device that receives customer questions in real time,

[0950] A system that includes this.

[0951] (Claim 2)

[0952] The system according to claim 1, wherein the generative machine learning means automatically prepares additional information for the next predicted customer question.

[0953] (Claim 3)

[0954] The system according to claim 1, wherein the speech recognition means performs emotion analysis from speech data and generates a response based on the customer's emotions.

[0955] "Example 2 of combining an emotion engine"

[0956] (Claim 1)

[0957] A speech recognition method that converts customer voice information into text information in real time,

[0958] A natural language processing means for analyzing the textual information and extracting the subject of the communication,

[0959] A means of analyzing customer emotions from voice information,

[0960] A database search method for retrieving relevant information from past contact history and knowledge bases,

[0961] A means for generating an optimal response that takes into account the customer's emotional state using an AI model based on the relevant information,

[0962] An information display means that displays the generated response on the worker's terminal in real time,

[0963] A storage means for recording information after a response as case information,

[0964] A system that includes this.

[0965] (Claim 2)

[0966] The system according to claim 1, wherein the generating AI model automatically prepares supplementary information for the next predicted customer question.

[0967] (Claim 3)

[0968] The system according to claim 1, wherein the emotion analysis means performs emotion analysis from voice information and generates a response based on the customer's emotions.

[0969] "Application example 2 when combining with an emotional engine"

[0970] (Claim 1)

[0971] A speech recognition method that converts customer voice information into text information in real time,

[0972] A natural language processing means for analyzing the character information and extracting the subject of the query,

[0973] Information retrieval methods that search for relevant information from inquiry history and knowledge bases,

[0974] A generation AI means that generates the optimal answer based on the relevant information,

[0975] An information display means that displays the generated answers in real time on the staff member's terminal,

[0976] A means of recording the information after the response as event information,

[0977] An emotion recognition means for detecting a customer's emotional state and generating a response based on that emotional state,

[0978] A system that includes this.

[0979] (Claim 2)

[0980] The system according to claim 1, wherein the generating AI means automatically prepares additional information for the next predicted customer question and presents the generated answer and information corresponding to the customer's emotions.

[0981] (Claim 3)

[0982] The system according to claim 1, wherein the information presentation means provides customized information based on the customer's emotional state and displays product information and emotional responses in an integrated manner. [Explanation of symbols]

[0983] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A speech recognition method that converts customer voice data into language data in real time, A natural language processing means for analyzing the language data and extracting the subject of the query, A memory retrieval method that searches for related information from inquiry history and knowledge bases, A generative machine learning means that generates the optimal answer based on the relevant information, An information presentation means that displays the generated response in real time on the person in charge's display device, Information storage means for recording information after a response as event information, An input method installed in a smart device that receives customer questions in real time, A system that includes this.

2. The system according to claim 1, wherein the generative machine learning means automatically prepares additional information for the next predicted customer question.

3. The system according to claim 1, wherein the speech recognition means performs emotion analysis from speech data and generates a response based on the customer's emotions.

Images